ABSTRACTS - icatsconf.org · ABSTRACTS E-ISBN: 978-605-67535-0-3. International Conference on Advanced Technology & Sciences ... ELIF DOGU, MICHELE CEDOLIN, Y ESRA ALBAYRAK EVALUATION

Science@

ICAT'17 International Conference on Advanced Technology & Sciences

ABSTRACTS May 9-12 2017

ISTANBUL / TURKEY Editors Omer Faruk BAY Ismail SARITAS

E-ISBN: 978-605-67535-0-3

International Conference on Advanced

Technology & Sciences

5th International Conference, ICAT’17

Istanbul, Turkey, May 9-12, 2017

Abstracts

Editors

Ismail SARITAS

Omer Faruk BAY

International Conference on Advanced Technology & Sciences, ICAT’17

5th International Conference on Advanced Technology & Sciences


ii International Conference on Advanced Technology & Sciences (ICAT’17)


International Conference on Advanced

Technology & Sciences

5th International Conference, ICAT’17


Abstracts

Editors

Ismail SARITAS

Omer Faruk BAY

E-ISBN: 978-605-67535-0-3

Saday Mühendislik Sertifika No: 35542

www.saday.com.tr

Saday Mühendislik

Kürden Mh. Temizciler Sk. No:5 Meram / KONYA

Tel: 0.332 323 07 39

May – 2017

International Conference on Advanced Technology & Sciences (ICAT’17) iii


iv International Conference on Advanced Technology & Sciences (ICAT’17)


EDITORS :

Ismail SARITAS

Selcuk University, Turkey

Depertmant of Electric-Electronic Engineering, Faculty of Technology

Alaeddin Keykubat Campus 42031 Konya, Turkey [email protected]

Omer Faruk BAY,

Gazi University, Turkey

Depertmant of Electronic - Computer Education

Teknikokullar, 06500, Ankara, Turkey

e-mail: [email protected]

ASSISTANT EDITORS :

Kemal TUTUNCU


Depertmant of Electric-Electronic Engineering, Faculty of Technology


Ilker Ali OZKAN


Depertmant of Computer Engineering, Faculty of Technology


Murat KOKLU


Depertmant of Computer Engineering, Faculty of Technology


International Conference on Advanced Technology & Sciences (ICAT’17) v


vi International Conference on Advanced Technology & Sciences (ICAT’17)


PREFACE

International Conference on Advanced Technology & Sciences (ICAT'17) has

been organized in Istanbul, Turkey on 9-12 May, 2017.

The main objective of ICAT'17 is to present the latest research and results of

scientists related to all sub-disciplines of Electrical Engineering. This conference

provides opportunities for the different areas delegates to exchange new ideas and

application experiences face to face, to establish business or research relations and to

find global partners for future collaboration.

We hope that the conference results provide significant contribution to the

knowledge in this scientific field. The organizing committee of conference is pleased

to invite prospective authors to submit their original manuscripts to ICAT'17.

All paper submissions will be double-blind and peer-reviewed and evaluated

based on originality, technical and/or research content/depth, correctness, relevance to

conference, contributions, and readability. All full paper submissions have been

double blind and peer reviewed and evaluated based on originality, technical and/or

research content/depth, correctness, relevance to conference, contributions, and

readability. Selected papers presented in the conference that match with the topics of

the journals will be published in the following journals:

International Journal of Intelligent Systems and Applications in Engineering

(IJISAE)

International Journal of Applied Mathematics, Electronics and Computers

(IJAMEC)

International Journal of Energy Applications and Technology (IJEAT)

At this conference, there are 319 paper submissions from 14 different countries

and 201universities. Each paper proposal was evaluated by two reviewers. And

finally, 227 papers will be presented at our conference.

In particular we would like to thank Prof. Dr. Mustafa SAHIN, Rector of Selcuk

University; Prof.Dr. Senay YALCIN, Rector of Bahcesehir University; Prof.Dr. Ali

GUNGOR, Vice Rector of Bahcesehir University; Prof.Dr. Mehmet OKKA, Vice

Rector of Selcuk University; Advanced Technology and Sciences, Academic

Publisher; International Journal of Intelligent Systems and Applications in

Engineering (IJISAE); International Journal of Applied Mathematics, Electronics and

Computers (IJAMEC); International Journal of Energy Applications and Technology

(IJEAT) and SADAY Organization. They have made a crucial contribution towards

the success of this conference. Our thanks also go to the colleagues in our conference

office.

Looking forward to see you in next ICAT.

Ismail SARITAS - Omer Faruk BAY

Editors

International Conference on Advanced Technology & Sciences (ICAT’17) vii


PROGRAMME COMMITTEES

HONORARY CHAIR :

Mustafa Sahin, Rector of Selcuk University, Turkey

Senay Yalcin, Rector of Bahcesehir University, Turkey

GENERAL CHAIR :

Omer Faruk Bay, Gazi University, Turkey

Ismail Saritas, Selcuk University, Turkey

CO-CHAIR :

Ali Gungor, Bahcesehir University, Turkey

Kemal Tutuncu, Selcuk Universitesi, Turkey

Ilker Ali Ozkan, Selcuk University, Turkey

Murat Koklu, Selcuk University, Turkey

PUBLICATION CHAIR : Mehmet Akif Sahman, Selcuk University, Turkey

Ali Yasar, Selcuk University, Turkey

Mustafa Buber, Selcuk University, Turkey

INTERNATIONAL ADVISORY BOARD

Abdullah Erdal Tumer, Turkey

Alexander Sudnitson, Estonia

Ali Gungor, Turkey

Ali Yasar, Turkey

Alla Anohina Naumeca, Latvia

Almoataz Youssef Abdelaziz, Egypt

Amar Ramdane Cherif, France

Anand Kumar, India

Anca Loana Andreescu, Bulgaria

Angel Smrikarov, Bulgaria

Anne Villems, Estonia

Antonella Reitano, Italy

Antonio Mendes, Portugal

Artan Luma, Macedonia

Asrun Matthiasdottir, Iceland

Biagio Lenzitti, Italy

Binod Kumar, India

Boris Akanaev, Kazakhstan

Cemil Sungur, Turkey

Cesare Valenti, Italy

Cigdem Eroglu Erdem, Turkey

Desislava Paneva Marinova, Bulgaria

Dimitris Dranidis, Greece

Domenico Tegolo, Italy

Eisha Akanksha, India

Elinda Kajo Mece, Romania

Ercan Erturk, Turkey

Fecir Duran, Turkey

Gabriel Luna Sandoval, Mexico

Halife Kodaz, Turkey

Harun Uguz, Turkey

Hayri Arabaci, Turkey

Heinz Dietrich Wuttke, Germany

Howard Duncan, Ireland

Hulusi Karaca, Turkey

Huse Fatkic, Bosnia and Herzegovina

Ilker Ali Ozkan, Turkey

Inan Guler, Turkey

Irina Noninska, Bulgaria

Ismail Saritas, Turkey

viii International Conference on Advanced Technology & Sciences (ICAT’17)


Ivan Jelinek, Czech Republic

Jan Vom Brocke, Liechtenstein

Janis Grundspenkis, Latvia

Janusz Jablonowski, Poland

Jiri Srba, Denmark

Kadir Sabanci, Turkey

Karl Jones, United Kingdom

Kemal Tutuncu, Turkey

Korhan Cengiz, Turkey

Laurentiu Cristian Deaconu, Romania

Lilia Georgieva, United Kingdom

Luca Lombardi, Italy

Mahdi Shahbakhti, United States

Majida Ali Abed Meshari, Iraq

Manik Sharma, India

Marco Porta, Italy

Markus Helfert, Ireland

Mehmet Cunkas, Turkey

Mehmet Akif Sahman, Turkey

Mehmet Rahmi Canal, Turkey

Michail Antchev, Bulgaria

Mirjana Ivanovic, Serbia

Mohamed Abdel Fattah Ashabrawy

Moustafa, Saudi Arabia

Muciz Ozcan, Turkey

Muhammad Zia Ur Rehman, Pakistan

Murat Koklu, Turkey

Mustafa Buber, Turkey

Mustafa Altin, Turkey

Mustafa Alci, Turkey

Mustafa Nevzat Ornek, Turkey

Nafiz Arica, Turkey

Natasa Hoic Bozic, Croatia

Novruz Allahverdi, Turkey

Omer Faruk Bay, Turkey

Onur Inan, Turkey

Pantha Ghosal, Australia

Pino Caballero Gil, Spain

Polyxeni Arapi, Greece

Raif Bayir, Turkey

Ridvan Saracoglu, Turkey

Rositsa Doneva, Bulgaria

Saadetdin Herdem, Turkey

Sakir Tasdemir, Turkey

Silyan Sibinov Arsov, Bulgaria

Simona Silvia Merola, Italy

Stavros Christodoulakis, Greece

Stavros Nikolopoulos, Greece

Tatjana Dulinskiene, Lithuania

Thomas Engel, Luxembourg

Virginio Cantoni, Italy

Wladimir Bodrow, Germany

Yuri Pavlov, Bulgaria

Zarifa Jabrayilova, Azerbaijan

ORGANIZING COMMITTEE

Omer Faruk Bay, Turkey

Ali Gungor, Turkey

Nafiz Arica, Turkey

Cigdem Bakir, Turkey

Ercan Erturk, Turkey

Ismail Saritas, Turkey

Kemal Tutuncu, Turkey

Ilker Ali Ozkan, Turkey

Murat Koklu, Turkey

Silyan Sibinov Arsov, Bulgaria

Dimitris Dranidis, Greece

Angel Smrikarov, Bulgaria

Sakir Tasdemir, Turkey

Mustafa Altin, Turkey Ridvan

Saracoglu, Turkey

Cigdem Eroglu Erdem, Turkey

Zekai Sen, Turkey

Fatih Saday, Turkey

viii International Conference on Advanced Technology & Sciences (ICAT’17)


Alla Anohina Naumeca, Latvia Abdulkadir Saday, Turkey

TECHNICAL COMMITTEE

Mehmet Akif Sahman, Turkey

Ali Yasar, Turkey

Mustafa Buber, Turkey

Esra Kaya, Turkey

Fatih Saday, Turkey

Abdulkadir Saday, Turkey

ix International Conference on Advanced Technology & Sciences (ICAT’17)


CONTENTS

COMPARATIVE HDL IMPLEMENTATIONS OF TWO CHAOTIC GENERATORS 1

ZEYNEP KAYA, EROL SEKE

THE INFERENCE PROCESS OF NONSTATIONARY FUZZY SETS WITH

UNCERTAINTY FOR CYBER PHYSICAL SYSTEMS 2

HASAN YETIS, MEHMET KARAKOSE

A COMPARISON OF SHORT TERM LOAD FORECASTING METHODS FOR

TURKISH DISTRIBUTION GRID 3

UGURCAN SENCAN, GURKAN SOYKAN, NAFIZ ARICA

NON LOCALLY COLOR IMAGE SEGMENTATION FOR REMOTE SENSING

IMAGES IN DIFFERENT COLOR SPACES BY USING DATA CLUSTERING

METHODS

4

MURAT KARAKOYUN, ALI SAGLAM, NURDAN AKHAN BAYKAN, ADEM ALPASLAN

ALTUN

IMPROVED RAIL SURFACE DETECTION AND CONDITION MONITORING

APPROACH WITH FPGA IN RAILWAYS 5

ORHAN YAMAN, MEHMET KARAKOSE, ERHAN AKIN

SENTIMENT ANALYSIS ON TWITTER SOCIAL MEDIA DATA 6

ISMAIL ISERI, GOKHAN KAYHAN, SALIH CAN OZEL

AMBIENT MEASUREMENT PLATFORM 7

NUR IREM EDINC, MEHMET CAN ALPHAN, LUTFI ARDA

IMAGE ENCRYPTION BASED ON HAAR WAVELET TRANSFORMATION AND

RC4 ALGORITHM 8

ELHAM YASIN, RIDVAN SARACOGLU, PARISA LARIBI

DYNAMICAL STRESS FIELD PROBLEM IN A PRE STRESSED PLATE STRIP

MADE BY COMPOUNDING TWO LAYERS SIDE BY SIDE 9

AHMET DASDEMIR

DEVELOPING INTELLIGENT SYSTEMS WITH CAPACITIVE TOUCH SENSOR

USING WIRELESS SENSOR NETWORKS 10

MURAT DENER

ESTIMATION OF SHORT PERIOD WIND POWER WITH ANN METHOD 11

BULENT YANIKTEPE, SAKIR TASDEMIR, A BURAK GUHER

FUZZY POSITION CONTROL APPROACH FOR AN AUTONOMOUS ROBOT

CONTROLLER 12

ALI CETINKAYA, NOVRUZ ALLAHVERDI, AHMED YASSER AFAGHANI, TUGBA SARAY

COST EFFECTIVE ELASTIC MANAGEABLE AND SCALABLE STREAM DATA

ANALYTICS ON DATA DRIVEN ARCHITECTURES 13

UGUR ARPACI

GIS BASED WHEAT QUALITY MAP KONYA PROVINCE APPLICATION 14

ABDULLAH OKTAY DUNDAR, MAHMUT TEKIN, KENAN PEKER, MEHMET AKIF

SAHMAN

x International Conference on Advanced Technology & Sciences (ICAT’17)


COMPARISON HYBRID METHODS IN FEED FORMULATIONS 15

MEHMET AKIF SAHMAN, ADEM ALPASLAN ALTUN, ABDULLAH OKTAY DUNDAR

THE MOBILE APP TO ASSIST WITH TATTOO ENCODED BABY

IMMUNIZATION RECORDS IN AFRICA 16

BONGKEUM JEONG, JESUS IBANEZ, NUNO NUNES, MONCHU CHEN

A NEW APPROACH BASED ON IMAGE PROCESSING FOR MEASURING

COMPRESSIVE STRENGTH OF STRUCTURES

17

MEHMET BAYGIN, SUAT GOKHAN OZKAYA, MUHAMMET ALPEREN

OZDEMIR, ILKER KAZAZ

A REVIEW OF SMART HOME SYSTEMS 18

MOHAMMED QAIMAZ ALI, ISMAIL SARITAS

A MULTI PURPOSE MOBILE APPLICATION FOR EMBEDDED SYSTEMS 19

FECIR DURAN, HATICE TIRASOGLU

OBJECT DETECTION AND SEGMENTATION ON APPLE TREE IMAGES BY

USING MORPHOLOGICAL OPERATIONS AND VERTICAL AND HORIZONTAL

HISTOGRAM VALUES

20

SUMEYYA ILKIN, MERVE TAFRALI, SUHAP SAHIN, BURAK INNER

HOSPITAL PERFORMANCE EVALUATION A FUZZY DATA ENVELOPMENT

ANALYSIS APPROACH 21

ELIF DOGU, MICHELE CEDOLIN, Y ESRA ALBAYRAK

EVALUATION OF CUSTOMER SATISFACTION FACTORS IN HOUSING

INDUSTRY AND SELECTION OF THE BEST PROJECT USING AHP AND

TOPSIS METHODS

22

MICHELE CEDOLIN, ELIF DOGU, Y ESRA ALBAYRAK

CLASSIFICATION OF ECG SIGNALS BY THE ENSEMBLE LERARNING

METHODS 23

CIGDEM BAKIR

INVESTIGATION OF E LEARNING APPLICATIONS IN CIVIL ENGINEERING 24

ALI ERDEM CERCEVIK, NIHAN KAZAK

DATA MINING APPLICATIONS IN DIAGNOSIS OF SYMPTOMS 25

HARUN OZKISI, MURAT TOPALOGLU

USING RDF SUMMARY GRAPH FOR KEYWORD BASED SEMANTIC

SEARCHES 26

SERKAN AYVAZ, MEHMET AYDAR

DESIGN OF WEB BASED MONITORING SYSTEM FOR TOMCAT

APPLICATION SERVER 27

GOKHAN KAYHAN, ISMAIL ISERI, HAKAN KABASAKAL

A NEW STUDY OF K MEANS ALGORITHM IN BOSPHORUS AND

DARDANELLES STRAITS DATABASE 28

AYLA SAYLI, TUGCE SERT

DECISION TREE RESEARCH APPLICATIONS OF ID3 AND C4 5 ALGORITHMS 29

AYLA SAYLI, TUGCE SENTURK

xi International Conference on Advanced Technology & Sciences (ICAT’17)


MP DDOSER IP KILLER TESTING BEHAVIOR OF IP KILLER DDOS BOT IN

THE DETER TESTBED 30

TUAGHMOHAMMAD KALTE, GOKHAN DALKILIC

A DEEP LEARNING APPROACH FOR OPTIMIZATION OF SYSTEMATIC

SIGNAL DETECTION IN FINANCIAL TRADING SYSTEMS WITH BIG DATA 31

H SERCAN KARAOGLU, SERKAN AYVAZ, UGUR ARPACI

A WEB AND MOBILE BASED APPLICATION MONITORING AIR POLLUTION

INSTANT PARAMETERS USING CURL IN PHP 32

ABDURRAHMAN OZBEYAZ, YAVUZ DEMIRCI

USING DATA MINING AND FUZZY RANKING METHODS FOR DETERMINING

SIGNIFICANCE LEVELS OF ASSOCIATION RULES 33

MELDA KOKOC, ADNAN AKTEPE, SULEYMAN ERSOZ

CLASSIFICATION OF TREE SPECIES IN A FOREST BY ARTIFICIAL NEURAL

NETWORK USING ASTER SATELLITE IMAGES 34

MUSTAFA BUBER, MUHAMMED FAHRI UNLERSEN, KADIR SABANCI

DIAGNOSIS OF MESOTHELIOMA DISEASE USING DIFFERENT

CLASSIFICATION TECHNIQUES 35

KEMAL TUTUNCU, OZCAN CATALTAS

A SIMPLE APPROACH FOR INFLUENCE MAXIMIZATION IN SOCIAL

NETWORKS 36

AYBIKE SIMSEK, RESUL KARA

DESIGN AND IMPLEMENTATION OF AN ARTIFICIAL NEURAL NETWORK

SOFTWARE TO DIAGNOSE DISEASE 37

ISMAIL SARITAS, KADIR TOK

INFORMATION SECURITY A HYBRID ALGORITHM STUDY 38

EGEMEN TEKKANAT, MURAT TOPALOGLU

THE FUNCTION ANALYSIS AND ITS APPLICATION TO DETERMINATION

OF FUZZY SUBSETS FOR MODEL HELICOPTER FUZZY CONTROL 39

MURAT HACIMURTAZAOGLU, NOVRUZ ALLAHVERDI

FRIENDSHIP ALGORITHM FOR DETECTING BLACK HOLE ATTACKS IN

MANET FADBAM 40

ADWAN YASIN, MAHMOUD ABU ZANT

HUMAN LIMB MOVEMENT ESTIMATION BY ARTIFICIAL NEURAL

NETWORK USING BRAIN WAVE SIGNALS 41

MURAT KOKLU, KADIR SABANCI, MUHAMMED FAHRI UNLERSEN

MALICIOUS WEB PAGE DETECTION WITH MACHINE LEARNING

TECHNIQUES 42

CANSU KADI, URAZ YAVANOGLU

IMPLEMENTATION OF 16 BIT ARITHMETIC LOGIC UNIT ON FPGA 43

ILKER ALI OZKAN, ABDULKADIR SADAY

A SUGGESTION PERFORMING AUTOMATIC BLOCK REALLOCATION

USING PARTICLE SWARM OPTIMIZATION IN LAND CONSOLIDATION

PROJECTS

44

ABDURRAHMAN OZBEYAZ, YASAR INCEYOL

xii International Conference on Advanced Technology & Sciences (ICAT’17)


A TOOL FOR COLLECTING DATA FROM THE DEVICES USING SNMP

AGENTS 45

ATIKE DEMIRBAS, YALCIN CEBI

FEATURE SELECTION FROM 3D BRAIN MODEL FOR SOME DEMENTIA

SUBTYPES USING GENETIC ALGORITHM 46

SAVAS OKYAY, NIHAT ADAR, KEMAL OZKAN, BAKI ADAPINAR

INTELLIGENT BACKUP SYSTEM BASED ON ARM MICROCONTROLLER 47

FATIH SADAY, ABDULKADIR SADAY

A CHANCE PLAY DESIGN WITH C EXAMPLE OF HORSE RACE

APPLICATION 48

SAKIR TASDEMIR, EMRE AVUCLU

A NOVEL OPTIMIZATION APPROACH BASED ON RECIPROCAL FORCES

FOR ASTRONOMICAL N BODY SIMULATION ON GPU 49

CELIL OZKURT, EYUP GEDIKLI

MULTISPECTRAL IMAGE FUSION USING A NEW TENSOR APPROACH 50

SAHIN ISIK, KEMAL OZKAN, GOLARA GHORBAN DORDINEJAD

DETECTION OF FOREIGN OBJECTS IN RAILWAY LEVEL CROSSINGS

USING IMAGE PROCESSING TECHNIQUES 51

CANAN TASTIMUR, MEHMET KARAKOSE, ERHAN AKIN

PREDICTION OF ION EXCHANGE EFFICIENCY OF COMERCIAL RESINS

STRONG BASE ION EXCHANGE RESINS WITH ARTIFICIAL NEURAL

NETWORKS

52

SERPIL EDEBALI, ABDULLAH ERDAL TUMER, MEHMET ALKANER

TRAINING OF ANFIS NETWORK BY GENETIC ALGORITHM FOR

DIAGNOSIS OF LEUKEMIA CANCER SUBTYPES USING GENE EXPRESSION

PROFILE

53

MUSTAFA TURAN ARSLAN, BULENT HAZNEDAR, ADEM KALINLI

AN INTELLIGENT CLASSIFICATION APPROACH FOR SOCIAL MEDIA DATA 54

FATIH ERTAM

DESIGN AND IMPLEMENTATION OF AN INTERPRETER FOR THE LEAST

SQUARES METHOD USING SYMBOLIC APPROACHES 55

NAWAL ABDULLAHI MOHAMED, HUSEYIN PEHLIVAN

THE ROLE OF INNOVATIVE TECHNOLOGY-INTEGRATED TEACHING :

TOWARDS A TRANSFORMATIVE PEDAGOGY 56

NAZ ONEL

TRAINING OF ARTIFICIAL NEURAL NETWORK USING METAHEURISTIC

ALGORITHM 57

SHAIMAA SAFAA AHMED ALWAISI, OMER KAAN BAYKAN

CUSTOMER CHURN ANALYSIS USING CLASSIFICATION ALGORITHMS IN

TELECOMMUNICATIONS INDUSTRY 58

MUHAMMET SINAN BASARSLAN, FATIH KAYAALP

IMPLEMENTATION AND EVALUATION OF FACE RECOGNITION BASED

IDENTIFICATION SYSTEM 59

FARUK CAN ELBIZIM, MUSTAFA CEM KASAPBASI

xiii International Conference on Advanced Technology & Sciences (ICAT’17)


REDUCING CBR DROPPING IN QOSBOX USING GREENCLOUD

SIMULATION IN CLOUD COMPUTING 60

SHAFIULLAH MALEKZAI, DOGAN YILDIZ, SERAP KARAGOL

BEACON BASED INDOOR POSITIONING AND ROUTING 61

TANER ARSAN

DETERMINING OPTIMAL NUMBER OF CLUSTERS IN LEACH LIKE

PROTOCOLS 62

SINAN TOKLU, MEHMET SIMSEK

BIG DATA AND RESTFUL BASED WEB APPLICATION FOR SMART HEALTH

APPLICATION 63

SINEM GUVEN SANTUR, YUNUS SANTUR, MEHMET KARAKOSE

DEEP LEARNING BASED ARTIFICIAL MANAGER FOR SMART CITY 64

YUNUS SANTUR, EBRU KARAKOSE, MEHMET KARAKOSE, ERHAN AKIN

RE ANALYSIS OF MOLD MAKING ERRORS WITH COMPUTER AIDED

DESIGN PROGRAMS AND RECOMMENDATIONS

65

MUSTAFA ALTIN, AZIZ TOLUNAY ARSLANBAS, SAKIR TASDEMIR

A COMPUTER ALGEBRA SYSTEM THAT CAN BE INTEGRATED INTO USERS

WORKING ENVIRONMENTS TO SUPPORT SYMBOLIC COMPUTATION

OPERATIONS

66

SEDA EFENDIOGLU, H EMRE EFENDIOGLU, HUSEYIN PEHLIVAN

COMPARISON THE PERFORMANCES OF DBSCAN CLUSTERING

ALGORITHM AND IMAGE PROCESSING TECHNIQUES FOR COUNTING

WHITE BLOOD CELLS

67

FATMA OZGE OZKOK, METE CELIK

THE DIAGNOSIS OF MELANOMA SKIN CANCER USING SOME FEATURE

DESCRIPTORS 68

KAYA TURGUT, CAGRI KILINC, SEMIH ERGIN

COMPARISON OF LINEAR DIMENSIONALITY REDUCTION METHODS ON

FACE RECOGNITION 69

ERAY YILDIZ, YUSUF SEVIM

ANALYSIS OF LRC PERFORMANCE USING LBP FEATURES IN FACE

RECOGNITION 70

SULENUR ERGUL, MEHMET KOC

DEVELOPMENT OF OPTICAL CHARACTER RECOGNITION MOBILE

APPLICATION OVER CARD VISITS WITH IMAGE PROCESSING METHODS 71

ISMAIL ISERI, GOKHAN KAYHAN, KEREM HACIFAZLIOGLU

FEATURE WEIGHTING USING GENETIC ALGORITHM FOR TEXT

CLASSIFICATION 72

PARISA LARIBI, RIDVAN SARACOGLU

HOW DO COLLEGE STUDENTS ADOPT SOCIAL NETWORKING SERVICES

UNDERSTANDING THE DETERMINANTS 73

HASAN USLU, MUSTAFA EREN YILDIRIM, YUCEL BATU SALMAN

xiv International Conference on Advanced Technology & Sciences (ICAT’17)


A NEW FAST MEASUREMENT AND INSPECTION APPROACH BASED ON

MACHINE VISION 74

MEHMET BAYGIN, MEHMET KARAKOSE, ALISAN SARIMADEN, ERHAN AKIN

OPTIMAL ROUTE DETECT BASED ON GENETIC ALGORITHM FOR

ERZURUM METROPOLITAN MUNICIPALITY 75

MEHMET BAYGIN, SALIH SERKAN KALELI, ABDULLAH NARALAN

A NEW SOFTWARE DESIGN FOR MUDEK ACCREDITATION PROCESS THE

APPLICATION OF KIRIKKALE UNIVERSITY 76

ATILLA ERGUZEN, SEYMA CIHAN, VOLKAN ATES, ERTUGRUL CAM

PREDICTION OF HEATING VALUES OF REFUSE DERIVED FUEL FROM

PROXIMATE ANALYSIS DATA 77

KEMAL OZKAN, SAHIN ISIK, AYSUN OZKAN, ZERRIN GUNKAYA, MUFIDE BANAR

ANALYSIS OF REAL TIME APPLICATION RESULTS OF BAYESIAN AND

ARTIFICIAL NEURAL NETWORKS ANN ALGORITHMS IN PRODUCING

CREDIT SCORE CARD

78

ALI TUNC, HUSEYIN KARADAG

THE TRUST MODEL FOR EVALUATION OF CLOUD SERVICE PROVIDERS

79

ATILA BOSTAN, DENIZ KORAY INCELER

SEED QUALITY CONTROL BY USING ARTIFICIAL NEURAL NETWORK 80

KADIR SABANCI, MUHAMMED FAHRI UNLERSEN, MURAT KOKLU

LOCAL T1 DISTANCES APPROACH SPACES 81

MEHMET BARAN, MUHAMMAD QASIM

DETERMINATION OF PLANT SPECIES USING VARIOUS ARTIFICIAL

NEURAL NETWORK STRUCTURES 82

MUHAMMED FAHRI UNLERSEN, MURAT KOKLU, KADIR SABANCI

COMPARISON OF OPTIMIZATION METHODS BASED ON GSA 83

NIHAN KAZAK, NESIBE YALCIN, ALI ERDEM CERCEVIK

A SMART PARKING LOT SENSOR SYSTEM 84

PINAR KIRCI, SELEN SAGLAMOZ, MURAT SENER

EMPIRICAL EXPLORATION FOR THE CORRELATION BETWEEN OBJECT

ORIENTED CLASS COHESION AND INHERITANCE 85

JEHAD AL DALLAL

A SMART CAR PARKING SYSTEM BASED ON GPS COORDINATES 86

GOKHAN KAYHAN, ISMAIL ISERI, SEVKI UTKU MALKOCOGLU

QR CODE BASED ENCRYPTION AND DECRYPTION OF TRIANGULAR

GEOMETRY PROBLEMS 87

CHEIKHNA LO

COMPARISION OF ORGANIZED INDUSTRIAL ZONES EFFICIENCY BY

USING A COMMON WEIGHT DEA BASED DECISION MODEL 88

BURCU DENIZ TULEK, NAZLI GOKER, MEHTAP DURSUN

MODELING OF SOCIAL PERFORMANCE METRICS WITH ARTIFICIAL

INTELLIGENCE TECHNIQUES 89

ENES CELIK, ADIL KONDILOGLU, HARUN BAYER, MUHAMMET ATALAY

xv International Conference on Advanced Technology & Sciences (ICAT’17)


IMPLEMENTATION OF DEFECT DETECTION ON MILITARY CARTRIDGE

WITH OPENMP 90

OMER FARUK BAY, SEMRA AYDIN, REFIK SAMET

NEAR FIELD COMMUNICATION BASED ROLL CALL SYSTEM 91

SEYIT ALPEREN CELTEK, MAHMUT DURGUN, KADIR SABANCI

SELECTION OF A SUITABLE WIND POWER PLANT WITH ECONOMICAL

ANALYSIS FOR AMASRA CITY TURKEY 92

YAGMUR ARIKAN, OZGE PINAR AKKAS, ERTUGRUL CAM

IN SITU UV EXCITATION STUDY ON MATRIX ISOLATED 2 CHLORO 6

FLUOROBENZOIC ACID 93

NIHAL KUS, RUI FAUSTO

P TYPE DYE SENSITIZED SOLAR CELLS ENHANCED PERFORMANCE WITH

A GRAPHENE CUO NANOSTRUCTURES 94

BAYRAM KILIC, SUNAY TURKDOGAN

INVESTIGATION OF FIRE EXTINGUISHER TUBE STRESS AND

DEFORMATION CONCEPTS UNDER THE OVERPRESSURE LOADING FOR

DIFFERENT MATERIALS

95

MEHMET HANIFI DOGRU, IBRAHIM GOV, KURSAD GOV, EYUP YETER, IBRAHIM

HALIL GUZELBEY

VOLTAGE BREAKDOWN AND LEAKAGE CURRENT STUDIES OF TIO2

AL2O3 CERAMIC NANO COMPOSITE CAPACITOR 96

ALARA ULAC, DOGAN AKCAN, MEHMET CAN ALPHAN, LUTFI ARDA, ALI GUNGOR

REACTION OF ACETOPHENONE FORMALDEHYDE RESIN WITH S

TRIAZONE COMPOUNDS FOR FIRE RESISTANCE POLYURETHANE 97

MINE MUTLU, NILGUN KIZILCAN, AHMET AKAR, NESRIN KOKEN

USING A NEW METHOD BASED ON FINSLER GEOMETRY FOR WIND SPEED

MODELLING 98

EMRAH DOKUR, SALIM CEYHAN, MEHMET KURBAN

EFFECT OF INDIUM INCORPORATION ON THE STRUCTURAL AND

OPTICAL PROPERTIES OF ZNS NANOSTRUCTURED FILMS USED AS

WINDOW LAYER IN SOLAR CELLS

99

TULAY HURMA

SYNTHESIS OF GLYCOLURIL DERIVATIVES FOR FIRE RESISTANCE

POLYURETHANE 100

NILGUN KIZILCAN, ESRA TUNCEL, AHMET AKAR

EXPERIMENTAL INVESTIGATION OF HEAT STORAGE BEHAVIOUR IN A

PHASE CHANGE MATERIAL UNDER SOLAR RADIATION 101

MEHMET EMIN CANLI, ADEM ACIR

INVESTIGATION OF NONLINEAR PATHS OF FLUIDS FLOW DETERMINED

BY FINSLER GEOMETRY THROUGH INHOMOGENEOUS POROUS MEDIA 102

DERYA ULUG, SALIM CEYHAN

THE EFFECTS OF INCENTIVES ON RENEWABLE ENERGY RESOURCES FOR

HOME USERS 103

FATIH BULUT, MURAT LUY, ERTUGRUL CAM

xvi International Conference on Advanced Technology & Sciences (ICAT’17)


FREQUENCY MODULATED IMPULSE RESPONSE PHOTOTHERMAL

REFLECTANCE THEORY OF JUNCTION VERTICAL PARALLEL SILICON

SOLAR CELL

104

GOKHAN SAHIN

TRANSMISSION EXPANSION PLANNING INCLUDING POWER LOSS COST

USING LINEARIZED AC MODEL 105

MEHMET FATIH CANKURTARAN, FARUK UGRANLI, ENGIN KARATEPE

EFFECT OF TRISO FUEL OF THE BURNING OF REACTOR GRADE

PLUTONIUM IN CANDU REACTOR 106

ADEM ACIR, TAHA HUSEYIN MERMER

SELECTION OF A SOLAR POWER PLANT LOCATION BY USING AHP

METHOD 107

OZGE PINAR AKKAS, YAGMUR ARIKAN, ERTUGRUL CAM

DETERMINATION OF SOME CHARACTERISTIC PARAMETERS OF

PHOTOVOLTAICS USING DIFFERENT ANALYTICAL METHODS 108

AYSEGUL TOPRAK, HAMDI SUKUR KILIC

SERVO MOTOR DRIVEN HUMAN GYROSCOPE STRUCTURE FLIGHT

SIMULATOR 109

KURSAD GOV, EYUP YETER, MEHMET HANIFI DOGRU, IBRAHIM GOV, IBRAHIM

HALIL GUZELBEY

SHORT TERM FORECASTING OF RENEWABLE ENERGY PRODUCTION

ACCORDING TO DAILY DATA USING ARTIFICIAL NEURAL NETWORK 110

BAYRAM AKDEMIR, FATIH MEHMET SENALP

COMPARISON BETWEEN HORIZONTAL AND VERTICAL PARALLEL

SILICON SOLAR CELL OF A PHOTOCURRENT DENSITY UNDER

MULTISPECTRAL ILLUMINATION

111

GOKHAN SAHIN

FATIGUE ANALYSIS OF LANDING GEAR OF A LIGHTWEIGHT AIRPLANE 112

IBRAHIM GOV, MEHMET HANIFI DOGRU, EYUP YETER, KURSAD GOV, IBRAHIM

HALIL GUZELBEY

STRESS ANALYSES OF DIFFERENT MECHANIC CONNECTION SHAPES

USED TO CONNECT COMPOSITE STRUCTURES 113

EYUP YETER, KURSAD GOV, IBRAHIM GOV, MEHMET HANIFI DOGRU, IBRAHIM

HALIL GUZELBEY

ENHANCEMENT OF PYROPLASTIC DEFORMATION BEHAVIOUR IN

CERAMIC SANITARYWARE FFC SLIPS 114

NAZIM KUNDURACI, IREM NUR GAMZE SIMSEK, ENGIN KOCAMAN

DESIGN AND IMPLEMENTATION OF 3D VISUAL LOCALIZATION SYSTEM

FOR QUADROTORS 115

TOLGA YUKSEL

A NEW BROADBAND REFLECTARRAY ANTENNA WITH MODIFIED

CIRCULAR PATCH 116

HANDE BODUR, SIBEL UNALDI, SIBEL CIMEN, GONCA CAKIR

xvii International Conference on Advanced Technology & Sciences (ICAT’17)


A LOW PROFILE MINIATURIZED BAND PASS FREQUENCY SELECTIVE

SURFACE WITH ANGULAR STABILITY 117

AYBIKE KOCAKAYA, GONCA CAKIR

DESIGN AND OPTIMIZATION OF FRACTIONAL SLOT PERMANENT

MAGNET ASSISTED SYNCHRONOUS RELUCTANCE MOTOR FOR WASHING

MACHINE

118

OMER FARUK PAYZA, GAMZE ODABAS, YASEMIN ERTUGRUL

POSITION CONTROL OF A BALL BEAM EXPERIMENTAL SETUP BASED ON

SLIDING MODE CONTROLLER 119

KAAN CAN, ABDULLAH BASCI

COMPARISON OF FRACTIONAL PI AND CLASSICAL PI CONTROLLER

BASED POSITION CONTROL OF BALL AND BEAM SYSTEM 120

KAAN CAN, ABDULLAH BASCI, BIROL SOYSAL, ADNAN DERDIYOK

MEANWHILE ANATOMY OF GASES WITH STUBBORN GAS SENSORS TO

INCREASE THE DETECTION EFFICIENCY 121

ALI MARDAN HAMEED QUTUB, ISMAIL SARITAS

WIND GENERATION ESTIMATION BASED ON WEATHER FORECASTING IN

KIRIKKALE TURKEY 122

MUSTAFA YASIN ERTEN, HUSEYIN AYDILEK, NIHAT INANC

THE INTERACTION OF THEORETICAL COURSES WITH PRACTICALS AND

THE EFFECTS ON SUCCESS CASE OF AN ELECTRONIC LECTURE 123

PIRIL TEKIN, RIZVAN EROL

IMAGE PROCESSING BASED WEATHER PREDICTION AND OUTDOOR

ACTIVITY DECISION FRAMEWORK WITH SUPPORT VECTOR MACHINES 124

HUSEYIN AYDILEK, MUSTAFA YASIN ERTEN, NIHAT INANC

EFFECTS OF INVERTER TOPOLOGIES AND MODULATION TECHNIQUES

ON POWER QUALITY ISSUES ON GRID CONNECTED PHOTOVOLTAIC

SYSTEMS

125

MEHMET CAN ALPHAN, GURKAN SOYKAN, LUTFI ARDA

ECONOMIC EVALUATION OF SOLAR PHOTOVOLTAIC SYSTEMS FOR A

SMART HOME 126

AHMET NUR, ABDULCELIL BUGUTEKIN

HISTOGRAM BASED VEHICLE LICENSE PLATE RECOGNITION WITH KNN

METHOD 127

SECKIN KARASU, AYTAC ALTAN, ZEHRA SARAC, RIFAT HACIOGLU

ESTIMATION OF FAST VARIED WIND SPEED BASED ON NARX NEURAL

NETWORK BY USING CURVE FITTING 128

SECKIN KARASU, AYTAC ALTAN, ZEHRA SARAC, RIFAT HACIOGLU

SIMULATION OF BROKEN ROTOR BAR FAULT IN INDUCTION MOTORS 129

CEMIL KOZKURT, MEHMET AKAR

RELIABILITY ANALYSIS OF ATAKUM SAMSUN DISTRUBUTION SYSTEM

BY USING MONTE CARLO SIMULATION 130

FATIH DURMUS, CAGRI KOCAMAN

xviii International Conference on Advanced Technology & Sciences (ICAT’17)


DESIGN OF A 400 HZ CURRENT SOURCE 1 PH INVERTER TOPOLOGY FOR

AVIONIC SYSTEMS 131

ERALP SENER, GURHAN ERTASGIN

DECOMPOSING LIGHTING COMPONENT OF AN IMAGE A REVIEW 132

UMIT CIGDEM TURHAL, YAVUZ BUYUKKOCAK

DESIGN AND ACTUALISATION OF EOG BASED INTERACTIVE

COMMUNICATIONS FOR ALS PATIENTS 133

UGUR FIDAN, NESE OZKAN, ALI ISIK

EFFECT OF THICKNESS ON THE STRUCTURAL AND OPTICAL

PROPERTIES OF SILAR DEPOSITED CUS THIN FILMS 134

TUBA CAYIR, YUNUS AKALTUN

BARRIER FREE DOOR FOR INDIVIDUALS WHO HAVE A PHYSICAL AND

VISUAL INSUFFICIENCY 135

FECIR DURAN, TURKAN TEKE, MUSTAFA TEKE, MEHMET GUNES

DESIGN OF DIFFERENTIAL RELAY APPLICATION WITH MATLAB

SIMULINK FOR PROTECTION OF TRANSFORMER IN ELECTRIC ARC

FURNACE EAF LOADS

136

MUSTAFA SEKER, MURAT CIKAN, ARIF MEMMEDOV, SERDAR KOCKANAT, HULYA

DOGAN

SYNCHRONIZATION OF THE CHAOTIC SYSTEM WITH GOLDEN

PROPORTION EQUILIBRIA ACTIVE CONTROL VERSUS SLIDING MODE

CONTROL

137

YILMAZ UYAROGLU, UGUR ERKIN KOCAMAZ

LYAPUNOV BASED NONLINEAR CONTROL OF THE LOTKA VOLTERRA

SYSTEM WITH FOUR COMPETITIVE SPECIES 138

GULTEKIN CAGIL, UGUR ERKIN KOCAMAZ, YILMAZ UYAROGLU

MULTIPLE TARGET TRACKING SIMULATION AT AN INTERSECTION

139

AYKUT GORKEM GELEN, AYTEN ATASOY

APPLICATION OF MULTI AGENT SYSTEMS IN MICROGRID CONTROL 140

ZIYODULLA YUSUPOV, ANIS AMHARIB, NURI ALMARGANI

DEMAND SIDE MANAGEMENT APPLICATION FOR THE KIRIKKALE

UNIVERSITY TURKEY 141

ERTUGRUL CAM, VOLKAN ATES, AYSEGUL AKSAKAL, YAGMUR ARIKAN

INVESTIGATION OF CURRENT AND VOLTAGE FLUCTUATIONS IN

ELECTRIC ARC FURNACES WITH THE HELP OF PROBABILITY THEORY 142

MUSTAFA SEKER, ARIF MEMMEDOV, MURAT CIKAN, HULYA DOGAN, SERDAR

KOCKANAT

SERIAL IN PARALLEL OUT DISCRETE FOURIER TRANSFORM 143

EROL SEKE, ZEYNEP KAYA

DESIGN OF X BAND SINGLE LAYER REFLECTARRAY WITH NOVEL

ELEMENT 144

SIBEL UNALDI, HANDE BODUR, SIBEL CIMEN, GONCA CAKIR

xix International Conference on Advanced Technology & Sciences (ICAT’17)


TIME RESOLVED FLUORESCENCE TECHNIQUE FOR STUDYING

SWELLING PROCESS IN HYDROGELS 145

DEMET KAYA AKTAS, FILIZ OZTEKIN

BACKGROUND SUBTRACTION TECHNIQUES TO DETECT HUMAN VITAL

SIGNS USING UWB GPR 146

CANSU BUYUKHAN, SAEID KARAMZADEH, ALPER CALISKAN, MELEK

ORHAN, INCI CILESIZ

DETECTION OF LESION BOUNDARIES IN BREAST MAGNETIC RESONANCE

IMAGING WITH OTSU THRESHOLDING AND FUZZY C MEANS 147

GOKCEN CETINEL, SEVDA GUL

PERFORMANCE COMPARISON OF REGULAR AND IRREGULAR PILOT

PLACEMENTS IN OFDM SYTEMS 148

CETIN KURNAZ, BEGUM KORUNUR ENGIZ, FATIH AYVENLI

AN AUTOMATIC SYSTEM FOR SEGMENTATION OF AGGREGATE AND

BITUMEN IN ASPHALT CONCRETE IMAGES 149

ONDER CIVELEK, MEHMET OZTURK

A WIRED SENSOR NETWORK FOR THE PURPOSE OF MULTI DETECTION

SYSTEMS 150

AYHAN YAZGAN, OGUZHAN CAKIR, HUSEYIN ATASOY

CURRENT VOLTAGE CURVE PREDICTION OF ANNEALED NI N GAAS SI IN

SCHOTTKY BARRIER DIODE USING ANFIS 151

SERDAR KOCKANAT, HULYA DOGAN, YAVUZ TURKAY, MUSTAFA SEKER, DERYA

BETUL UNSAL CELIMLI, MURAT CIKAN

DESIGN AND SIMULATION OF MINIATURE MICROSTRIP ANTENNA USING

CIRCULAR MUSHROOM EBG STRUCTURE FOR WLAN PAN APPLICATION 152

BELABBAS KHEDEDJA, KHEDROUCHE DJAMEL, HOCINI

ABDESSELAM, BOUGOUTAIA TAHAR

RAPID PROTOTYPING LOW COST SOLAR SIMULATOR FROM OF THE SELF

COMPONENTS

153

LUTFI ARDA, M C ALPHAN, D AKCAN, ALI GUNGOR

PID TYPE STR BASED ON SVR FOR NONLINEAR SYSTEMS 154

KEMAL UCAK, GULAY OKE GUNEL

THE EFFECT OF DISTURBANCES ON GRID TIED POWER CONVERTERS

AND DSOGI BASED POWER CONTROLLER SOLUTION 155

DOGAN CELIK, M EMIN MERAL

PREVENTIVE PROPERTIES OF METAL COMPOSITE FABRICS AGAINST

ELECTROMAGNETIC RADIATION 156

RUSLAN ABDULLA

AN EXPERIMENTAL COMPARISON AND PERFORMANCE ASSESSMENT OF

NONLINEAR CONTROL TECHNIQUES FOR AN INDUSTRIAL AIR HEATING

SYSTEM

157

NECDET SINAN OZBEK, ILYAS EKER

xx International Conference on Advanced Technology & Sciences (ICAT’17)


THD COMPARISON OF VENTURINI AND SVM ALGORITHMS OF MATRIX

CONVERTERS FOR VARYING SWITCHING FREQUENCY 158

TOLGA SEN, ATA SEVINC

GENETIC ALGORITHM BASED FUZZY LOGIC SYSTEM FOR SHORT TERM

LOAD FORECASTING 159

MURAT LUY, VOLKAN ATES, NECAATTIN BARISCI, ERTUGRUL CAM

DESIGN AND MOTION EQUATIONS OF A ROBOT DEVELOPED FOR USE IN

LAPAROSCOPIC SURGERY 160

SERHAT AKSUNGUR, OGUZ YAKUT

INDUCTANCE AND LOSS BEHAVIORS OF MEDIUM FREQUENCY HIGH

POWER GAPPED CORE INDUCTORS 161

SELAMI BALCI, M BAHA BAYRAM, NECMI ALTIN, IBRAHIM SEFA

PARTIAL GAIN INCREASE AT DESIRED FREQUENCY RANGE FOR UWB

MICROSTRIP ANTENNAS USING BAND STOP FSS 162

KOMRON HOTAMI, CENGIZHAN MUSTAFA DIKMEN, GONCA CAKIR, AYBIKE

KOCAKAYA, SIBEL CIMEN

FRACTIONAL ORDER DISCRETE TIME MODEL ORDER REDUCTION BASED

ON SVD APPROACH 163

KARIMA GHARBI, DAOUD BERKANI

ADAPTIVE CRUISE CONTROL SYSTEM FOR ELECTRIC VEHICLES USING

FUZZY LOGIC 164

FATIH KARAGOZ, MUZEYYEN SARITAS

MULTIMODE LASER DIODE DRIVER DESIGN FOR PHOTODYNAMIC

THERAPY 165

MUHAMMED GULER, ALI FURKAN KAMANLI, MUSTAFA ZAHID YILDIZ, AYLA

EREN OZDEMIR, HYUN SOO LIM

REVIEW OF EXTREMELY LOW FREQUENCY (ELF) ELECTROMAGNETIC

FIELD MITIGATION TECHNIQUES 166

KENAN OZEL, ILHAN KOSALAY

GRAPHICAL STABILIZATION APPROACH FOR UNSTABLE FIRST ORDER

LINEAR SYSTEMS WITH TIME DELAY 167

ABDELKADER MADDI, ABDERREZAK GUESSOUM, DAOUD BERKANI

THROUGHPUT ENHANCEMENT OF MULTIPATH TCP BY SUBFLOW

BANDWIDTH MEASUREMENT IN WIRELESS NETWORKS

168

MIN SUB KIM, KI MOON HAN, JAE YOUG LEE, BYUNG CHUL KIM

A MOBILE APPLICATION DESIGN FOR INTRODUCTION OF DEVICES OF

THE PHYSICS LABORATORY 169

NIGMET KOKLU, DUNDAR YENER

DESIGN OF INTERACTIVE E BOOK APPLICATION FOR PHYSICS COURSE 170

NIGMET KOKLU, DUNDAR YENER

INVESTIGATION OF ATMOSPHERIC ATTENUATION FACTOR FOR HIGH

SPEED TERAHERTZ WIRELESS COMMUNICATION LINKS AT 0 275 THZ 171

AYHAN YAZGAN

xxi International Conference on Advanced Technology & Sciences (ICAT’17)


ANNEALING TEMPERATURE DEPENDENT CHARACTERIZATION OF NI AU

N GAN SCHOTTKY DIODE USING ARTIFICIAL BEE COLONY ALGORITHM 172

HULYA DOGAN, SERDAR KOCKANAT, YAVUZ TURKAY, MUSTAFA SEKER, MURAT

CIKAN, DERYA BETUL UNSAL CELIMLI

PERFORMANCE COMPARISON BETWEEN IP AND MPLS NETWORKS USING

OPNET MODELER 173

ALI GEZER, MARWA KHALEEL

CONTROL OF TRIGLIDE ROBOT WITH ARTIFICIAL NEURAL NETWORKS 174

MUHAMMET AYDIN, OGUZ YAKUT

CENTRE OF MASS MOTION AND THE MOTT TRANSITION IN LIGHT

NUCLEI 175

AREEJ ABDELRAHMAN, MAHMOUD STATY, HENRRY JAQAMAN

CLASSIFICATION OF COMMONLY HEARD LUNG SOUNDS RECORDED IN

SINGLE CHANNEL 176

SELIM ARAS, MEHMET CINEL, ALI GANGAL

DOGAN YILDIZ, SERAP KARAGOL

SELJUK STAR PATCH ANTENNA DESIGN ON WOOLENFELT MATERIAL

FOR WEARABLE 2400 MHZ ISM BAND APPLICATIONS 178

DILEK UZER, SEYFETTIN SINAN GULTEKIN, RABIA TOP

A COMPARATIVE STUDY FOR COOPERATIVE MISSION PLANNING OF UAV

SWARM 179

EBRU KARAKOSE

DETERMINING TWO COMMON POWER QUALITY DISTURBANCES

VOLTAGE SAG AND VOLTAGE SWELL 180

CAGRI KOCAMAN, MUAMMMER OZDEMIR

TIME FREQUENCY GRAY LEVEL CO OCCURRENCE MATRIX

DESCRIPTORS FOR DECEPTION DETECTION 181

ABDULKADIR SENGUR, SAMI EKICI, YAMAN AKBULUT, TURGUT KAVAS

CHARACTERIZATION OF ZNO THIN FILMS DEPOSITED BY PULSED LASER

DEPOSITION PLD TECHNIQUE 182

AHMET TOPRAK, SERAP YIGIT GEZGIN, HAMDI SUKUR KILIC

IMPACT OF BACKGROUND LANGUAGE ON TURKISH TEXT DEPENDENT

SPEAKER VERIFICATION 183

HAVVA CELIKTAS, CEMAL HANILCI

A NEW MODIFICATION ON FORMULATION OF BROADBAND IMPEDANCE

MATCHING 184

METIN SENGUL

PROPERTIES OF MULTILAYERED MG BASED GUIDED MODE RESONANCE

FILTERS GMRFS 185

DUAA ABUMURRA, EDWARD SADER

OBTAINING THE EFFECTS OF THE VOLTAGE SAG DURATION ON

FLUORESCENT LAMPS 186

MEHLIKA SENGUL

xxii International Conference on Advanced Technology & Sciences (ICAT’17)


A MICROSTRIP PATCH ANTENNA DESIGN TO DETERMINE THE LEVEL OF

OCCLUSION IN HEART VEIN 187

RABIA TOP, SEYFETTIN SINAN GULTEKIN, DILEK UZER

IMPROVEMENT OF A SOLENOID FOR ELECTROMECHANICAL VALVE

EMV SYSTEMS 188

NIDA BIRGUL

UNDERSTANDING OF BIPOLAR DISORDER ON RAT BEHAVIORS VIA

NONLINEAR STATE ESTIMATION 189

AYSEN CETIN KARDESLER, MERIC CETIN, MEHMET DOGAN ELBI

PETRI NETS MODELLING OF SMART AGRICULTURAL SYSTEM 190

HANIFE APAYDIN OZKAN

PHOTOELECTRICAL PARAMETER CALCULATION OF METAL ORGANIC

INORGANIC PHOTO DIODE USING AN APPROACH BASED MABC

ALGORITHM

191

HULYA DOGAN, SERDAR KOCKANAT, YAVUZ TURKAY, MURAT CIKAN, DERYA

BETUL UNSAL CELIMLI

ELECTRICAL PARAMETER DETERMINATION OF AL P SI SCHOTTKY

DIODE FROM I V AND C V CURVES USING MABC BASED METHOD 192

SERDAR KOCKANAT, HULYA DOGAN, YAVUZ TURKAY, MURAT CIKAN, DERYA

BETUL UNSAL CELIMLI

LEACH AND AHP BASED CLUSTERING ALGORITHM FOR WIRELESS

SENSOR NETWORKS 193

GOKCE HACIOGLU, ERHAN SESLI, VOLKAN GANGAL, VOLKAN AYDIN

IMPROVEMENT OF AN ELECTROMECHANICAL VALVE EMV SYSTEM

DYNAMIC ANALYSIS AND DETERMINATION OF WORKING LIMITS AT

DIFRENT LIFTINGS

194

NIDA BIRGUL

A COMPARISON OF PSO GA AND NELDER MEAD OPTIMIZATION

ALGORITHMS ON BUSBAR TRANSMISSION LINE PARAMETER

EXTRACTION

195

ZEYNEP HASIRCI, MEHMET OZTURK, RIFAT BENVENISTE, HAYDAR KAYA, ISMAIL

HAKKI CAVDAR

COMPARATIVE PERFORMANCE EVALUATION OF MULTI SOGI PLL FOR

MINIMUM SETTLING TIME AND LOW UNIT VECTOR DISTORTION UNDER

DISTORTED CONDITIONS

196

TUGCE DEMIRDELEN, TAYLAN OZDEMIR, MEHMET TUMAY

SLOW LIGHT IN PHOTONIC CRYSTAL WAVEGUIDE 197

MAACHE MOHSSIN, HOCINI ABDESSELAM, KHEDROUCHE DJAMEL

MOVE YOUR WHEELCHAIR WITH YOUR EYES 198

GOKCEN CETINEL, ZAFER TIRYAKI, ENES KUZU, MELTEM MIILIGUNEY, SEVDA

GUL

ELECTROMAGNETIC DESIGN OF LINE START AXIAL FLUX PERMANENT

MAGNET SYNCHRONOUS MOTORS 199

MUSTAFA EKER, MEHMET AKAR

xxiii International Conference on Advanced Technology & Sciences (ICAT’17)


WIRELESS POWER TRANSFER VIA AIR GAP TRANSFORMER 200

MUSTAFA OZSOY, ZAFER DOGAN

I V AND C V CHARACTERISTICS OF P SI NZNO NRS PHOTODIODES

DEPOSITED BY MW CBD 201

YASEMIN CAGLAR, KAMURAN GORGUN, MUJDAT CAGLAR, SALIHA ILICAN

EFFECT OF MN DOPANT ON THE STRUCTURAL AND MORPHOLOGICAL

PROPERTIES OF ZNO FILMS BY SPIN COATING 202

SALIHA ILICAN, YASEMIN CAGLAR, MUJDAT CAGLAR

COMPARATIVE STUDY ON THE STRUCTURAL MICROSTRUCTURAL AND

OPTICAL CHARACTERISTICS OF DIFFERENT DOPED ZNO NANOPOWDERS

BY MW HTS

203

MUJDAT CAGLAR, KAMURAN GORGUN, SEVAL AKSOY, SALIHA ILICAN, YASEMIN

CAGLAR

MEASUREMENT OF ENERGY CONSUMPTION OF WSN FOR INDOOR

ENVIRONMENT 204

SEYMA AYMAZ, GOKCE HACIOGLU

MULTILAYER PLASMONIC ABSORBER BASED METAMATERIAL FOR

REFRACTIVE INDEX SENSING AND SURFACE ENHANCED SPECTROSCOPY

APPLICATIONS

205

ERDEM ASLAN, EKIN ASLAN, OMER GALIP SARACOGLU, MUSTAFA TURKMEN

REVIEW OF REDUCTION AND SIMPLIFICATION METHODS FOR

TRANSMISSION NETWORKS 206

ALI AJDER, KADIR DOGANSAHIN, SELIM AY

DESIGN AND APPLICATION OF PRINTED CIRCUIT TRANSFER MACHINE 207

MEHMET RAHMI CANAL, MEVLUT AHMET HAS, FECIR DURAN, MERAL OZARSLAN

YATAK

DETERMINATION AND ANALYSIS OF ELECTRIC FIELD STRENGTH

LEVELS IN SAMSUN CITY CENTER 208

CETIN KURNAZ, BEGUM KORUNUR ENGIZ

ILLUMINANT DIRECTION ESTIMATION METHODS 209

ABDULLAH TAHA ARSLAN, EROL SEKE

DESIGNING A 2 D PAINTING ROBOT 210

HUSEYIN DEDE, HUSEYIN EMRE GUNER, KADIR ERGUNER, BEKIR

KARATAS, OMER AKBAS, H METIN ERTUNC

A NEW MODEL FOR MEASURING DIMENSIONAL ACCURACY BRIDGING

AND OVERHANG PERFORMANCE OF FDM 3D PRINTERS 211

SERDAR KUCUK, NURETTIN GUNES

EVALUATING THE PERFORMANCE OF DELTA TYPE 3D PRINTER

CONSIDERING THE INTERNATIONAL TOLERANCE VALUE 212

SERDAR KUCUK, NURETTIN GUNES

AN EASY TO USE JOYSTICK CONTROLLER DESIGN FOR INDUSTRIAL

ROBOTS 213

SINAN OZCAN, MUSTAFA COBAN, GOKHAN GELEN

xxiv International Conference on Advanced Technology & Sciences (ICAT’17)


ROBOT COLLISION AVOIDANCE CONTROL AND MANAGEMENT METHOD

FOR INDUSTRIAL ROBOTS IN AUTOMOTIVE INDUSTRY

214

CENGIZ DENIZ, MUSTAFA CAKIR

INFLUENCES OF ROTOR SLOT DISTANCE ON THE PERFORMANCE OF

SINGLE PHASE INDUCTION MOTOR 215

AHMET FENERCIOGLU, MERVE SEN KURT

MULTI OBJECTIVE H2 HINFINITY CONTROL OF BLADE VIBRATIONS

USING PIEZOELECTRIC ACTUATOR 216

FEVZI CAKMAK BOLAT, SELIM SIVRIOGLU

DETERMINATION OF WEIGHT COEFFICIENTS USING FUZZY LOGIC IN

WEIGHTED 3D DISCRETE WAVELET TRANSFORM 217

GUZIN OZMEN, BURAK YILMAZ, SERAL OZSEN

DUAL BAND PERFECT ABSORBER FOR BIO SENSING APPLICATIONS WITH

CARDINAL POINT STAR SHAPED NANOPARTICLES 218

AYTAC ONUR, MUSTAFA TURKMEN

DESIGN AND DEVELOPMENT OF A BIOTELEMETRY SYSTEM FOR

MONITORING BEDRIDDEN PATIENTS 219

ILKAY TASLI, ILKER ALI OZKAN

WEB BASED FUZZY EXPERT SYSTEM DESIGN FOR DETECTION OF

RHEUMATOID ARTHRITIS DISEASE SEVERITY 220

FATIH TARAKCI, ILKER ALI OZKAN, SEMA YILMAZ

SEMI AUTOMATIC DETECTION OF STOMACH CANCER BY REGIONAL

SEGMENTATION METHODS 221

ALI YASAR, ISMAIL SARITAS

MEASUREMENT OF THE OPTICAL PENETRATION DEPTH IN CHICKEN

BREAST TISSUE 222

YASAR BARIS DOLUGAN, HALIL ARSLAN, MUSTAFA ZAHID YILDIZ, AYLA EREN

OZDEMIR, ALI FURKAN KAMANLI, AYSE NUR AY

THE EVALUATION OF TOTAL OXIDANT AND ANTIOXIDANT LEVELS IN

INACTIVE HEPATITIS B PATIENTS WITH DELTA POSITIVE AND NEGATIVE 223

ILHAN SABANCILAR, TUGBA GUR

MACROMOLECULAR INVESTIGATION OF INTERMOLECULAR

INTERACTIONS DRIVING THE SELF ASSEMBLY OF COLLAGEN BY DLVO

THEORY

224

ESMA ERYILMAZ

ELECTRONIC HEALTH REMOTE EHR CONTROL OF PATIENTS BY

MEDICAL STUDENTS WITH MOBILE AND SENSOR TECHNOLOGY 225

PINAR KIRCI

AUTOMATIC DETECTION OF EPILEPSY USING EEG ENERGY AND

FREQUENCY BANDS 226

SAMEH BELLGDI, SAMER ARAFAT

SERVO MOTOR CONTROL USING PLC 227

SERHAT AKSUNGUR, TARKAN KOCA

1 International Conference on Advanced Technology & Sciences (ICAT’17)


COMPARATIVE HDL IMPLEMENTATIONS OF TWO CHAOTIC

GENERATORS

ZEYNEP KAYA1, EROL SEKE2

1 Bilecik Seyh Edebali University, Turkey; 2 Eskisehir Osmangazi University, Turkey

ABSTRACT

Chaos theory based pseudo-random sequence generators find use in secure communication

systems and cryptography. In this paper, comparative FPGA (Field Programmable Gate

Array) implementations of Lorenz and Sprott 94 G chaotic generators are presented.

Designs in HDL (Hardware Description Language) employ 8+40 bits signed fixed point

arithmetic due to the required precision for long sequences. Matlab and ISIM simulations

and comparisons are provided showing that both chaotic generators can be effectively used

in real-time cryptography and secure communication applications.

KEYWORDS - Chaotic generators, FPGA implementation, Sprott 94 G chaotic system,

Lorenz chaotic system



THE INFERENCE PROCESS OF NONSTATIONARY FUZZY SETS WITH

UNCERTAINTY FOR CYBER PHYSICAL SYSTEMS

HASAN YETIS1, MEHMET KARAKOSE1

1 Firat University, Turkey

ABSTRACT

Because of their reliable, scalable and flexible structure, the applications of Cyber-

Physical Systems (CPS) are increasing day by day. CPS is a system consisting of cyber

and physical layers, and communication layer between these layers. CPS which have a

wide application area from industrial to medical, are generally used in large-scale and

critical systems. Considering the systems they use, it is very important that the system

operates correctly and consistently. Uncertainties are undesirable situations that pose a

threat to the consistency of the system. To ensure consistency in CPS, the uncertainties

that may arise in these systems are a matter should be addressed. Because CPS has more

than one component and subcomponents, it has more uncertainty source than classical

systems. As is known, fuzzy logic systems consist of clusters used to model uncertainties.

Fuzzy logic systems proposed in type-1, type-2, or different types to model high-order

uncertainties. Nonstationary fuzzy systems are also a fuzzy inference system used to

reduce the computational complexity of type-2 fuzzy logic systems. In this study, a new

approach is proposed to increase the effectiveness especially in CBS by adding random

uncertainty to nonstationary fuzzy systems. The proposed approach first adds random

uncertainties modeled by equations constructed by taking into account the characteristics

of nonstationary fuzzy sets in CPS. Then the nonstationary fuzzy system structure is

realized according to this uncertainty effect which is algorithmically combined. The results

obtained in various scenarios for the approach verified by simulation studies demonstrate

the effectiveness, sensitivity and performance of the proposed approach compared to the

type-1, type-2 and nonstationary fuzzy systems.

KEYWORDS - cyber-pyhsical systems, uncertainty, nonstationary fuzzy, fuzzy systems



A COMPARISON OF SHORT TERM LOAD FORECASTING METHODS FOR

TURKISH DISTRIBUTION GRID

UGURCAN SENCAN1, GURKAN SOYKAN1, NAFIZ ARICA1

1 Bahcesehir University, Turkey

ABSTRACT

Short Term Load Forecasting (STLF) is a crucial issue for energy market participants, in

various reasons including energy generation planning, networks’ safeness and generation

capacity reserve management. Especially, in the liberalized energy markets, STLF is

mandatory for distribution network utilities. If the forecast is underestimated or

overestimated, it results in financial penalties in the energy markets. All system’s

compensation must have been done according to the STLF results given by demand side to

day-ahead market. If the imbalance occurs in the system due to large forecasting errors,

the responsible side has to meet the imbalance penalty costs. The margin of error on the

hourly demand forecasting should be kept low to minimize the imbalance costs of the

system. In this study, we compare the most popular STLF methods, namely exponential

smoothing, multi-layer perceptron (MLP) and recurrent neural network (RNN), on Turkish

distribution grid in Istanbul’s hourly electricity consumption data of 2015. The forecasting

models use historical load, temperature and calendar days as input. While thirty-nine

weeks’ hourly data is selected as training data, thirteen weeks’ hourly data is used for

analyzing the performance of the methods.

KEYWORDS - Short-term load forecasting, Recurrent Neural Networks, Exponential

Smoothing, Multi-Layer Perceptron



NON LOCALLY COLOR IMAGE SEGMENTATION FOR REMOTE SENSING

IMAGES IN DIFFERENT COLOR SPACES BY USING DATA CLUSTERING

METHODS

MURAT KARAKOYUN1, ALI SAGLAM1, NURDAN AKHAN BAYKAN1, ADEM

ALPASLAN ALTUN1

1 Selcuk University, Turkey

ABSTRACT

Image segmentation which is generally one of the first step of algorithms for image

processing is an important process that renders the data more meaning full for users. It is a

crucial task for image processing applications. The solution of this task bases on the

principles of the data-clustering problem. Data clustering is to separate the data according

to the similarities or dissimilarities between the data objects. There are many popular data-

clustering algorithms, such as K-means, Fuzzy C-means, and K-medoids, that rely on the

global properties of the data and are iterative, i.e. the result of the clustering algorithms

converge the best score gradually. The features of the data which will be used for data-

clustering process is also important as much as the algorithm that is to be used. In color

image segmentation, the color values of the pixels of the image are widely used as data

features. The color values vary by different color space systems. We segmented some

remote sensing images, which are taken by airborne sensors getting from ISPRS WG III/4

2D Semantic Labeling dataset, by using different data-clustering algorithms (K-means,

Fuzzy C-means and K-medoids) and different color space systems (RGB, HSV and

L*a*b*). In this paper, we show the accuracy rates of the algorithms for the color space

systems.

KEYWORDS - clustering, image segmentation, k-means, fuzzy c-means, k-medoids, color

spaces



IMPROVED RAIL SURFACE DETECTION AND CONDITION MONITORING

APPROACH WITH FPGA IN RAILWAYS

ORHAN YAMAN1, MEHMET KARAKOSE1, ERHAN AKIN1


ABSTRACT

Railway transport is the most important means of transport widely used in developed

countries. In recent years, electric trains have been produced in the direction of developing

technologies. Electric trains are comfortable, fast, reliable and environment friendly

compared to other transport vehicles. It requires continuous checking of the rail line for

the safety of railway transport. Failures in the railway can lead to larger failures if not

diagnosed early. Therefore, it is necessary to continuously check the components

constituting the railway line. In this study, an FPGA based method is proposed for rail

surface detection in railway. The proposed method is realized by image processing with

FPGA. The image is taken on the railway line with the camera attached to the FPGA

development board. Pre-processing is performed on the obtained image. Edge extraction is

applied to the image after pre-processing. The rail surface is detected using the image

obtained as a result of edge extraction. The proposed method works in real time to monitor

and diagnose faults. It detects many defects on the track surface. In this study, FPGA

based condition monitoring device was developed. An architecture has been developed for

implementing the proposed method with FPGA. This work using FPGA technology is low

cost and fast compared to other methods. The proposed method is quite advantageous

because of its real-time operation.

KEYWORDS - Railway, Rail surface detection, Condition monitoring, Embeded system,

Image processing



SENTIMENT ANALYSIS ON TWITTER SOCIAL MEDIA DATA

ISMAIL ISERI1, GOKHAN KAYHAN1, SALIH CAN OZEL1

1 Ondokuz Mayis University, Turkey

ABSTRACT

Due to the widespread use of social media tools over the last few years, massive data has

been generated around the world. According to the Domo’s report, in 2011, 100,000 twits

per minute were thrown, and in 2014, 277,000 twits were thrown per minute on twitter.

Worldwide growth in annual data volume is expected to be 59% and growth will continue

to increase. Researhers are trying to analyse and extract the meaningfull information on

this huge amount of data. There are many different studies this data by using paralel

computing , cloud computing and big data solutions. In this study, sentiment analysis is

performed on data obtained from twitter social media tool by using Hadoop. The aim of

the study is to distinguish or classify the tweets taken with a particular keyword as positive

or negative. To achieve this goal, twitter data was simultaneously obtained by using

Hadoop which is one of the wide used big data tool. After obtaining data from twitter, the

map and reduce algorithms was performed for analysing and calssification of the data. As

a result of the study seventy percent of accuracy value was obtained in the classification of

twits.

KEYWORDS - Social Media Data, Twitter, Sentiment Analysis, Hadoop, Big Data



AMBIENT MEASUREMENT PLATFORM

NUR IREM EDINC1, MEHMET CAN ALPHAN1, LUTFI ARDA1


ABSTRACT

With new technologies introduced in our lives, the interaction between machines and

people is increasing. Smart homes, smart transportation, smart appliances, smart grid

systems are part of one of the latest technology branches: Internet of Things, which can be

referred as IoT. Internet of things is a combined system of sensors, actuators and objects

which are connected via services, platforms and networks. Through this communication

structure, IoT allows people to receive or send data, control over mechanical systems,

which aids us to have the latest innovations in not only in household and cities, but also in

healthcare, agriculture, energy management and automation. This paper presents a design

of a IoT system of sensors connected via a web server which allows to be connected

locally or remotely from a Raspberry Pi. The aim of the system is monitoring temperature,

humidity, light intensity and gas measurements in an environment to a website. Also, the

printed data is stored in a database, which allows the user to reach to the past sensor data.

In this design, Raspberry Pi operates as a host of the web server and the database. Sensors

are connected to an Arduino to collect and process the data. Processed data is sent to the

Raspberry Pi via Wi-Fi to provide mobility and possibility of increasing the amount of the

sensor circuits to be used in different ambients, then stored and printed through website.

KEYWORDS - IoT system, Ambient sensing, Smart building



IMAGE ENCRYPTION BASED ON HAAR WAVELET TRANSFORMATION

AND RC4 ALGORITHM

ELHAM YASIN1, RIDVAN SARACOGLU1, PARISA LARIBI1

1 Yuzuncu Yil University, Turkey

ABSTRACT

Security and privacy of information or data at the show time have turned into a critical

concern. Progressed techniques for secure transmission and recovery of advanced pictures

are progressively required for various military, restorative, country security, and different

applications. Different sorts of strategies for increment security information or data as of

now is created, one basic path is by cryptographic systems. Keep up the security of the

message by changing information or data into an alternate frame, so the message can't be

perceived. This study aims to expect to upgrade security image arrange by joining the

Haar wavelet transformation and RC4 encryption algorithm.Haar wavelet transformation

is used to compress the image for simplicity and high-speed performance. RC4 encryption

is utilized to encrypt the image for image security and transform image safely.

KEYWORDS - Cryptography, Haar Transform, Image compression, Image encryption,

Information security, RC4 stream cipher



DYNAMICAL STRESS FIELD PROBLEM IN A PRE STRESSED PLATE STRIP

MADE BY COMPOUNDING TWO LAYERS SIDE BY SIDE

AHMET DASDEMIR1

1 Kastamonu University, Turkey

ABSTRACT

Within the scope of the piecewise homogeneous body model with utilizing of the three

dimensional linearized theory of elastic waves in initially stressed bodies (TLTEWISB)

the forced vibration of a bi-layered pre-stressed plate-strip under the action of a time-

harmonic force resting on a rigid foundation is investigated. The considered plate-strip is

constituted by compounding two discrete plate-strips side-byside. It is assumed that there

is complete contact interaction at the interface plane between the layers. To investigate the

concrete examples, the materials such as Aluminum and Steel are selected. The

mathematical modeling of the problem under consideration is introduced, and the

governing system of the partial differential equations of motion is numerically solved by

employing Finite Element Method (FEM). In particular, the numerical results

exemplifying the influence of a change in the aspect length of each layer on the dynamic

response of the plate-strip are investigated.

KEYWORDS - Bi-layered plate-strip, finite element method, initial stress, time-harmonic

force, forced vibration



DEVELOPING INTELLIGENT SYSTEMS WITH CAPACITIVE TOUCH

SENSOR USING WIRELESS SENSOR NETWORKS

MURAT DENER1

1 Gazi University, Turkey

ABSTRACT

As technology develops, the usage areas of Wireless Sensor Networks are also increasing.

Wireless Sensor Networks technology can be used in most intelligent systems because it is

more suitable than wired communication in nature conditions, it is reliable, it has flexible

structure, it provides low cost solution, it can be self-organizing, it can use energy in a

balanced way within the network and it is easy to install. In this study, driver software for

the MPR121 Capacitive Touch Sensor is written and integrated into the WiSeN sensor

node. This sensor is used in many applications. These applications include; General

Purpose Capacitance Detection, Switch Replacements, Touch Pads, Touch Wheel, Touch

Slide Bar, Capacity Near Proximity Detection. Thanks to this study, the values read from

the MPR121 sensor can be transmitted wirelessly and monitored.

KEYWORDS - Wireless Sensor Networks, Capacitive Touch Sensor, Intelligent Systems,

Sensor Node, Gateway Node



ESTIMATION OF SHORT PERIOD WIND POWER WITH ANN METHOD

BULENT YANIKTEPE1, SAKIR TASDEMIR2, A BURAK GUHER1

1 Osmaniye Korkut Ata University, Turkey; 2 Selcuk University, Turkey

ABSTRACT

As an alternative to carbon-based energy sources, there has been a widespread return to

clean and long-lived resources, called renewable energy, in the world over the past decade.

This can be attributed to the adverse effects of carbon-based resources on the environment

and the existence of certain life-times. Wind energy, which has an important place in

renewable energy resources, stands out in this scope and wind potential studies are

realized. The more stable and higher the wind power in a region, the higher the energy

potential of the investment. For this reason, the most important condition for investing in a

region and on the spot is to determine the wind potential of that region. Due to the

meteorological and geographic parameters affecting the wind power, determining the wind

potential is a difficult and time consuming process. One of the common methods used to

calculate this potential today is artificial neural network method. In this study, estimation

of wind power was investigated by using ANN (Artificial Neural Network) method with

short-period hourly data obtained from meteorological station at Osmaniye Korkut Ata

University. The effect of the Artificial Neural Networks method was successfully

demonstrated by comparing the estimated results of the tested place with the actual

measured values as a statistic.

KEYWORDS - ANN, Renewable energy, Wind Power



FUZZY POSITION CONTROL APPROACH FOR AN AUTONOMOUS ROBOT

CONTROLLER

ALI CETINKAYA1, NOVRUZ ALLAHVERDI1, AHMED YASSER

AFAGHANI1, TUGBA SARAY1

1 KTO Karatay University, Turkey

ABSTRACT

In this study, the application of position control on a fixed track by an autonomous robot

fuzzy controller approach is realized. For this purpose, a system is designed in which the

real time data exchange can be done both for hardware and software. The embedded

system (Atmel 2560 Integrity) is used in the hardware section to achieve the target

attainment and position control with the fuzzy controller. DC motor, servo motor,

ultrasonic sensor, Bluetooth module and optical sensor are controlled using this system. In

addition, the desktop software is designed using the microcontroller software as well as

Visual Studio 2015 platform using c# language. The coordinates (x, y) and path

information sent by Altu robot (The name of the robot designed for this study) on the

designed track in real time are processed with the fuzzy controller. The obtained data is

displayed with the desktop software, position information, and robot motion map of Altu

robot. Applications using the fuzzy approach are also tested with the conventional control

method and the results are compared. Thus, the fuzzy logic approach has proved that it can

provide more precise control than the conventional approach.

KEYWORDS - Position Control, Autonomous robot, Fuzzy logic



COST EFFECTIVE ELASTIC MANAGEABLE AND SCALABLE STREAM

DATA ANALYTICS ON DATA DRIVEN ARCHITECTURES

UGUR ARPACI1


ABSTRACT

Data analysis plays a vitally important for businesses to understand their functionality and

market’s requirements. The data mentioned here is user generated, therefore it shows their

behaviour, requirements, demands from the first hand. Almost on every business or

practical cases this is the common case. However analyzing and implementing such

infrastructure comes with challenges and investments in terms of finance and time. Since

this is a multi-disciplinary business area, the analysis of data could be applied by multiple

different teams with different professionalism and educational backgrounds. On the other

hand, the common practical part of these disciplines is the demand of flexible, cost

effective and scalable tools and services so that their analysis can run without any glitch.

This study states a set of applicable and battle tested infrastructures about big data analysis

by always holding the following concepts in mind: cost effectivity, flexibility and being

generic. Architectures mentioned here will be ‘serverless’ and data driven which does

require the least operational effort to execute a big data pipeline. Future work will be

automating and deploying the architecture in a couple of minutes in a scalable

configuration.

KEYWORDS - Big Data, Stream Analytics, Serverless Architectures, Data Driven, Cloud

Computing



GIS BASED WHEAT QUALITY MAP KONYA PROVINCE APPLICATION

ABDULLAH OKTAY DUNDAR1, MAHMUT TEKIN1 , KENAN PEKER2 , MEHMET

AKIF SAHMAN1

1 Selcuk University, Turkey; 2 Firat University, Turkey

ABSTRACT

Wheat production costs in Turkey are high because wheat growing enterprises are small in

area and the production is generally not carried out at barren land. The increase in wheat

production cost is deflected to wheat-based production enterprises and then to consumers.

Having an important position, especially in world flour exports, Turkey’s sustainable

competitive advantage is of great importance for the country’s economy. Enterprises in the

wheat sector are having difficulty accessing the quality wheat they want: they need a

wheat-quality map. The preparation of a wheat quality map in Konya, the grain warehouse

for Turkey, will help suppliers with supply chain arrangements, warehouse location

selection, and transportation costs. Costs will be saved by managing resources properly. In

this study, the Konya province Bread Wheat Quality Map was established for use in

management of the wheat supply chain. Accordingly, bread wheat samples were collected

from 3100 farmers in a range of production capacities in Konya province. Sample analysis

was conducted using the NIR device by Ova Wheat Factory, Inc., which is among the top

500 industrial establishments in Turkey. The results of this analysis were processed in

MapInfo Professional Release Build 16 Version 11.5 program; this created theKonya

province Bread Wheat Quality Map. The bread wheat quality map will assist efficient

management of the wheat supply chain, resulting in quality improvement and cost

reductions. This study is a part of the research which is founded by the Ministry of

Science, Industry and Technology as SANTEZ project No: 0509.STZ.2013-2 for a Ph.D.

thesis.

KEYWORDS - bread wheat, quality map, GIS, wheat supply cahin management



COMPARISON HYBRID METHODS IN FEED FORMULATIONS

MEHMET AKIF SAHMAN1, ADEM ALPASLAN ALTUN1, ABDULLAH OKTAY

DUNDAR1


ABSTRACT

Feed costs represent the largest expense for companies engaged in animal husbandry.

Animal feed-producing enterprises face problems that include raw material costs, quality-

standard product production, and sustainability. Overcoming these problems will

contribute greatly to both feed-producing and feed-consuming enterprises. At present,

companies that produce feed must establish factories in different locations to be close to

raw material resources or livestock firms. The common and effective use of these raw

materials in the factories is important. This study proposes a mixed integer linear

programming model to provide data to help solve these problems for the feed producers.

The proposed model will be analyzed the by Hybrid Linear Binary PSO (H-LBP) and H-

LBD (Hybrid Linear Binary DSA) hybrid approaches, providing decision makers with

alternative solutions that can support sustainability. By comparing the results generated by

the proposed approaches, the differences between them will be revealed.

KEYWORDS - Feed formulations, Hybrid methods



THE MOBILE APP TO ASSIST WITH TATTOO ENCODED BABY

IMMUNIZATION RECORDS IN AFRICA

BONGKEUM JEONG1, JESUS IBANEZ2, NUNO NUNES3, MONCHU CHEN4

1 Madeira Iti University of Madeira, Portugal; 2 Madeira Iti University of Madeira,

Spain; 3 Institution Superior Tecnico, Portugal; 4 Hewlett Packard, United States

ABSTRACT

Abstract— This paper describes the design and implementation of a mobile application

that assists with the interpretation and generation of tattoos which encode immunization

records. That explains the application of ethnography design, in a broad way, to solve the

issues of the vaccination system for children in Africa. We have derived the “tattoo of

Africa” concept from a design that reflects African cultural backgrounds and have linked it

with the health service design process. Tattoos, which have been common in Africa for

thousands of years, have been redefined as a new ritual for child health records. Our

design is applied to two different artefacts: a baby guide book and a baby bracelet. We

also describe our first prototype of the baby bracelet and its related App. The bracelet,

which contains an NFC tag with information about the baby’s vaccinations, can be read

from or written to any NFC-enabled Android smartphone via the visual interface of the

related App.

KEYWORDS - Ethnography Design Methods; Cultural Implications; Tattooing as a New

Ritual; Child Health Record; Mobile App with NFCEthnography Design Methods;

Cultural Implications; Tattooing as a New Ritual; Child Health Record; Mobile App with

NFC



A NEW APPROACH BASED ON IMAGE PROCESSING FOR MEASURING

COMPRESSIVE STRENGTH OF STRUCTURES

MEHMET BAYGIN1, SUAT GOKHAN OZKAYA1, MUHAMMET ALPEREN

OZDEMIR2, ILKER KAZAZ3

1 Ardahan University, Turkey; 2 Igdir University, Turkey;

3 Erzurum Technical University, Turkey

ABSTRACT

The compressive strength factor in civil engineering is a very important parameter used to

determine the performance of structures. The stability of structures can be tested with this

parameter which is used to measure the performance of concrete under different loads.

This parameter, which should be determined for the safety of the structures, is usually

based on experimental analyses performed in the laboratory environment. In this study, a

new approach to compressive strength measurement in civil engineering is proposed. With

this approach, which is based on image processing, measurement of compressive strength

parameter of concrete samples taken from structures is performed. For this purpose,

images of concrete specimens with different strengths are taken and these images are

divided into two groups as training and test set. Then, image processing algorithms are

applied to these images and the compressive strength of concrete specimens is calculated.

It has been determined that the approach suggested in the test runs performed with an error

rate of about 1-2%.

KEYWORDS - Image processing, Compressive Strength, Concrete



A REVIEW OF SMART HOME SYSTEMS

MOHAMMED QAIMAZ ALI1, ISMAIL SARITAS1


ABSTRACT

The general state of intelligent home systems, which are now becoming very public-

specific for residential buildings, is that they are simply designed for comfort. This is not

enough, with the general state being correct that they are designed as comfort. Smart home

systems can have comfort and rescue features in situations where living spaces can

threaten human life and property with comfort. In this article, it is important to note that

the dialogue of intelligent home systems can be put in place for the control and

management of possible emergency situations and the information required to design the

infrastructure of the houses / buildings accordingly, and to identify the deficiencies during

the projecting phase and to share them with project managers, operators and supervisory

colleagues, Withdrawal is intentional.

KEYWORDS - Smart, Home System, Smart Home System



A MULTI PURPOSE MOBILE APPLICATION FOR EMBEDDED SYSTEMS

FECIR DURAN1, HATICE TIRASOGLU1


ABSTRACT

Nowadays, Embedded Systems are used in many areas to perform many specific tasks.

These systems usually communicate with mobile devices. Also they have wireless control

and monitoring function. Each embedded system designer develops own algorithm and

mobile software according to their needs, causing unnecessary time loss. In this paper, the

functional mobile application has been presented for monitoring and control of the

embedded systems via Bluetooth. In addition an accessibility protocol developed for

communication to all kinds of embedded the system. This application provides real-time

instant value monitoring, graphics based monitoring and control functions to the

developers.

KEYWORDS - Mobile Application, Android, Bluetooth, Remote Control, Embedded

Systems



OBJECT DETECTION AND SEGMENTATION ON APPLE TREE IMAGES BY

USING MORPHOLOGICAL OPERATIONS AND VERTICAL AND

HORIZONTAL HISTOGRAM VALUES

SUMEYYA ILKIN1, MERVE TAFRALI1, SUHAP SAHIN1, BURAK INNER1

1 Kocaeli University, Turkey

ABSTRACT

In this paper, an algorithm for detection of apples in apple tree images is proposed. We

describe an approach that allowing the user to count apples on tree images interactively

using an application. In our algorithm, analysis of apples in tree images is made by using

basic morphological operations and vertical and horizontal histogram values. Our

algorithm begins with take apple tree images from a digital camera. And then in this

algorithm, basic morphological operations are used in apple detection process for counting

apples on the trees. In this way it is possible to allowing the user to obtain a general idea

about number of apples which he has in his apple trees. By using this information the user

can manage to estimate to earnings of his goods. The proposed study compares the output

of the algorithm with the ground truth and measures the differences according to

application results. We apply this methodology in order to assess how well it can detect

apples in apple tree images. The experimental results show that the proposed algorithm is

quite efficient for gaining the information of number of apples.

KEYWORDS - Vertical and horizontal histogram, morphological operations, pattern

recognition, object detection, image processing.



HOSPITAL PERFORMANCE EVALUATION A FUZZY DATA ENVELOPMENT

ANALYSIS APPROACH

ELIF DOGU1, MICHELE CEDOLIN1, Y ESRA ALBAYRAK1

1 Galatasaray University, Turkey

ABSTRACT

Service performance evaluation is still a contemporary interest area that relies mutual

parties, namely providers, clients and policy makers. With the new technological and

economic developments and increasing competitive conditions in health care sector, the

patients’ expectations are rising above their needs, and thus the performance evaluation of

health care systems and institutions orientates to a more uncertain environment. In the

literature, there are several articles that evaluate clinics, physicians or hospitals,

considering criteria that can be evaluated quantitatively (number of beds, number of

physicians…), whereas the expectations of customers, that generally reflect to the models

via criteria that assessed qualitatively (empathy, reliability…) are ignored in most cases. In

this paper, throughout a literature survey, the criteria that affect both technical efficiencies

(crisp data set) and patients’ expectations (fuzzy data set) are determined in order to

evaluate hospital performances of Istanbul. Alternatives are assessed by decision makers

and a non-parametric decision making tool that incorporates fuzziness into the problem is

employed to reach a ranking among the alternatives.

KEYWORDS - Service Perfomance, Hospital Evaluation, MCDM, DEA, Fuzzy Decision

Making, Quality Assessment



EVALUATION OF CUSTOMER SATISFACTION FACTORS IN HOUSING

INDUSTRY AND SELECTION OF THE BEST PROJECT USING AHP AND

TOPSIS METHODS

MICHELE CEDOLIN1, ELIF DOGU1, Y ESRA ALBAYRAK1


ABSTRACT

Housing industry is one of the leading and fastest growing industries in Turkey, in recent

years. Growing sectors represent highly competitive market structures. For acquiring

competitive advantage, companies need to clearly define their target markets and

positioning strategies. In order to increase their customer satisfaction and customer

loyalty, the companies of this industry focus on Customer Relationship Management

(CRM) activities. CRM represent all the actions that a company makes in order to

determine, acquire, develop and retain loyal and profitable customers by delivering the

right product or service to the right customer, from the right channel, at the right price.

High-budget projects of housing industry have a greater risk and therefore need to fully

meet customer expectations. There exist many factors influencing new project decisions;

selecting the best project is a Multiple Criteria Decision Making (MCDM) problem. In this

study, in the first step the factors directly affecting customer satisfaction in the housing

industry are determined by taking expert opinions and conducting literature research.

Three criteria and thirteen sub-criteria are defined and evaluated; their weights are

calculated using AHP method. In the second step, these weights are used as inputs and

among three different new project alternatives; the best project is selected with project

data and expert decisions using the TOPSIS method.

KEYWORDS - MCDM, AHP, TOPSIS, Housing industry



CLASSIFICATION OF ECG SIGNALS BY THE ENSEMBLE LERARNING

METHODS

CIGDEM BAKIR1

1 Yildiz Technical University, Turkey

ABSTRACT

In this study, non-linear dimension reduction methods such as PCA, LDA were applied to

ECG signals and success of such dimension reduction techniques for the classification and

segmentation of ECG signals were discussed. As classification method, such as bagging,

boosting and random subspace, ensemble methods have been chosen. Because ensemble

learning methods are powerful and popular classifiers. Also, segmentation of data through

neighbourhood feature extraction (NFE) method were enabled by transiting from high

dimensioned space to low dimension space by considering the longitudinal combination of

ECG signals. Results classification results of NFE algorithm performed through

longitudinal combination and as a newly developed method were compared with

classification results of ECG signals obtained through dimension reduction by taking one

ECG instance. Results of NFE dimension reduction technique performed by considering

the neighbour ECG instances, advantage of effect on segmentation of ECG signals were

presented at empirical results section and the success of suggested method was indicated.

In addition, ensemble learning methods results were presented comparatively.Results

obtained by performed study are promising for the studies to be conducted in further

period.

KEYWORDS - Ensemble learning, ECG, Classification



INVESTIGATION OF E LEARNING APPLICATIONS IN CIVIL ENGINEERING

ALI ERDEM CERCEVIK1, NIHAN KAZAK1

1 Bilecik Seyh Edebali Universty, Turkey

ABSTRACT

E-learning has emerged with the increasing use of computers and internet worldwide. E-

learning is a personal interactive learning platform. Active learning can be done at home at

any time with this system. Civil engineering education is also generally based on theory.

In the recent past, e-learning have been considered as effective learning techniques for all

lessons except practice and fieldwork. However, studies and technological developments

have shown that e-learning can be used for fieldwork. In this study investigated that e-

learning for civil engineering. The applications in our country and world, advantages-

difficulties are discussed. The use of new technologies in e-learning has been researched

such as virtual reality. It has been shown that E-learning is very effective learning and

practise for future civil engineering education.

KEYWORDS - E-learning, Civil Engineering, Active Learning, Engineering Education,

Virtual Reality.



DATA MINING APPLICATIONS IN DIAGNOSIS OF SYMPTOMS

HARUN OZKISI1, MURAT TOPALOGLU1

1 Trakya University, Turkey

ABSTRACT

Taking into account the dietary habits of today, it has become very common for people to

have diabetes. When the importance of human health is considered, the correct diagnosis

of a person suffering from any illness is of great importance for the positive outcome of

treatment. In addition, the duration of treatment for a correctly diagnosed disease is

shortened and the negative situations that may arise as a result of misdiagnosis are

avoided. The application of data mining in the field of accurate diagnosis in the field of

health reveals significant results. Decision trees are obtained by analyzing the actual data

obtained from this field by means of data mining methods. Decision trees are created with

decision support systems to prevent misdiagnosis. In this study, it is aimed to apply data

mining related to the diagnosis of diabetes according to the symptoms. The data recorded

by a health facility was categorized by expert opinion before being analyzed. Then, the

data mining method was used to obtain decision trees from these data. The analysis was

performed using the WEKA package program. As a result, in this study, a decision tree is

revealed about the diagnosis of the disease with the help of diabetes findings.

KEYWORDS - Data Mining, Decision Tree, Diabetes.



USING RDF SUMMARY GRAPH FOR KEYWORD BASED SEMANTIC

SEARCHES

SERKAN AYVAZ1, MEHMET AYDAR2

1 Bahcesehir University, Turkey; 2 Kent State University, United States

ABSTRACT

The Semantic Web began to emerge as its standards and technologies developed rapidly in

the recent years. The continuing development of Semantic Web technologies has

facilitated publishing explicit semantics with data on the Web in RDF data model. This

study proposes a semantic search framework to support efficient keyword-based semantic

search on RDF data utilizing near neighbor explorations. The framework augments the

search results with the resources in close proximity by utilizing the entity type semantics.

Along with the search results, the system generates a relevance confidence score

measuring the inferred semantic relatedness of returned entities based on the degree of

similarity. Furthermore, the evaluations assessing the effectiveness of the framework and

the accuracy of the results are presented.

KEYWORDS - Semantic Web, Semantic Search, RDF, Graph Summarization, Ontology



DESIGN OF WEB BASED MONITORING SYSTEM FOR TOMCAT

APPLICATION SERVER

GOKHAN KAYHAN1, ISMAIL ISERI1, HAKAN KABASAKAL1


ABSTRACT

Servers are just as important as the useful applications for Information Technologies.

Server Management Tools are very crucial and helpful for management of application

servers. System administrators need to monitor all the features of servers by using web

applications. In this study, a monitoring tool which is web based Java Application was

developed for management of Tomcat Application Server. Java Management Extension

(JMX) must be used to read and write administrative information for the application

server. It is sufficient to enable remote access for JMX with given Java Virtual Machine

(JVM) parameters when the tomcat server starts. As a result, many meta data such as

response times and ready times of the Tomcat Application server can be monitored with

the developed tool and the applications running on the server are managed more

effectively according to the resource usage.

KEYWORDS - Application Servers, Server Monitoring, Server Management, JMX, JVM



A NEW STUDY OF K MEANS ALGORITHM IN BOSPHORUS AND

DARDANELLES STRAITS DATABASE

AYLA SAYLI1, TUGCE SERT1


ABSTRACT

Data mining information and knowledge from large datasets in order to find data analytics

and use them for better designing and modelling in many areas such as market sales,

market basket analysis, customer relationship management, medical science, finance

analysis, fraud detection. A model has been generated by clustering algorithms in order to

cluster the data according to similarity. Clustering plays a very important role in exploring

data, creating approximations and also it is practical for grouping together users with

similar behaviours’ so that mutual models can be built for them. Many clustering

algorithms have been developed and are categorized from several aspects which are

Partitioning Methods, Hierarchical Agglomerative, Density based methods, Grid based

methods, and Model based methods. In this study, K-Means clustering algorithm is

focused on. This algorithm is one of Partitioning methods. It has been coded in Windows

Forms Application on Microsoft SQL Server database and also implemented in R-Studio.

Elbow method is implemented in R-Studio and then it estimates the right number of

clusters for the algorithm. Dataset used in this study is obtained from Maritime

Transportation Data of Turkish Statistical Institute. This data is analysed in our application

and clustered according to the number of vessels and total tonnage carried by the vessels

passed through the Bosphorus (Istanbul) and the Dardanelles (Canakkale) for last decade.

Computational results shows that this algorithm is very promising and interesting.

KEYWORDS - Data Mining, Cluster Analysis, K-means, Windows Form Application, R-

Studio



DECISION TREE RESEARCH APPLICATIONS OF ID3 AND C4 5

ALGORITHMS

AYLA SAYLI1, TUGCE SENTURK1


ABSTRACT

Data mining is the process of acquiring useful, previously unknown and potentially

applicative information from the data in order to make future decisions better. It allows

users to analyse data from many categories and to summarise relationships identified.

Several methods & algorithms are available in Data Mining which are collected under four

common classes of techniques: Classification, Clustering, Association Rule Learning and

Regression. Classification which is the process of separating observations into classes is

the most widespread method for finding the mine rule from huge databases. Various

algorithms exist in classification such as; KNN, Bayes Classifier, Artificial Neural

Networks, Genetic Algorithms. The simplest one is Decision Tree method which plays a

fundamental role to classify and to predict for data analyses due to its simple structure to

understand& interpret .In other words, the decision tree method is easier way by analysing

data partly faster when it is compared to the other classification algorithms. In this project,

ID3 and C4.5 algorithms which were introduced by J. R. Quinlan have been used for “Red

Wine Quality” dataset to determine which attributes of this dataset affect the its quality

hierarchically . The objective of this paper is to present comparison of these two

algorithms that have been coded in Visual Studio Console Application using C# as

programming language for ID3 algorithm and in WEKA for C4.5 algorithm. As a

consequence, its results are obtained and they show that these algorithms can be used

effectively.

KEYWORDS - Data Mining, Classification, Decision Tree, ID3, C4.5, Visual Studio

Console Application, WEKA



MP DDOSER IP KILLER TESTING BEHAVIOR OF IP KILLER DDOS BOT IN

THE DETER TESTBED

TUAGHMOHAMMAD KALTE1, GOKHAN DALKILIC1

1 Dokuz Eylul University, Turkey

ABSTRACT

The denial of service (DoS) attack and its distributed sort, the DDoS attack, are potentially

very dangerous attacks, which can put down web and application servers throughout the

World. Several provider companies support their customers via the Internet, yet those

individuals who have malicious intentions are threatening them. The DDoS attackers use

different methods to attack their victims, one popular choice among these is the use of bot

tools. Researchers intensely try to find out solutions to DDoS attacks, so they analyze the

behavior of these sort of attacks. But they have issues for testing scenarios in the real

implementations, hence testers use simulation or emulation methods to address this

situation. This study is focused on one of the attacking tools called IP-Killer. IP-Killer (or

MP-DDoser) is solely a DDoS bot, which contains many different DDoS attack rules.

Here we test its attacking rules with an emulation environment to show the impact of the

tool. Thus we have decided to use DETER Cybersecurity Testbed for our attacking tests,

which has many special abilities for testing cyber-attacks and this might represent a real

world scenario better. This paper covers the analysis of IP-Killer and its behavior during

DDoS attacks, and measurement of the outcomes. We have taken the bandwidth, the CPU

and RAM usages into consideration. Then we showed the test results in the graphs. Lastly,

we discovered that UDP method consumes the largest bandwidth, while RAM usage and

CPU utilization are consumed most by the HTTP GET method.

KEYWORDS - DDoS, Botnet, MP-DDoser (IP-Killer), The DETER, DDos



A DEEP LEARNING APPROACH FOR OPTIMIZATION OF SYSTEMATIC

SIGNAL DETECTION IN FINANCIAL TRADING SYSTEMS WITH BIG DATA

H SERCAN KARAOGLU1, SERKAN AYVAZ2, UGUR ARPACI1

1 Foreks Bilgi Iletisim Hizmetleri A. S., Turkey; 2 Bahcesehir University, Turkey

ABSTRACT

Expert systems for trading signal detection have received considerable attention in recent

years. In financial trading systems, investors’ main concern is determining the best time to

buy or sell a stock. The trading decisions are often influenced by the emotions and feelings

of the investors. Therefore, investors and researchers have aimed to develop systematic

models to reduce the impact of emotions on trading decisions. Nevertheless, the use of

algorithmic systems face another problem called “lack of dynamism”. Due to dynamic

nature of financial markets, trading robots should quickly learn and adapt as human

traders. Recently, a solution for detecting trading signals based on a dynamic threshold

selection was proposed. In this study, we extend this approach by adopting several

different rule based systems and enhancing it by using the Recurrent Neural Network

algorithm. Recurrent Neural Networks learn the connection weights of subsystems with

arbitrary sequences of inputs that make them a great fit for time series data. Our model is

based on Piecewise Linear Representation and Recurrent Neural Network with the goal of

detecting potential excessive movements in noisy stream of time series data. We use an

exponential smoothing technique to detect abnormalities. Trading signals are produced

using fixed time interval data from Istanbul Stock Exchange. The evaluations indicated

that our model produces successful results in trading data. Future work will focus on

further improvements and scalability of the model.

KEYWORDS - Big Data, Machine Learning, Recurrent Neural Networks, Expert Systems,

Trading Systems.



A WEB AND MOBILE BASED APPLICATION MONITORING AIR POLLUTION

INSTANT PARAMETERS USING CURL IN PHP

ABDURRAHMAN OZBEYAZ1, YAVUZ DEMIRCI1

1 Adiyaman University, Turkey

ABSTRACT

Air pollution is one of the main problems for humanity. Especially, this pollution occurs

via transportation, warming, agriculture and industry. The quality of air in terms of

pollution changes according to the intensity of air pollutants in atmosphere. Monitoring air

quality is crucial for asthma sufferers and people affected by air pollution. These persons

want to monitor by their mobile phones. In Turkey, air quality monitoring stations are

established by Environment and Urban Ministry for each province. These stations have

been hourly measuring air pollutant parameters according to its places. While some of

stations only measure PM10 and SO2, another one measures PM10, SO2, NO2 and etc.

According to these parameters, air quality indexes are instantly calculated for each station

daily. These parameters have been publishing on the web page of ministry. In this study, a

new web based application publishing PM10, SO2 and AQI values belonging to

Adıyaman province instantly is developed in PHP with CURL. This application is parsing

the current web address. Furthermore; developed application has been offering service in a

web server located in Adiyaman University. Also, the mobile application has been

developed in Android. Thus, the air quality parameters can be instantly monitored from

our mobile phones. In this paper, such a system with web and mobile base is detailed

described. And this study reveals such a need including mobile based application

measuring air quality parameters according to provinces.

KEYWORDS - Air Pollution, Monitoring System, Web programming, PHP, Android



USING DATA MINING AND FUZZY RANKING METHODS FOR

DETERMINING SIGNIFICANCE LEVELS OF ASSOCIATION RULES

MELDA KOKOC1, ADNAN AKTEPE1, SULEYMAN ERSOZ1

1 Kirikkale University, Turkey

ABSTRACT

Data mining methods have become more prominent with the spread of the notion of big

data. These techniques are used for filtering and analyzing the information required in

production and service systems. Fuzzy logic, which allows for different analyses to be

integrated with data mining, facilitates the analysis of incomplete or subjective data. In

this study, data mining and fuzzy ranking methods are applied in an emergency service.

Support and confidence values of inter-departmental of emergency service relations were

found by applying Apriori algorithm, GRI algorithm, and CARMA algorithm, using in

SPSS Clementine software. Confidence values for the three algorithms are divided into

quarter intervals and four intervals are obtained in order to facilitate the calculation of the

significance level of the relations. This interval is one of three criteria considered for

ranking the significance level of relationships between emergency services departments.

Other two criteria, "the number of algorithms" and "the direction of the relationship" were

calculated according to the three algorithms in SPSS Clementine. As a result of the

analysis of these criteria, the fuzzy ranking was used to determine the significance levels.

The fuzzy ranking was carried out using Fuzzy Logic Designer module of Matlab R2009a.

In the program, firstly the input units (criteria) and the output units (groups) of the

algorithm were considered. In order for the module to be able to decide, the decision rules

created through the expert opinion used. Since the properties of relations were inputted on

the resulting interface of the program, the importance of each relation was found and

ranked. With the help of this study, data mining and fuzzy ranking were used to determine

signification levels of the association rules and it was applied in emergency service.

KEYWORDS - Fuzzy Logic, Data mining, Emergency Service, Service system,Fuzyy

Ranking



CLASSIFICATION OF TREE SPECIES IN A FOREST BY ARTIFICIAL

NEURAL NETWORK USING ASTER SATELLITE IMAGES

MUSTAFA BUBER1, MUHAMMED FAHRI UNLERSEN2, KADIR SABANCI3

1 Doganhisar Vocational School, Turkey; 2 Necmettin Erbakan University, Turkey;

3 Karamanoglu Mehmetbey University, Turkey

ABSTRACT

Remote sensing is information acquisition from objects without any physical contact. It is

used in various areas from earth sciences like hydrology, ecology, oceanography,

glaciology etc. to military applications. One of the areas where remote sensing is most

useful is the forest investigations. Forests always have areas which is very difficult to

achieve. Due to this fact, the remote sensing can be very useful to use the classification of

tree types in a forest. In this study, photographs taken by ASTER satellite imaging have

been used to determine tree types to create a forest map. The ASTER satellite imagery has

contains images that taken in various wave length. The dataset is taken from a forested

area in Ibaraki Prefecture, Japan. The most dominant tree species in this forest are

Cryptomeria japonica (Sugi) and Chamaecyparis obtusa (Hinoki). In this study, the

estimation of tree species has been done by using many types of Artificial Neural

Networks (ANN). Number of neurons in the hidden layer of the ANN has been changed

from 1 to 20 to determine the best ANN structure which estimate the tree species. The

effects of change of activation function of the layers on the results have been investigated.

Hyperbolic tangent sigmoid (Tansig), Logarithmic Sigmoid (LogSig) and Linear (PureLin)

transfer functions have been used as activation function. The prepared ANN structures

have been trained various training functions like Resilient backpropagation

(traingdx),Gradient descent with momentum and adaptive learning rate backpropagation

(traingdx), Bayesian regularization backpropagation (trainbr), Conjugate gradient

backpropagation with Fletcher-Reeves updates (traincgf), Gradient descent

backpropagation (traingd), Gradient descent with momentum backpropagation (traingdm),

Levenberg-Marquardt backpropagation (trainlm), Scaled conjugate gradient

backpropagation (trainscg). All of the results of the researches have been compared and

the best results have been obtained as 0. 501 RMSE and 0.225 MAE. The structure that

has the best results, has consisted of 11 neurons in the hidden layer with PureLin

activation function. And this structure has been trained by Conjugate gradient

backpropagation with Fletcher-Reeves updates function.

KEYWORDS - Classification of Tree,Artificial Neural Network,ASTER Satellite Images



DIAGNOSIS OF MESOTHELIOMA DISEASE USING DIFFERENT

CLASSIFICATION TECHNIQUES

KEMAL TUTUNCU1, OZCAN CATALTAS1


ABSTRACT

Mesothelioma, which is a disease of the pleura and peritoneum, is an asbestos-related

environmental disease in undeveloped countries although developed countries in the west

have an asbestos-related illness. In spite of the incidence of this disease is lower than that

of lung cancer, the reaction it creates in society is very high. In this study, 10 different

classification algorithms were applied to the Mesethelioma data set obtained from real

patients in Dicle University, Faculty of Medicine and loaded into UCI Machine Learning

Repository, and the results were compared. When the obtained results were compared, It

has been seen that Artificial Neural Network correctly classifies 321 examples of 324 and

catches %99.0740 success among other classification techniques.

KEYWORDS - Mesothelioma Disease, Classification Techniques, Data Mining.



A SIMPLE APPROACH FOR INFLUENCE MAXIMIZATION IN SOCIAL

NETWORKS

AYBIKE SIMSEK1, RESUL KARA1

1 Duzce University, Turkey

ABSTRACT

People are social entities and need to communicate with each other. They are affected by

the other people or affect them. People can influence each other on a variety of issues,

from buying an item from a supermarket to voting in an election. A reflection of this social

environment in real life emerges as a social networking platform, such as Twitter, Flicker,

etc. Today, the real life social environments and positions of people have been moved to

social platforms. Individuals often communicate on social platforms, engage in new

interactions, and organize their social environment. The influence of some individuals may

be greater than others due to their position, the institutions or charities they represent.

These people are called as influencers. Detection of the influencers offers many benefits.

For example, advertising a product by the influencers can increase product sales. Such

influencers can be used to mold public opinion on a particular subject, too. In addition,

detection of people who affect social networks can be used to create security

vulnerabilities, as spreading a computer virus. There are many studies in the literature

about the detection of influencer users. The main purpose of these studies is to identify the

top-k influencers who create maximum influence within a particular social environment.

In this study, we developed a simple approach for the influence maximization by detecting

top-k influencers in a Erdös-Renyi synthetic graph. The graph is directed and weightless, it

has 150 nodes and 1500 edges. It has been tried to detect influencers using only

neighborhood relations. The top-k influencers are detected successfully with the proposed

method.

KEYWORDS - Social Networks, Influence Maximization, Graph Theory



DESIGN AND IMPLEMENTATION OF AN ARTIFICIAL NEURAL NETWORK

SOFTWARE TO DIAGNOSE DISEASE

ISMAIL SARITAS1, KADIR TOK1


ABSTRACT

In this study, method of artificial neural networks is used in order to support medical

diagnostics. A software, which has a visual interface, has been designed with C#.NET

programming language. With this software, success analysis of the results can be made

and the performance of the results can be measured. This software was designed to be able

to be used in academic studies and in applications which are related to artificial neural

networks. In the studies, the datas which are received from the patients with rheumatoid

arthritis and from the people who are not suffering from rheumatoid arthritis are used as

training and test data sets. With the software, training takes place according to the desired

number of neurons, learning rate, momentum coefficient and number of iterations which

are determined dinamically. It was found that the network, on average, reaches the zero

error rate, in the range of 500 and 1500 iterations. In the iterations which were performed

1500 times and more in computers with different processors, 82% to 100% success rate

has been obtained. In this study, the importance of artificial neural networks and the

advantages of the software, that can learn, in computer technologies have been examined.

Also, backpropagation algorithm was examined in Matlab environment for the diagnosis

of rheumatoid arthritis. The results found with the backpropagation algorithm and the

perceptron algorithm have been compared in terms of performance. While %82 accuracy

percentage is obtained with the Backpropagation algorithm, the accuracy percentage is

found as %71 with Perceptron algorithm.

KEYWORDS - Artificial neural networks, disease diagnosis, artificial intelligence



INFORMATION SECURITY A HYBRID ALGORITHM STUDY

EGEMEN TEKKANAT1, MURAT TOPALOGLU1

1 Trakya Universitesi, Turkey

ABSTRACT

Information represents an important place since human existence. Technology is

progressing every day because of the age of information and technology that we live in.

Secure storage of information has become very important. Especially with the information

thefts that have emerged in recent years, the secure storage and transmission of

information has become very important. Encryption, as well as information hiding

technologies, are also an important part of storage information more safely. In the time

from past to now, cryptography is widely used. Many encryption methods have been

developed since the time of human existence to now. Moreover today, widely used public

key cryptography systems and steganography methods can be used together to increase

information security. Steganography is the science is the hiding of a secret message with

in an ordinary message. In this study, we have developed a new hybrid algorithm with

steganography technologies together with existing encryption algorithms.

KEYWORDS - İnformation Security, Cryptology, Steganography, Theft



THE FUNCTION ANALYSIS AND ITS APPLICATION TO DETERMINATION

OF FUZZY SUBSETS FOR MODEL HELICOPTER FUZZY CONTROL

MURAT HACIMURTAZAOGLU1, NOVRUZ ALLAHVERDI2

1 RTE University, Turkey; 2 KTO Karatay University, Turkey

ABSTRACT

In this study, an analysis of the function that determines the sub-clusters of the fuzzy

controller for model helicopter control is performed. For this purpose, a model helicopter

and a test platform were established in which the results of the function analysis can be

seen and the application results will be examined. Arduino control card is used for model

control. Analysis results are analyzed on computer using c # program. The computer

connection with the model is carried out via the Usb port. Input parameters for control are

obtained from the sensors on the model. The output values are calculated as net values to

control the motors on the model. Real-time application results are seen as simulations on

the model.

KEYWORDS - Fuzzy subset function, fuzzy control, fuzzy logic control system, model

helicopter control



FRIENDSHIP ALGORITHM FOR DETECTING BLACK HOLE ATTACKS IN

MANET FADBAM

ADWAN YASIN1, MAHMOUD ABU ZANT1

1 Arab American University, Palestinian Territory

ABSTRACT

Security of Mobile Ad-hoc Network (MANET) is one of the hottest research topics in the

network fields. MANET was originally used in military, and it is consists of mobile nodes

connected via wireless links (radio, microwaves signals) with no fixed infrastructure to

control the communication between nodes, Black hole attack is one of the most wide

spread attacks that used to affect the connectivity and integrity of MANET. A black hole

node (Malicious node) drops all incoming packets from the source node instead of

transmitting them to destination node. In this paper we propose a new technique to fight

against this type of attacks by revealing and isolating the black hole nodes from the

network.

KEYWORDS - MANET; AODV; Black-hole Attacks; Security; Connectivity; Secure ad

hoc routings protocols



HUMAN LIMB MOVEMENT ESTIMATION BY ARTIFICIAL NEURAL

NETWORK USING BRAIN WAVE SIGNALS

MURAT KOKLU1, KADIR SABANCI2, MUHAMMED FAHRI UNLERSEN3

1 Selcuk University, Turkey; 2 Karamanoglu Mehmetbey University, Turkey;

3 Necmettin Erbakan University, Turkey

ABSTRACT

In our daily life, our limbs are indispensable. But many people lost some limbs due to

accidents or illnesses. To make life of people like that easy, some robotic solutions could

be applied. Electroencephalography (EEG) may be used to control that kind of robotic

limbs. In our study, EEG signals have been used to determine status of a human limb. The

limb whose status is asked to be determined, is eyelid in this study. The dataset used to

determine eyelid status were recorded brain waves from 14 electrodes by using a

commercial EEG headset. The status of eyelid was added into the dataset manually. The

dataset has been analyzed by using many types of Artificial Neural Networks (ANN).

Number of neurons in the hidden layer of the ANN has been changed from 1 to 20 to

determine best ANN structure which estimate the eyelid status. At the same time the

activation function of neuron layers have been investigated for the same purpose.

Hyperbolic tangent sigmoid (Tansig), Logarithmic Sigmoid (LogSig) and Linear (PureLin)

transfer functions have been used as activation function. The prepared ANN structures

have been trained various training functions like quasi-Newton backpropagation (trainbfg),

Bayesian regularization backpropagation (trainbr), Conjugate gradient backpropagation

with Fletcher-Reeves updates (traincgf), Gradient descent backpropagation (traingd),

Gradient descent with adaptive learning rate backpropagation (traingda), Gradient descent

with momentum backpropagation (traingdm), Levenberg-Marquardt backpropagation

(trainlm), Scaled conjugate gradient backpropagation (trainscg). All of the results of the

researches have been compared and the best results have been obtained as 0.262 RMSE

and 0.328 MAE. The structure that has the best results, has consisted of 11 neurons in the

hidden layer with PureLin activation function. And this structure has been trained by

Bayesian regularization backpropagation function.

KEYWORDS - Electroencephalography, Artificial Neural Networks



MALICIOUS WEB PAGE DETECTION WITH MACHINE LEARNING

TECHNIQUES

CANSU KADI1, URAZ YAVANOGLU1


ABSTRACT

Web technologies which is changing and increasing day by day become a target by

attackers who take advantage of web users growing attention. Users can do not only

important things require interaction such as shopping, banking, reservation, paying bills

but also get information. All these procedures they can share sensitive/critical informations

about them. Even if it only entered into this web page it is possible to access these

informations. To ensure the security of the web page is needed to prevent these attacks.

For this purpose it is used different diagnosis tools and approaches. A dataset has been

created with malicious and benign web pages. It contains lexical and popularity features of

URL, HTML features and JavaScript features. Features have been selected with Fisher’s

Discriminant Ratio (FDR), Discrete Cosine Transform (DCT) and Principal Component

Analysis methods (PCA). And then, k-Nearest Neigbor (kNN), Support Vector Machines

(SVM) machine learning methods have been used for classification. Experimental

evaluation demonstrates that the accuracy of detection system is %97,14 and false positive

rate is 0,02 with using FDR and SVM.

KEYWORDS - Malicious and benign web pages, static and dynamic approaches, URL

lexical features, HTML features, JavaScript features, machine learning, support vector

machines, k-nearest neighbor



IMPLEMENTATION OF 16 BIT ARITHMETIC LOGIC UNIT ON FPGA

ILKER ALI OZKAN1, ABDULKADIR SADAY1


ABSTRACT

Electronic devices have a Central Processing Unit (CPU) which controls the operation of

the device and carries out the operations to be carried out respectively. A CPU has basic

operations such as arithmetic functions, logical operations. All of this is done based on the

control input selection on the Arithmetic Logic Unit (ALU), which is one of the most

important parts of the CPU. In this work, we present a 16-bit ALU application using the

VHDL language and the Quartus II 13.0 tool. This ALU performs arithmetic operations

such as Addition, Subtraction, Increment, Reduction, Transpose as well as logical

operations such as AND, OR, XOR, NOT. All of these operations are subject to

verification to see if they match theoretically.

KEYWORDS - 16 bit ALU, FPGA, VHDL, ALU Design



A SUGGESTION PERFORMING AUTOMATIC BLOCK REALLOCATION

USING PARTICLE SWARM OPTIMIZATION IN LAND CONSOLIDATION

PROJECTS

ABDURRAHMAN OZBEYAZ1, YASAR INCEYOL1


ABSTRACT

Block reallocation has been a challenge problem in Land consolidation projects today.

Block reallocation is manually done by specialists in current works. If a robust algorithm

which is capable automatic reallocation is designed using in land consolidation projects,

both fair reallocation is provided and this challenge work will be easier. Many algorithms

have been proposed which can perform block reallocation in the land consolidation

projects in literature. Although the successes of the proposed algorithms in literature have

been substantially good, the achievements of some algorithms have been constantly

investigated. In this study, Automatic block reallocation using Particle Swarm

Optimization (PSO) algorithm which is an optimization algorithm in land consolidation

projects has been investigated. The steps of the algorithm for the task are developed for

PSO. An arable land in Dot village in Adıyaman has been employed as a project study.

Landholdings preferences and places of fixed installations are taken into consideration in

the design of the steps of algorithm. The land includes 143 blocks and 225 landholders.

405 fixed areas weren’t used in the project; only 293 parcels are subjected to reallocation

in the study. Consequently, this paper presents a model algorithm which can automatically

reallocate parcels to blocks using Particle Swarm Optimization.

KEYWORDS - Automatic Block Reallocation;,Land Consolidation, Lanholding, PSO,

Optimization



A TOOL FOR COLLECTING DATA FROM THE DEVICES USING SNMP

AGENTS

ATIKE DEMIRBAS1, YALCIN CEBI1

1 Dokuz Eylul University, Turkey

ABSTRACT

This paper introduces an overview for network management system in order to collect and

analyze data from active devices on network and it gives a summary about network

management function model. According to the function model, a special tool is developed

using Java and MySQL with SNMP for monitoring system. One of the purposes of this

tool is; tracking the active device on network, discovering new device which is added to

system, controlling the devices and etc. The other purpose of this tool is to develop an

open source and simple tool because the other open-source software are either very

complex or very simple. The system is based on manager device, agent devices and

Network Management System. The manager device control and monitor agent devices is a

device, the agent device is managed by the manager. Network Management System is

software installed on the manager device. All communication between manager and agents

are made possible by NMS.

KEYWORDS - SNMP, Java, Network Management, Network Monitoring System,

Performance Management



FEATURE SELECTION FROM 3D BRAIN MODEL FOR SOME DEMENTIA

SUBTYPES USING GENETIC ALGORITHM

SAVAS OKYAY1 , NIHAT ADAR1 , KEMAL OZKAN1 , BAKI ADAPINAR1

1 Eskisehir Osmangazi University, Turkey

ABSTRACT

Brain scans that are appropriate to the medical standards are obtained from magnetic

resonance imaging devices. Through image processing techniques, 3D brain models can

be constructed by mapping medical brain imaging files structurally. Physical

characteristics of patient brains can be extracted from those 3D brain models.

Characteristics of some specific brain regions are more efficacious in predicting the type

of the disease. For that reason, researches are made for finding the worthwhile features out

using cortical volumes, gray volumes, surface areas, and thickness averages for left and

right brain parts separately or together. The main objective of this work is determining

more influential sections throughout the entire brain in establishing the clinical diagnosis.

To that end, among all the measurements exported from 3D models, the significant brain

features that are effective in identifying some dementia subtypes are sought. The dataset

has 3D brain models generated from magnetic resonance scans of 63 samples. Each

sample is labeled with one of the following three disease types: Alzheimer’s disease (19),

frontotemporal dementia (19), and vascular dementia (25). The genetic algorithm based

wrapper feature selection method with various classifiers is proposed to select the features

that state the aforementioned dementia subtypes best. The tests are performed by applying

cross validation technique and confusion matrices are shown. At the end, the best features

are listed, and the accuracy results up to %95.2 are achieved.

KEYWORDS - Genetic algorithm, Feature selection, Dementia subtypes, 3D brain model,

Magnetic resonance imaging.



INTELLIGENT BACKUP SYSTEM BASED ON ARM MICROCONTROLLER

FATIH SADAY1 , ABDULKADIR SADAY2

1 KTO Karatay University, Turkey ; 2 Selcuk University, Turkey

ABSTRACT

The intelligent backup system, which has been designed considering the problems

experienced by enterprises and individuals in recent years, will protect backups against

encryption viruses unlike existing systems. The systems and products used for back-up

purposes in the market contain some deficiencies in different directions. For example,

many systems have major problems with data loss such as not having the ability to clean

up old backups, creating security vulnerabilities due to continuous access vulnerability,

forgotten backups, and so on. Cost-effective solutions that include all the features are

needed to get rid of these problems by the users. Customers want to store their data in a

physically accessible back-up system with minimal workload and high reliability. Systems

for needing backup are only available as foreign products on the market and do not have

all the features together. The intelligent backup system to be developed with the embedded

system is aimed at enabling enterprises to fully respond to the need for backup.

KEYWORDS - backup system,arm,adaptive scan,intelligent backup



A CHANCE PLAY DESIGN WITH C EXAMPLE OF HORSE RACE

APPLICATION

SAKIR TASDEMIR1 , EMRE AVUCLU1


ABSTRACT

Digital games are leisure-time computer software with certain rules. In leisure time, people

are involved in activities to relax. The existence of chance thinking and people's beliefs,

thoughts and experiences are at the basis of luck games. Chance games appeal to many

people because they show a stereotyped behavioral characteristic adopted by a significant

segment of society. Based on spending leisure time, people have been searching computer-

aided to contribute to their different behaviors. In this study, real time a horse racing game

was developed in Visual Studio C # environment. The C # language was chosen because it

was designed as a simple, modern, general-purpose, object-oriented programming

language. With visually programmed play, the user can play several options. These

options have made the game more attractive and made it more enjoyable. In the game,

some adjustments can be made user-sided (weight, lucky horse etc.). Depending on the

settings made by the user, the excitement situation is further increased during the race. At

the end of the game, according to the race performance the user or the computer can win

the sum of the points laid out.

KEYWORDS - Digital games, Chance games, Game programming, Horse racing.



A NOVEL OPTIMIZATION APPROACH BASED ON RECIPROCAL FORCES

FOR ASTRONOMICAL N BODY SIMULATION ON GPU

CELIL OZKURT1 , EYUP GEDIKLI1

1 Karadeniz Technical University, Turkey

ABSTRACT

Astronomical N-body simulations are widely used to investigate the formation and

evolution of various astronomical systems, such as planetary systems, global clusters,

galaxies, galaxy clusters, and large-scale constructions of the universe. In this systems

each body continuously interacts with every other body via gravity forces. The all-pairs

method is a brute-force technique that calculates all pairwise interaction among the N

bodies. Although it is a simple method, it is not used in large-scale simulations since the

computational complexity is O (N2). However, it can be used for close range calculations

when combined with hierarchical methods that are used in long distance calculations.

Reciprocal forces fij = -fji can be used to reduce the number of force calculation in the all-

pairs method by half. However, until March 2015, it was difficult to implement this

optimization effectively with CUDA. With the release of the concurrent streams feature in

CUDA 7, this optimization has become effective. In this paper, the classical CUDA

application with the reciprocal forces optimized CUDA application was compared in terms

of memory usage and calculation time. It has observed that reciprocal forces optimization

uses about 42% less memory than classical CUDA implementation and about 19% faster

when the number of body is 73,728. In addition, as the number of body increases, the

reciprocal forces optimized CUDA application runs even faster.

KEYWORDS - Astronomical N-body; Reciprocal forces; Parallel Programming;

Optimization with CUDA; GPU Programming



MULTISPECTRAL IMAGE FUSION USING A NEW TENSOR APPROACH

SAHIN ISIK1 , KEMAL OZKAN1 , GOLARA GHORBAN DORDINEJAD1


ABSTRACT

In this study, we have introduced a new multispectral image fusion methodology with

Common Matrix Approach based tensor decomposition. The performance of three well-

known tensor decomposition approaches including Two Dimensional Principal

Component Analysis, Two Dimensional Singular Value Decomposition and Common

Matrix Approach are compared by conducting an experiment on a special Multispectral

Image Database to examine the effectiveness of each image fusion method. The statistical

and visual results are ground on in case of performance evaluation. The obtained results

obviously indicate that using Common Matrix Approach algorithm for image fusion gives

satisfactory numerical results and good appearance as well as preserving meaningful

information stated on each channels.

KEYWORDS - common matrix approach; multispectral image fusion; Two Dimensional

Component Principal Analysis; Two Dimensional Singular Value Decomposition.



DETECTION OF FOREIGN OBJECTS IN RAILWAY LEVEL CROSSINGS

USING IMAGE PROCESSING TECHNIQUES

CANAN TASTIMUR1 , MEHMET KARAKOSE1 , ERHAN AKIN1


ABSTRACT

Railroad transportation is considered one of the safest means of transportation. Train

accidents happened every year in the world due to heavy task. And the train accidents

resulted in serious destruction of property and injury or death of passengers and crew

members. Many of the railway transport accidents happen because of driver’s tiredness,

bad weather conditions, and defective rail components, etc. To prevent these accidents,

importance is attached to the detection of faulty regions in the tracks, and other rail

components. Safety of railroad transportation can be enhanced by utilizing intelligent

systems that provide additional information about the exact location of the train, its speed

and upcoming obstacles. A collision avoidance system for railway vehicles is required to

specify their location in the railroad network precisely and reliably. Level crossings in

railroad are important safety regions. The level crossings are common areas used by trains

and motor vehicles. Therefore there is a risk of collision of trains and motor vehicles at

these points. Foreign objects in the level crossings constitute a risk of accident for both

trains and motor vehicles. Moreover, the deterioration of the shape and surface structure of

the level crossings threatens the safety of railway access. Faulty level crossings should be

detected with non-contact image processing techniques. In this study, the level crossings

have been detected by using Pre-processing such as color format conversion, image

normalization, Feature extraction, Hough transform, Elimination of noisy pixels, and some

image processing techniques. It has been determined whether or not the foreign objects are

in the detected level crossings. It has been decided whether the detected foreign objects

were of a threat to the security of the transportation. The foreign object detected for this

process has been examined by using morphological feature extraction techniques. So that

the train driver can be alerted beforehand about the foreign object that has been checked.

KEYWORDS - Rail transportations; Level crossings detection; Hough transform; Feature

extraction; Foreign object detection



PREDICTION OF ION EXCHANGE EFFICIENCY OF COMERCIAL RESINS

STRONG BASE ION EXCHANGE RESINS WITH ARTIFICIAL NEURAL

NETWORKS

SERPIL EDEBALI1 , ABDULLAH ERDAL TUMER2 , MEHMET ALKANER2

1 Selcuk University, Turkey; 2 Necmettin Erbakan University, Turkey

ABSTRACT

In this study, the ion-exchange efficiency of Lewatit®M500 and Lewatit®MP500 strong

base anion exchange resins was predicted with artificial neural networks by means of

experimental results obtained during laboratory studies. Artificial neural networks have

been frequently used in many fields for various targets such as prediction, modeling,

classification and data association. Similarly, they have been applied successfully in

environmental engineering studies (modeling of sorption processes, prediction of ozone

concentrations and water losses, etc.). One of the most interesting fields related with

environmental engineering is ion exchange process. In this treatment process which is

based on exchange of anion or cation on the surface of unsoluble solid material with

similar charged ions, ion exchange resins are used. In this study, experiments were carried

out where Lewatit®M500 and Lewatit®MP500 ion exchange resins were used for the

removal of Cr(VI) from aqueous solutions. The data (time, resin amount, pH value,

concentration, sorption percentage) obtained from the experiments were firstly trained by

artificial neural networks and then they were tested for the prediction of ion exchange

efficiency of ion exchange resins. As a result of this study, it was determined that

Lewatit®MP500 strong base anion exchange resin had better removal efficiency.

KEYWORDS - Artificial neural networks, Modeling, Ion Exchange, Base Anion exchange

resin, Non-linear Modeling, Feed forward back propagation Neural Networks,Comersial

resins



TRAINING OF ANFIS NETWORK BY GENETIC ALGORITHM FOR

DIAGNOSIS OF LEUKEMIA CANCER SUBTYPES USING GENE EXPRESSION

PROFILE

MUSTAFA TURAN ARSLAN1 , BULENT HAZNEDAR2 , ADEM KALINLI3

1 Mustafa Kemal University, Turkey; 2 Hasan Kalyoncu University, Turkey;

3 Erciyes University, Turkey

ABSTRACT

In this study, subtypes of Leukemia cancer has classified by using microarray gene

expression profiles. An approach is proposed to train Adaptive Neuro Fuzzy Inference

System (ANFIS) network by using a population-based Genetic Algorithm (GA) to classify

this cancer data. The classification success of the proposed model has compared with the

successes of Backpropagation (BP)-ANFIS and Hybrid-ANFIS, which are derivative

based ANFIS models. According to obtained results, GA-ANFIS model has performed

very well on leukemia cancer, with 90.91% in the classification study. For the same data,

BP-ANFIS and Hybrid-ANFIS models have performed poorly with 63.63% and 59.09%,

respectively.

KEYWORDS - Microarray Gene Expression Profile, ANFIS, Genetic Algorithm,

Classification



AN INTELLIGENT CLASSIFICATION APPROACH FOR SOCIAL MEDIA

DATA

FATIH ERTAM1


ABSTRACT

Social media has a great place in today's internet usage. In particular, it is important for

companies that want to advertise on social media to invest in which area. Facebook has the

highest usage rate among social media tools. In this study, artificial intelligence techniques

were used to classify a data set generated with Facebook data. In order to achieve this

purpose, support vector machines and extreme learning machines are compared. The

results of the studies were evaluated with accuracy values and ROC curves. In addition,

the processing speeds of the classifiers were also checked and compared. In this study, an

intelligent classifier that could be preferred for social media data was tried to be offered.

Different activation functions for both classifiers were selected and compared among

themselves.

KEYWORDS - social media classification, machine learning, support vector machine,

extreme learning machine, ROC curves



DESIGN AND IMPLEMENTATION OF AN INTERPRETER FOR THE LEAST

SQUARES METHOD USING SYMBOLIC APPROACHES

NAWAL ABDULLAHI MOHAMED1, HUSEYIN PEHLIVAN1


ABSTRACT

The least squares method is an important technique of regression analysis that fits a

mathematical or statistical model to a particular data set. This work describes the

development of an interpreter for the least squares method, using symbolic computation

methods and automatic code generation tools. The development process starts with the

construction of a context-free grammar that denotes the mathematical curves. Then, a

parser which is generated via the JavaCC tool for this grammar is employed to represent

the curves with object structures and to determine their parameters. Through these object

structures, the curves are analyzed and the parameters to be computed by the least squares

method are determined. For the curves with specific function components, such as

exponential, logarithmic and rational functions, some symbolic computation tasks are

performed, which transform those curves into polynomials.

KEYWORDS - Symbolic computation, Curve fitting, Least squares method, Context-free

grammars



THE ROLE OF INNOVATIVE TECHNOLOGY-INTEGRATED TEACHING:

TOWARDS A TRANSFORMATIVE PEDAGOGY

NAZ ONEL1

1 Stockton University, United States

ABSTRACT

Today’s educators are using technology in many different ways both inside and outside the

classroom. In order to develop effective and engaging learning environment, it is crucial to

integrate technology using innovative ideas, which can create active, stimulating,

collaborative, and informative experiences for students. Existing research shows that to

help improve undergraduate education, it is necessary to use active learning techniques,

encourage contacts between students and faculty, and develop reciprocity and cooperation

among students, which all could be possible with the adoption of technology in innovative

ways. This study covers a number of unique instructional methods that are currently being

implemented as a distinct pedagogical approach. The study provides ideas and tools for

online and in-class teaching with a number of concrete examples specifically geared

towards educators in variety of disciplines and at all levels. Approaches and some good

practices that have been used at a U.S.-based higher education institution (HIU) are also

described in detail.

KEYWORDS - in-class teaching,learning,pedagogical approach,online teaching

,technology in classroom



TRAINING OF ARTIFICIAL NEURAL NETWORK USING METAHEURISTIC

ALGORITHM

SHAIMAA SAFAA AHMED ALWAISI1 , OMER KAAN BAYKAN2

1 Science, Iraq; 2 Selcuk University, Turkey

ABSTRACT

This article clarify enhancing classification accuracy of Artificial Neural Network (ANN)

by using metaheuristic optimization algorithm. Classification accuracy of ANN depends

on the well-designed ANN model. Well-designed ANN model Based on the structure,

activation function that are utilized for ANN nodes, and the training algorithm which are

used to detect the correct weight for each node. In our paper we are focused on improving

the set of synaptic weights by using shuffled frog metaheuristic optimization algorithm

which are determine the correct weight for each node in ANN model. We used ten well

known datasets from UCI machine learning repository. In order to investigate the

performance of ANN model we used datasets with different properties. These datasets

have categorical, numerical and mixed properties. Then we compared the classification

accuracy of proposed method with the classification accuracy of back propagation training

algorithm. The results showed that the proposed algorithm performed better performance

in the most used datasets.

KEYWORDS - Artificial Neural Network, Metaheuristic Optimization Algorithm, Back

propagation algorithm, Shuffled Frog Optimization Algorithm



CUSTOMER CHURN ANALYSIS USING CLASSIFICATION ALGORITHMS IN

TELECOMMUNICATIONS INDUSTRY

MUHAMMET SINAN BASARSLAN1, FATIH KAYAALP1


ABSTRACT

One of the customer-focused sectors is the Telecommunication sector. Telecommunication

companies want to win customers and lose their customers. At this point, they work to

predict the loss of customers in a variety of ways. In this study, a study was carried out

with classification algorithms to estimate the allocation of customers on data received

from a telecommunication company in Turkey.

KEYWORDS - Keywords: Data mining Machine Learning.Churn, R Tool,

Telecommunication



IMPLEMENTATION AND EVALUATION OF FACE RECOGNITION BASED

IDENTIFICATION SYSTEM

FARUK CAN ELBIZIM1, MUSTAFA CEM KASAPBASI1

1 Istanbul Commerce University, Turkey

ABSTRACT

Face recognition has been widely used and implemented to many systems for the purpose

of authentication, identification, finding faces, etc. In this study Yale face database [1] is

used which consist of 15 different people. For each of person there are 11 different images

with different face expressions. In this study images are categorized as normal, normal and

center light, normal and happy, normal with left light and right light. In order to recognize

these faces 4 different face recognition methods namely Eigenface, Fisherface, LBPHface

and SURF are utilized in the developed environment. In order to test the mentioned face

recognition algorithms a software is developed using EmguCV in .NET environment.

After evaluating and comparing the obtained confusion matrix amongst other the

LBPHface method was found to be superior method with an average accuracy of 99%, it

was ~98% SURF, ~97% for EigenFace and FisherFace. FicherFace was slightly better

then the Eigenface method

KEYWORDS - Face recognition,eigenFaces,fisherfaces,SURF



REDUCING CBR DROPPING IN QOSBOX USING GREENCLOUD

SIMULATION IN CLOUD COMPUTING

SHAFIULLAH MALEKZAI1, DOGAN YILDIZ2 , SERAP KARAGOL2

1 77 Construction Company Kabul Afghanistan, Afghanistan; 2 Ondokuz Mayis University,

Turkey

ABSTRACT

Cloud computing is the latest and greatest thing of our era. It was introduced when local

resources were incapable any more to fullfil the access requirements along with the

increase of network bandwidth and variety of services breaks constraints of physical

placement and deliver diversities of services in its virtualized hardware and software type.

GreenCloud simulator as an extension of the Network Simulator (NS-2) cloud computing

simulators, execute an energy-aware cloud computing scenario with data center virtualized

components (switches, routers, links, etc.). This paper converses a cloud architecture that

delivers virtualized network as per demand from cloud user, with the explanation of QoS-

based GreenCloud simulation that to demonstrate the performance and behavior of the

nework performance in cloud computing environment. The QoSbox is used as PC-routers

over a network. QoSbox is expected as a desired solution for the gradual extension to the

problem of providing service differentiation in a scalable manner.

KEYWORDS - Cloud Computing; GreenCloud; NS-2; QoS; QoSbox



BEACON BASED INDOOR POSITIONING AND ROUTING

TANER ARSAN1

1 Kadir Has University, Turkey

ABSTRACT

This research paper is about beacon-based indoor positioning and routing system and its

Android application to guide the visitors for University Campus Area. Bluetooth Low

Energy (BLE) beacons are located in strategically designed points. The system has been

implemented on approximately 9,000m2 indoor area with 150 beacons. The application

software has been developed with Android Studio and Special Android Indoor Software

Development Kit (SDK). Android SDK uses path loss model and trilateration for

localization. Trilateration method is used to compute the distance of the traveller from

each of three beacons. The position of the visitor is determined as the point of intersection

of three circles, each centred at one of the beacons. The radius of each of these circles is

the estimated distance of the visitor from that beacon. The system allows users to navigate

in the campus with a 3D map while providing them directions to the daily events that take

place. Also an augmented reality has been created using camera view of the mobile device

to further help with the indoor navigation. On the other hand, reliable indoor localization is

a must for the development of the Internet of Thing (IoT). Sensors that are a part of the

IoT can benefit from accurate indoor location determination. That is why we have been

studying on indoor positioning systems for 3 years and during this research period of time,

we obtain significantly better results in average error point of view. For this purpose,

Fingerprint algorithm is developed and applied to beacon-based indoor positioning system

and the results are discussed in this research paper.

KEYWORDS - Indoor positioning, indoor routing, trilateration, mobile application

development, software development



DETERMINING OPTIMAL NUMBER OF CLUSTERS IN LEACH LIKE

PROTOCOLS

SINAN TOKLU1, MEHMET SIMSEK1


ABSTRACT

Clustering of sensor nodes to reduce energy consumption in wireless sensor networks

(WSN) is a very common approach. The selected number of clusters directly affects the

performance of the network. There are some suggestions for the number of clusters in the

literature. To reduce energy consumption, determining the number of clusters according to

the network parameters would be correct. In this study, we examined the effect of some

network parameters on the selection of cluster numbers.

KEYWORDS - wireless sensor networks, clustering, energy efficiency



BIG DATA AND RESTFUL BASED WEB APPLICATION FOR SMART HEALTH

APPLICATION

SINEM GUVEN SANTUR1, YUNUS SANTUR2, MEHMET KARAKOSE2

1 Papatyasoft, Turkey; 2 Firat University, Turkey

ABSTRACT

Smart city applications are generally the whole of concepts that benefit from technology in

terms of management and society. They provide benefits along with the effect of increase

in new services or in the quality of existing service in terms of society with the purposes of

efficient use of resources in terms of management, obtaining valuable information through

data mining and reducing costs. Smart cities have components such as smart

transportation, smart park, smart buildings, smart park and smart health. In this study, a

Big Data and Restful web based software proposal was presented for the purpose of

storing, processing and analyzing the data obtained from information sources for a smart

health management in a smart city.

KEYWORDS - big data, pregnancy, data mining, machine learnin



DEEP LEARNING BASED ARTIFICIAL MANAGER FOR SMART CITY

YUNUS SANTUR1 , EBRU KARAKOSE1 , MEHMET KARAKOSE1 , ERHAN AKIN1


ABSTRACT

Nowadays, deep learning is commonly used in many areas such as natural language

processing, data mining, image processing and interpretation. The use of technology in

city management for the purposes of effective resource management, improving the

quality of service and reducing costs have led to smart city concept. The data produced by

automation systems as well as internet-connected objects such as sensor, camera and

mobile device are also used for smart city management. It is difficult to analyze such a big

sized data by processing with conventional methods and to use them in decision-making

mechanisms. In this study, deep learning based data mining was performed on big data

obtained from different types of sources for smart city management and an approach to

ensure that the results can be analyzed was proposed.

KEYWORDS - Deep Learning, Deep Mining, Internet of Things, Smart City, Big Data



RE ANALYSIS OF MOLD MAKING ERRORS WITH COMPUTER AIDED

DESIGN PROGRAMS AND RECOMMENDATIONS

MUSTAFA ALTIN1, AZIZ TOLUNAY ARSLANBAS1, SAKIR TASDEMIR1


ABSTRACT

Computer-Aided Design (CAD) programs are widely used in construction technologies.

CAD programs are constantly being developed and with these software most accurate and

clear solutions can be achieved. By creating concrete static projects, alternative solutions

can be seen, most accurate and economical decisions can be taken. Better time

management is also provided in this way. No doubt, one of the most important productions

in the construction business is mold making. Mold making masters should have good

training and should be good at their jobs. A small mistake can lead to irreversible

consequences and can cause to various troubles. Bad manufacturing and workmanship will

cause accidents in the construction phase and after that will cause problems in calculations

done regarding safety. The construction industry comes first in work accidents and an

important part of the accidents happen because of molding and scaffolding errors. The cost

of these accidents to the employers and the state is at serious levels. In this study, a

construction made with values gained from a project prepared with boundary values is

examined. In this roughly finished construction, incomplete productions and missing

section heights have been determined due to mold workmanship. It has been observed that

there are major faults such as column installation errors and columns not overlapping.

Errors in beam column joints have been eliminated by resolution of project, i.e. CAD

program. Due to great advantages provided by the CAD program, a solution has been

reached shortly. It has been decided to stop the construction of the existing building and

carry out a mandatory strengthening project. CAD programs have been used in the

strengthening project and the most suitable project has been prepared by working on

alternatives to minimize the cost of strengthening. It is important that the CAD programs

provide instant alternatives at the solution stage and comply with the comparison

conditions. The economic benefit provided by different alternatives for reducing the

strengthening cost is significant.

KEYWORDS - Computer-Aided Design, Strengthening, Molding errors, Building defect,

Cost analysis



A COMPUTER ALGEBRA SYSTEM THAT CAN BE INTEGRATED INTO

USERS WORKING ENVIRONMENTS TO SUPPORT SYMBOLIC

COMPUTATION OPERATIONS

SEDA EFENDIOGLU1, H EMRE EFENDIOGLU1, HUSEYIN PEHLIVAN1


ABSTRACT

Algebra is one of the important tools for facilitating calculations used in all branches of

science in which mathematics is involved. Numerous computer programs have been

developed that can quickly perform algebraic calculations without errors. However, since

these programs are designed to be used only with their own user interfaces, they can not be

integrated into other computing environments or run as a component of those

environments. In this study, the design and development stages of a system that can be

integrated into users' working environments to support algebraic computational operations

are presented. Formal language definitions have been used to represent an algebraic

expression. First, grammar rules are defined for the forms of algebraic expressions, and

abstract syntax trees of related expressions are generated by applying language parsing

operations. These trees are converted to expressions in JavaScript Object Notation (JSON)

data interchange format so that they can be used by all programming languages. Then

simplification operations are applied to obtain object representations of these expressions.

Finally, various numerical and symbolic computation operations are defined on these

object representations to support the simplification of algebraic expressions and the

generation of similar expressions.

KEYWORDS - computer algebra system, automatic expression generation, simplification,

symbolic computation, JSON,formal grammars



COMPARISON THE PERFORMANCES OF DBSCAN CLUSTERING

ALGORITHM AND IMAGE PROCESSING TECHNIQUES FOR COUNTING

WHITE BLOOD CELLS

FATMA OZGE OZKOK1, METE CELIK1


ABSTRACT

White blood cells (WBCs) are an important part of immune system of human body. The

number of low or high WBCs give significant information about many diseases and so

correct determination of number of WBCs is very critical. For example number of WBCs

can increase in bacterial infections while number of WBCs can decrease in viral

infections. Traditionally, tedious and time consuming manual methods based on visual

examinations have been used. Recently, automated methods for counting WBCs have

begun to gain importance in literature. In the study, we compared the performances

DBSCAN clustering technique and image processing methods (i.e., watershed approach

and morphological operators) on counting WBCs. DBSCAN is a density based clustering

algorithm and can discover arbitrary shaped clusters without using number of cluster as

input parameter. Experimental results show that DBSCAN clustering algorithm presents as

promising results as image processing techniques of watershed and morphological

operations.

KEYWORDS - White Blood Cells, White Blood Cell Count, Clustering, DBSCAN, Data

Mining, density- based clustering



THE DIAGNOSIS OF MELANOMA SKIN CANCER USING SOME FEATURE

DESCRIPTORS

KAYA TURGUT1, CAGRI KILINC1, SEMIH ERGIN1


ABSTRACT

In this paper, a computer aided diagnosis (CAD) method to classify melanoma diseases

which are one of the most mortal types among skin cancer varieties. A tissue including

melanoma is sorted as malignant or benign cases using traditional ABCD (Asymmetry-

Border-Color-Differential Structure) rule from dermoscopic images. In the classification

of melanoma cases, a lesion segmentation step is vital since an appropriate segmentation

significantly affects the accuracy of recognition results. After the separation of a lesion

from skin background in a dermoscopic image using Otsu thresholding approach, the

feature vectors including shape, color and texture properties are extracted from the

corresponding lesion. While shape features such as asymmetry, aspect ratio, compactness,

etc. are extracted from border of a lesion, color features are computed by analyzing the

color distribution on the corresponding lesion. Finally, texture features are extracted from

the GLCM (Gray-Level Co-Occurrence Matrix) of the raw pixel matrix of a dermoscopic

image. An image database consisting of 81 benign and 81 malignant melanoma cases is

utilized in this paper and lesions. The preferred classification method is the one against

one first order Support Vector Machines (SVM) algorithm. The proposed computer aided

diagnosis system offers a recognition accuracy of 82.26% on the test image set.

KEYWORDS - pattern classification, feature extraction, support vector machines,

melanoma detection



COMPARISON OF LINEAR DIMENSIONALITY REDUCTION METHODS ON

FACE RECOGNITION

ERAY YILDIZ1, YUSUF SEVIM1

1 Karadeniz Technical Universtiy, Turkey

ABSTRACT

The analysis of high dimensional data is encountered in many areas, such as face

recognition. In the face recognition, dimensionality reduction methods are used to work

with fewer dimensions. There are a lot of dimensionality reduction methods. Because of

time and space complexities, linear dimensionality reduction methods are more preferred

generally. In this paper, popular linear dimensionality reduction methods and their

performance are investigated on face recognition. These methods are principal component

analysis (PCA), linear discriminant analysis (LDA), neighbourhood preserving embedding

(NPE) and locality sensitive discriminant analysis (LSDA).

KEYWORDS - Face Recognition, principal component analysis, linear discriminant

analysis, dimensioanlity reduction, locality sensitive discriminant analysis, neighborhood

preserving embedding



ANALYSIS OF LRC PERFORMANCE USING LBP FEATURES IN FACE

RECOGNITION

SULENUR ERGUL1, MEHMET KOC1

1 Bilecik Seyh Edebali University, Turkey

ABSTRACT

Local Binary Patterns (LBP) is a popular method used for feature extraction in pattern

recognition problems such as texture recognition and face recognition. Features are

extracted by comparing the grey level values of a center pixel and its neighbors. In face

recognition, image is divided into several regions and LBP feature distributions are

evaluated from each region separately. Then the classification is done according to

dissimilarity measures such as Chi-square. Linear Regression Classification (LRC) is a

popular holistic method used in face recognition problems. In this method, it is assumed

that the feature vectors in a class lie on a linear subspace. Then the regression coefficients

are calculated using least squares method. In LRC, intensities of image pixels are used as

features. The unknown query is assigned to the class where the minimum reconstruction

error occurs. In this work, we use LBP features to describe the face images and use them

to classify the images using LRC. In the experiments performed in AR face database,

LRC-based classification outperformed Chi-square based one.

KEYWORDS - Linear regression classification, local binary patterns, chi-square, face

recognition, feature extraction



DEVELOPMENT OF OPTICAL CHARACTER RECOGNITION MOBILE

APPLICATION OVER CARD VISITS WITH IMAGE PROCESSING METHODS

ISMAIL ISERI1 , GOKHAN KAYHAN1 , KEREM HACIFAZLIOGLU1


ABSTRACT

In this study, it is worked on transfering personel information to mobile devices by using

image processing and character recognition methods on card visits. Although the square-

coded card visits are widely used, it is difficult to use because it’s necessity to have a data

matrix reader on mobile devices. So that it is developed an mobile application by using

Open Source Computer Vision Library (OPENCV) and Open Source Optical Caracter

Recognition framework Tesseract and Android Platform. OpenCV is an open source

computer vision and machine learning software library. Tesseract is an optical character

recognition engine for various operating systems which is free software released under the

Apache License. As a result of the study, an innovative application was developed using

optical character recognition technology and obtained deep knowledge about optical

character recognition.

KEYWORDS - OpenCV, Tesseract OCR, Edge Detection, Canny, Scanner, Gray Scale,

Image Pyramids, Image Perspective



FEATURE WEIGHTING USING GENETIC ALGORITHM FOR TEXT

CLASSIFICATION

PARISA LARIBI1, RIDVAN SARACOGLU1

1 Yuzuncu Yil University, Turkey

ABSTRACT

It is important to organize, process and extract useful information of textual documents in

a convenient way. The most important thing to fulfill these aims is the text classifier. In

this study, documents were passed through pre-processing. At this point, a mathematical

model of the documents was established. In this model, the data size is soo high. The

fuzzy similarity approach is used to reduce the data size. The k-Nearest Neighbor

algorithm is used for the next step, classification. Weights have been assigned to the

attribute values of the documents in order to increase classification performance. Genetic

algorithms are used to find the best weight values.

KEYWORDS - Genetic Algorithm, K Nearest Neighborhood, Text Classification



HOW DO COLLEGE STUDENTS ADOPT SOCIAL NETWORKING SERVICES

UNDERSTANDING THE DETERMINANTS

HASAN USLU1, MUSTAFA EREN YILDIRIM1, YUCEL BATU SALMAN1


ABSTRACT

Technological developments have introduced significant changes to the way in which

individuals interact with each other. People tend to spend more time on social networking

activities than any other content on the Internet. Social media centred research has become

an increasingly significant interest which is capable of supporting several modes of

communication compared to traditional methods. This paper aims identifying and

analysing the influential factors of social networking adoption among Turkish college

students in order to explore the user behaviours in this context. An extended Technology

Acceptance Model with the integration of attitude, ease of use, usefulness, trust and social

influence is generated as the research model. Data is collected by an online survey with the

participation of 417 subjects. The results are analysed by statistical methods. The validity

and reliability of collected data is verified. Perceived usefulness, perceived ease of use,

trust, attitude, and social influence showed significant effect on the adoption of social

networking services.

KEYWORDS - social networking services, technology acceptance model, influence,

adoption



A NEW FAST MEASUREMENT AND INSPECTION APPROACH BASED ON

MACHINE VISION

MEHMET BAYGIN1 , MEHMET KARAKOSE2 , ALISAN SARIMADEN3 , ERHAN

AKIN2

1 Ardahan University, Turkey; 2 Firat University, Turkey; 3 Medel Electronic, Turkey

ABSTRACT

Machine vision based quality control and high-precision measurement systems provide

users with significant advantages such as high speed, accuracy, low installation and

maintenance costs. A machine vision system capable of non-contact, fast and high-

precision measurement has been proposed and developed in this paper. For this purpose,

two cameras were used to capture images of products passing through a conveyor. Firstly,

a reference image is taken in the system. The product to be passed over the conveyor is

placed in the middle of the two cameras and the conveyor for metric calibration. In this

way, images are taken from two cameras. After this process, image processing and

mathematical operations is applied to images and measurement values are obtained. The

proposed approach works in real time and error amount of the system is less than +-2 mm.

The most important feature of this approach which provides high-precision and non-

contact measurement is the ability to measure dimension from camera characteristics and

lens, product type and position of products.

KEYWORDS - Machine Vision, Image Processing, Measurement, Inspection



OPTIMAL ROUTE DETECT BASED ON GENETIC ALGORITHM FOR

ERZURUM METROPOLITAN MUNICIPALITY

MEHMET BAYGIN1, SALIH SERKAN KALELI1, ABDULLAH NARALAN2

1 Ardahan University, Turkey; 2 Ataturk University, Turkey

ABSTRACT

Optimization problems are constantly met in everyday life and the main goal in solving

these problems is to determine the solution which gives the best benefit. In this paper, the

road public transport routes of the Erzurum Metropolitan Municipality were examined and

tested by means of an optimization algorithm. The genetic algorithms are used as the

optimization algorithm. The data used in the approach suggested in the study are obtained

from the Erzurum Metropolitan Municipality Public Transportation Directorate and

contains the currently used routes. As a result of the optimization process in the proposed

approach, some of the currently used routes have been determined to be problematic and

accordingly, a new route has been proposed. In the proposed approach, the performance of

the algorithm has been tested and a performance increase of approximately 25% was

achieved.

KEYWORDS - Genetic Algorithm, Route Optimization, Traveler-Salesman Problem



A NEW SOFTWARE DESIGN FOR MUDEK ACCREDITATION PROCESS THE

APPLICATION OF KIRIKKALE UNIVERSITY

ATILLA ERGUZEN1, SEYMA CIHAN1 , VOLKAN ATES1 , ERTUGRUL CAM1


ABSTRACT

Association for Evaluation and Accreditation of Engineering Programs (MÜDEK) is a

non-governmental organization operating for the enhancement of the quality of

engineering education in Turkey in different disciplines. They are trying to create invisible

links and unity between the universities in the developed countries and the universities

accredited by MÜDEK in our country. This link is established with the agreements made

with the European Network for Accreditation of Engineering Education (ENAEE) and

Washington Accord Signatory (WAS) accreditation agencies. However, there are some

requirements, rules and procedures to be followed in this process. In this study, MÜDEK

preparing process conducted by Kırıkkale University Engineering Faculty were

mentioned. In addition, the requirements of MUDEK for accreditation and the software

developed by the faculty were introduced. In conclusion, it is believed that the developed

software will support effective training activities and measurement procedures during and

after the accreditation process of MÜDEK and will benefit other universities that will

apply in the future.

KEYWORDS - engineering education, MUDEK, accreditation of engineering



PREDICTION OF HEATING VALUES OF REFUSE DERIVED FUEL FROM

PROXIMATE ANALYSIS DATA

KEMAL OZKAN1, SAHIN ISIK1 , AYSUN OZKAN3 ,

ZERRIN GUNKAYA2 , MUFIDE BANAR2

1 Eskisehir Osmangazi University, Turkey; 2 Anadolu University, Turkey

ABSTRACT

Refuse-derived fuel (RDF) is a fuel produced by shredding and mixing municipal solid

waste or industrial waste. In recent years, RDF are used in waste to energy plants and

cement factories as alternative fuel. Determination of heating values of fuels needs

delicate and expensive calorimeter, and is time-consuming. It’s quite desirable to develop

quick and easy model predicting heating values of fuel. So, new empirical correlations

based on proximate analyses of RDF used for prediction of higher heating value (HHV)

are presented in this paper. For this purpose, the moisture (MC), ash (AC) and volatile

(VC) content of the RDF samples were determined according to ASTM D-3173-85,

ASTM D-3174-82 and ASTM D-3172-73, respectively. Also, the HHV of the RDF were

determined using an IKA C200 calorimeter (ASTM D-5865). The correlations have been

developed via linear regression method by using proximate analysis results. The best

correlation, which has least errors, is selected and compared with real HHV analysis

results. The correlation based on first order (HHV = -0.1332*MC + 0.009*AC- 0.009*VC

+ 7) has a total error of 0.92 %. The other correlation which is based on second order

(HHV = -0.2688*MC – 0.007575*AC + 0.4829*VC + 0.0025*MC² + 0.0033*AC² -

0.01441*VC² + 5) is indicate that it has good HHV predictive capability.

KEYWORDS - heating value modelling, linear regression; proximate analysis; RDF.



ANALYSIS OF REAL TIME APPLICATION RESULTS OF BAYESIAN AND

ARTIFICIAL NEURAL NETWORKS ANN ALGORITHMS IN PRODUCING

CREDIT SCORE CARD

ALI TUNC1, HUSEYIN KARADAG1

1 Kuveyt Turk Participation Bank, Turkey

ABSTRACT

Credit scorecard production and development work is a very important part of the

financial sector.Software products called credit scoring systems generally consist of

collecting the customer's data according to previously determined credit factors,

processing the obtained data with various statistical or machine learning techniques and

determining the final credit decision by making a credit risk analysis.These scorecards,

developed in line with the customer's information, are based on any financial action

requested.According to this information, it is decided whether the customer can allocate

necessary loan and fund utilization.In this context, it is tried to be analyzed by finance

institutions in the best way to produce the most accurate results. In addition to the old

customers, the credit status of new customers should be determined.In the absence of too

much financial activity belonging to the customer, statistical results are needed according

to the existing records when the financial and scorecard of this customer needs to be

established.In this study, a training was performed using the data obtained from a private

financial institution and Bayesian and Artificial Neural Networks(ANN) algorithms.The

information that can or cannot be used for the new customer is estimated by including

instant clients on the learned data.The proposed method is run in a number of operations,

and the performance and success rates of the method are presented in real time.The study,

which consists of variables belonging to the customer, was processed by machine learning

and statistical technique and tried to determine the credit value of the customer.Bayes and

Artificial Neural Networks methods and classification problem.According to the obtained

information, the creditability of the customers has been decided.

KEYWORDS - Credit Score Card, Real Time Machine Learning,Artificial Neural

Networks ,Bayesian,Classification



THE TRUST MODEL FOR EVALUATION OF CLOUD SERVICE PROVIDERS

ATILA BOSTAN1, DENIZ KORAY INCELER1

1 Atilim University, Turkey

ABSTRACT

Cloud services are spreading wide and becoming a global need. The most important

criteria for people who are receiving and sending data, using cloud storage for

information, transferring processes to cloud computing environment has now come to the

fore as a trust. The users mostly doubt about non-technical trust issues instead of technical

Encryption, confidentiality, availability and so on. Either selecting the services or whether

to get cloud services or not. Measuring trust is quite troublesome. Personal view comes

into prominence. It is considered that a measurement and evaluation method which is as

common as possible that aids to evaluate by sorting and categorizing the services. In

accordance with this purpose, we developed a scale that evaluates the trust in cloud

services. We gathered expert opinions. As of the moment the expert standpoints and the

scale will be compared and a fine tune will be applied.

KEYWORDS - Cloud, Trust,service-to-service trust,trust model



SEED QUALITY CONTROL BY USING ARTIFICIAL NEURAL NETWORK

KADIR SABANCI1, MUHAMMED FAHRI UNLERSEN2, MURAT KOKLU3

1 Karamanoglu Mehmetbey University, Turkey; 2 Necmettin Erbakan University, Turkey;


ABSTRACT

Due to the growing markets, it is important to present unmixed species of goods and have

standard quality of goods. To ensure these standards the computer aided quality control

systems gain importance in marketing and producing. One of the most important thing in

wheat production is the seeds. The quality of seeds and being seeds consist of only wanted

species of wheat is so important. With low quality seeds it is inevitable to achieve low

efficiency. Additionally, mixes of unwanted species in the seeds creates extra costs on the

result product to select them. To avoid these problems, a machine vision based selection

study has been proposed. In this study some characteristic parameters that have been

obtained from images taken from seeds, have been used to determine class of wheat. The

attributes of this dataset are area, perimeter, compactness, length, width asymmetry

coefficient and length of kernel groove. 60% of the dataset has been used to train an

Artificial Neural Network (ANN) and rest of the dataset has been used for tests. To obtain

best structure of ANN, the number of neurons in the hidden layer of ANN has been

changed within 1 to 40. And the activation function of hidden layer and output layer has

been investigated for best result. Hyperbolic tangent sigmoid (Tansig), Logarithmic

Sigmoid (LogSig) and Linear (PureLin) transfer functions have been used as activation

function. The prepared ANN structures have been trained various training functions. The

Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) have been used as

performance criteria. All of the results of the researches have been compared and the best

results have been obtained as 0.1598 RMSE and 0.0273 MAE. The structure that has the

best results, has consisted of 25 neurons in the hidden layer with hyperbolic tangent

sigmoid activation function. And this structure has been trained by Levenberg-Marquardt

backpropagation function.

KEYWORDS - Artificial Neural Network, Quality Control, Seed



LOCAL T1 DISTANCES APPROACH SPACES

MEHMET BARAN1, MUHAMMAD QASIM1


ABSTRACT

In the field of theoretical computer science, mathematical structures have been developed

to model semantics of programming language [14]. In 1969, Dana Scott [11] introduced a

domain theory that could provide a mathematical foundation for denotational semantics.

Moreover, in 1972, D. Scott [12] proved that there is one-to-one connection between

continuous lattices and specialization orders of injective T0 topological spaces. This result

shows that there is a deep relationship between order-theoretic properties and topological

properties. Continuous directed complete partial orders (dcpos) endowed with the Scott

topology [13], [14] help to model lambda calculi as well as the denotational semantics of

computer programs. In 1989, Robert Lowen [15] introduced theory of approach spaces

which is based upon point-to-set distances rather than point-to-point distances. In 2011, E.

Colebunders et al. [16] have proved that approach structure on domains induces Scott

topology and characterized some fixed point theorems on these domains in [17]. In 1991,

Baran [4] introduced local separation properties in set-based topological categories and

later, they are generalized to point free definitions by using the generic element method of

topos theory [21] or [22]. In addition to, one of the fundamental usage of local separation

properties is to define the notion of strong closedness in set-based topological categories

that has been used in the notion of Hausdroffness ([4], [6]), Compactness [7], regular,

completely regular, and normal objects in ([8], [9]). In this paper, we characterize local T1

distance-approach spaces and investigate the relationship with usual T1 distanceapproach

spaces. ACKNOWLEDGMENT This research was supported by the Scientific and

Technological Research Council of Turkey (TUBITAK) under Grant No: 114F299 and the

Erciyes University Scientific Research Center (BAP) under Grant No: 7174.

KEYWORDS - Topological category, distance-approach spaces, gauge-approach spaces,

local T1 spaces.



DETERMINATION OF PLANT SPECIES USING VARIOUS ARTIFICIAL

NEURAL NETWORK STRUCTURES

MUHAMMED FAHRI UNLERSEN1, MURAT KOKLU2, KADIR SABANCI3

1 Necmettin Erbakan University, Turkey; 2 Selcuk University, Turkey;

3 Karamanoglu Mehmetbey University, Turkey

ABSTRACT

In daily life, humankind surrounded with many kinds of plants. But knowing all of the

species and characteristic of these plants is impossible. An application that gives

information about plants from its database could be very attractive. Not only botanist but

also anyone who loves plant/bass would interest on an application that determine species

or families of a plant automatically by using a photo of leaves taken instantly. In this

study, a dataset by using many species of plants leaf image has been created. The

attributes of dataset consist of some morphological and color based properties obtained by

image processing. There are 14 attributes with 340 instances. To obtain best results with

Artificial Neural Network (ANN) many structures have been investigated. Root Mean

Square Error (RMSE) and Mean Absolute Error (MAE) have been used as performance

criteria. While number of neurons in the hidden layer has been changing from 1 to 20, the

performance criteria has been observed. All of the tested structures mentioned above has

been trained with various training functions. Best structure and training function is

obtained when there are 15 neurons in the hidden layer with LogSig activation function.

For best situation the RMSE and MAE are 0.0007 and 0.0001 respectively. The training

function is scaled conjugate gradient backpropagation.

KEYWORDS - Artificial Neural Network, Determination of Plant



COMPARISON OF OPTIMIZATION METHODS BASED ON GSA

NIHAN KAZAK1, NESIBE YALCIN1, ALI ERDEM CERCEVIK1


ABSTRACT

In recent years, many heuristic evolutionary optimization algorithms have been developed.

Gravitational search algorithm (GSA) is one of these algorithms. It is inspired by

Newton’s law of universal gravitation. In this paper, we compare two modified algorithms

based on GSA. One of the algorithms is called Member-Satellite algorithm. Members are

randomly positioned in the search space and a certain amount of satellites are assigned

around the members in the predetermined region. Members and their satellites are used to

find a near optimal solution all together. The second one is MSS-GSA. When

interconnected objects are used, it is possible to obtain a solution closer to the optimum

point. For this reason mass spring system is integrated into the GSA. Three benchmark

functions are used to compare performance. Experimental results show that the highest

performance is obtained with Member-Satellite algorithm.

KEYWORDS - gravitational search algorithm, heuristic algorithm, optimization, swarm

intelligence



A SMART PARKING LOT SENSOR SYSTEM

PINAR KIRCI1, SELEN SAGLAMOZ1, MURAT SENER1

1 Istanbul University, Turkey

ABSTRACT

Today, technology improves and facilitates our lives.Technology takes place in daily life

with smart phones, smart cars, smart homes and smart clocks. Thus, these smart systems

and devices take part during lifetime. To become a smart program or a system, some

characteristics should be included: - To make decision - To sense - To learn - To solve

problem - Reasoning - To recognize figures or pictures - To understand natural language

The capability of carrying out variable jobs and applications simultaneously makes smart

systems indispensable.Smart portable systems may provide almost all of the applications

which are provided by a stationary computer system. We considered main requirements of

customers in a parking lot; Automatic barrier system, empty/occupied area sensing and

monitoring. Empty parking lots are monitored with sensors that are placed on the parking

lot floors, in our project.Also, the number of the empty parking lots are presented at the

entrance of the parking lots. Thus, if there is not an empty parking lot then the customer

does not enter the parking garage and leave.So, time and money is saved. As soon as a

customer leaves its parking lot, its place will be realized by the system. After it leaves

from the exit, its place will shown as empty for preventing the crowd. At the entrance and

exit, possible disorders are prevented. We utilized an Arduino Mega 2560 which is an

ATmega2560 base circuit board. It is programmed over Arduino IDE. CNY 70 konstrat

sensors are used to decide the empty parking lots. CNY 70 sensor emits infrarosso and

produces analog outputs according to taking infrarosso back or not. It helps us to

determine if there is a reflective surface in front of the sensor. If there is a surface in front

of the sensor, we understand that there is a car parked on the considered parking lot. The

sensors which are occupied by cars are represented by red lamplights which are on. In the

first scenario, at the entrance the number of occupied parking lots are given as three at the

table of the parking lot together with empty parking lot as one. For this reason, the

automatic barrier is opened for the fourth car to enter in the parking lots area.

KEYWORDS - Sensors, wireless networks, parking lots, cost



EMPIRICAL EXPLORATION FOR THE CORRELATION BETWEEN OBJECT

ORIENTED CLASS COHESION AND INHERITANCE

JEHAD AL DALLAL1

1 Kuwait University, Kuwait

ABSTRACT

The overall quality of the object-oriented software is greatly affected by the design of the

internal relationships among object-oriented class members (i.e., attributes and methods)

and the external relationships among classes. To maintain high quality and reduce code

redundancy, developers are advised to build classes with high degree of cohesion (i.e.,

relatedness among class members) and implement class inheritance concepts whenever

applicable. In object-oriented systems, instead of redefining the attributes and methods

that are included in other classes, a class can inherit these attributes and methods and only

implement its unique attributes and methods, which results in reducing code redundancy

and improving code testability and maintainability. In this paper, we empirically

investigate the relation between class cohesion and inheritance using classes of three open-

source Java systems. We consider five cohesion and two inheritance measures. The

empirical study results show that cohesion and inheritance internal quality attributes are

inversely correlated. In addition, the results indicate that the cohesion and inheritance

quality attributes are almost orthogonal.

KEYWORDS - object-oriented class, software quality, class cohesion measure, class

inheritance measure



A SMART CAR PARKING SYSTEM BASED ON GPS COORDINATES

GOKHAN KAYHAN1, ISMAIL ISERI1, SEVKI UTKU MALKOCOGLU1


ABSTRACT

While technology has evolved very quickly, it has brought along many innovations and

conveniences. In this regard, technology has become an indispensable part of our lives.

Nowadays, most people use smartphones and the applications developed for smartphones

make people's business easier and managable. In this study, the technological facilities

provided by smartphones were used to solve the problem of finding parking space in the

parking lots.The proposed parking system solves two fundamental problems. The first

problem is to find the location of the suitable parking areas in the parking lots. The second

problem is to help the drivers to find their previously parked car in the parking lot. The

system also guides to the nearest free parking place in a multi-storey car park by using the

altitude property of the device’s GPS module. The car parking system which uses mobile

device’s GPS module allows the drivers to find the nearest available parking space in the

parking lot. This car parking system will reduce time spent looking for a parking spaces in

parking lots and help to make city parking less stressful and optimally distribute parking

spaces.

KEYWORDS - Android, Car Parking System, Google Maps API, Volley API, GPS

Module.



QR CODE BASED ENCRYPTION AND DECRYPTION OF TRIANGULAR

GEOMETRY PROBLEMS

CHEIKHNA LO1


ABSTRACT

This study focuses on the design and implementation of a QR code-based encryption and

decryption system for triangular geometry problems. The encryption stage starts with the

development of a context-free grammar to describe triangular problems in a formal

language. Triangle descriptions are then encrypted using a traditional cryptography

algorithm and lastly the corresponding QR code is generated. In the decryption stage, the

QR code is decrypted and a parser, which is automatically produced by the JavaCC tool, is

used to analyze and show the related problem graphically. The developed system provides

QR code reader-like tool and presents an elegant way to be able to display mathematical

problems on various devices easily. With such a tool, the whole questions of an exam can

be encoded in QR code and stored in smaller sizes. This would significantly increase the

security of exams and decrease question data, which can be particularly transmitted on

mobile devices, in size.

KEYWORDS - Formal grammar, encryption, QR code, security, geometry



COMPARISION OF ORGANIZED INDUSTRIAL ZONES EFFICIENCY BY

USING A COMMON WEIGHT DEA BASED DECISION MODEL

BURCU DENIZ TULEK1, NAZLI GOKER1, MEHTAP DURSUN1


ABSTRACT

Organized Industrial Zones (OIZs) are the production areas of goods and services that are

established to provide planned industrialization and planned urbanization by structuring

the industry in suitable areas, to prevent environmental problems and to provide efficient

use of resources. They began to be established and supported by the Turkish government

since 1960s and at the present time there are one or more OIZs in all cities of Turkey

(except Artvin). The low occupancy rates of these OIZs make the question of whether the

OIZs work efficiently. This study aims to evaluate the efficiencies of OIZs located in the

Eastern Black Sea Region of Turkey by using a data envelopment analysis based common

weight decision model, which is a powerful method to measure the relative efficiencies of

decision making units.

KEYWORDS - Common weight DEA based decision model, Decision support systems,

Multi-criteria decision-making, Organized industrial zones;, Performance evaluation



MODELING OF SOCIAL PERFORMANCE METRICS WITH ARTIFICIAL

INTELLIGENCE TECHNIQUES

ENES CELIK1, ADIL KONDILOGLU2, HARUN BAYER1, MUHAMMET ATALAY1

1 Kirklareli University, Turkey; 2 Beykent University, Turkey

ABSTRACT

The social media yield a huge flood of heterogeneous data that is voluntarily published

and shared by billions of individual users all over the world. As a result, the representation

of an entity in this data may consist of various data types, including location and other

numeric attributes, textual descriptions, images, videos, social network information and

other types of information. Searching similar entities in this multi enriched data exploiting

the information of multiple representations simultaneously promises to yield more

interesting and relevant information than searching among each data type individually.

This study presents a research approach using artificial intelligence techniques for model

the performance metrics of posts published in brands' Social media pages. Firstly, the

performance metrics data was classified with machine learning algorithms and then

training and test results were analysed. The estimation results of the performance metrics

were compared with similar data and similar studies. We have drawn a decision process

flow from model, which by complementing the sensitivity analysis information may be

used to support manager's decisions on whether to publish a post.

KEYWORDS - machine learning, social network, Decision support system, metrics,

Artificial Intelligence



IMPLEMENTATION OF DEFECT DETECTION ON MILITARY CARTRIDGE

WITH OPENMP

OMER FARUK BAY1, SEMRA AYDIN1, REFIK SAMET2

1 Gazi University, Turkey; 2 Ankara University, Turkey

ABSTRACT

Image processing techniques requires long processing time. It is known that time is crucial

in real time applications. Serial image processing does not satisfy real time conditions

adequately. To decrease processing time, parallel computing techniques can be used.

Parallel computing techniques have two types of memory model named shared and

distributed. OpenMP is a parallel programming model for shared memory machines. This

paper describes the implementation of defect detection algorithm on military cartridge

using OpenMP. Experimental results showed that using OpenMP platform give us an

acceleration of 6.34 times in comparison with serial computing.

KEYWORDS - Parallel processing, parallel computing, shared memory model, OpenMP,

defect detection



NEAR FIELD COMMUNICATION BASED ROLL CALL SYSTEM

SEYIT ALPEREN CELTEK1, MAHMUT DURGUN1, KADIR SABANCI1

1 Karamanoglu Mehmet Bey University, Turkey

ABSTRACT

Near-field communication (NFC) is a one of the wireless communication technologies that

enable two way interactions between electronic devices, one of which is usually a portable

device such as a smart device. This research has developed a roll call system for school

using NFC technology. The proposed systems consist of hardware and software

components. The smartphone supported NFC used as a hardware. The android program

developed for the read student card and record attendance information to internet. The

NFC enabled roll call systems aims at reduce paper work and waste of time. Also

authorities and parents are able to learn real time attendance information. The proposed

NFC roll call system will be a valuable guide for universities and schools which would

like to change conventional roll call model to smart attendance system.

KEYWORDS - NFC, Attendance Sytems, Roll Call System



SELECTION OF A SUITABLE WIND POWER PLANT WITH ECONOMICAL

ANALYSIS FOR AMASRA CITY TURKEY

YAGMUR ARIKAN1 , OZGE PINAR AKKAS1 , ERTUGRUL CAM1


ABSTRACT

In this study, the wind energy potential of the city of Amasra in Turkey was analysed.

Data from the General Directorate of Turkish Meteorology for the years of 2012-2013

were taken. The data had been recorded for a height of 10 m by the Directorate. In the

analyses, annual average wind speed of the city was found as 5.77 m/s. In order to

calculate the annual wind energy generation and capacity factor of the city, a 600-kW

turbine belonging to Nordex company and 900 kW and 2300 kW turbines belonging to

Enercon company were used. Thus, investment feasibility with 3 different turbines was

analysed. In the economic analysis, payback and net present value methods were used. In

conclusion, it became clear that the use of 2300 kW turbine in the region would be more

profitable.

KEYWORDS - wind energy, wind turbines, capacity factor, cost analysis



IN SITU UV EXCITATION STUDY ON MATRIX ISOLATED 2 CHLORO 6

FLUOROBENZOIC ACID

NIHAL KUS1, RUI FAUSTO2

1 Anadolu University, Turkey; 2 Coimbra University, Portugal

ABSTRACT

2-Chloro-6-fluorobenzoic acid was studied by matrix isolation IR spectroscopy, and the

analysis of the experimental vibrational data was supported by DFT(B3LYP) calculations.

According to the theoretical calculations, the ClFBA molecule exists in three different

conformers. Conformers II and III are higher in energy than I by 17.1 and 17.3 kJ mol−1

(16.0 and 16.5 kJ mol−1, after consideration of zero-point corrected energy), respectively.

According to the energy data predicted theoretically only conformer I was expected to be

present in the cryogenic xenon matrix, a result that was confirmed experimentally. Laser

UV ( = 235 nm) excitation of matrix-isolated ClFBA led to prompt decarboxylation of

the compound, with production of CO2 and 1-chloro-3-fluorobenzene (ClFB), whose

vibrational signature could be doubtlessly identified in the spectra of the photolysed

matrix.

KEYWORDS - 2-Chloro-6-fluorobenzoic acid, Matrix isolation spectra, DFT(B3LYP)

calculations, FTIR spectroscopy, 1-Chloro-3-fluorobenzene, UV-induced photolysis



P TYPE DYE SENSITIZED SOLAR CELLS ENHANCED PERFORMANCE

WITH A GRAPHENE CUO NANOSTRUCTURES

BAYRAM KILIC1, SUNAY TURKDOGAN1

1 Yalova University, Turkey

ABSTRACT

A photoactive CuO electrode for p-type dye-sensitized solar cells (p-DSCs) has been

designed by the introduction of graphene coatings layer on FTO substrate. This study

examined the facile process for the fabrication of a compact CuO nanostructures. The

N719-dye used as the photosensitizer for the CuO electrode in the p-DSCs device and their

performances have been analyzed. The enhancement of both short-circuit photocurrent

(Jsc) and fill factor (FF) resulted in a twofold increase in the power conversion efficiency

(h). Dye-sensitized photocathodes (p-DSCs) operate in an inverse mode, where dye-

excitation is followed by rapid electron transfer from a p-type semiconductor to the dye

(dye-sensitized hole injection). Here we show that p-DSCs can convert absorbed photons

to electrons with resulting in a sevenfold increase in energy conversion efficiency

compared with previously reported photocathodes.

KEYWORDS - Graphene, CuO, DSSC,P-TYPE DSSC



INVESTIGATION OF FIRE EXTINGUISHER TUBE STRESS AND

DEFORMATION CONCEPTS UNDER THE OVERPRESSURE LOADING FOR

DIFFERENT MATERIALS

MEHMET HANIFI DOGRU1, IBRAHIM GOV1, KURSAD GOV1, EYUP

YETER1, IBRAHIM HALIL GUZELBEY1

1 Gaziantep University, Turkey

ABSTRACT

In industrial applications, four different types of fire extinguisher tubes are used. These are

water fire extinguisher, foam fire extinguisher, dry powder fire extinguisher, and CO2 fire

extinguisher. Overpressure concept is so crucial for the fire extinguisher tube in filling

time or when the tube exposed to high temperature. Body materials are also important to

prevent tearing for these type of tubes. In this study, fire extinguisher tube, which is 1.3 kg

CO2 fire extinguisher, is investigated in terms of the overpressure loading. Three different

materials, which are ck45 steel, aluminum, and glass epoxy composite, are used to

perform this study. Symmetrical tube model is created according to Turkish standards,

which are TS 1519 and TS 11169. Numerical simulation of the study is performed using

ANSYS-Workbench, which is finite element program. Analyses system, which are static

structural and fatigue analysis, is used to perform numerical study. Stress and deformation

concepts are investigated. Factor of safety and fatigue life are also investigated for the fire

extinguisher tube.

KEYWORDS - Fire extinguisher tube, finite element analysis, overpressure, stress, factor

of safety.



VOLTAGE BREAKDOWN AND LEAKAGE CURRENT STUDIES OF TIO2

AL2O3 CERAMIC NANO COMPOSITE CAPACITOR

ALARA ULAC1, DOGAN AKCAN1, MEHMET CAN ALPHAN1, LUTFI ARDA1, ALI

GUNGOR1


ABSTRACT

Despite capacitors which have a fairly simple structure have many applications in fields

like energy distribution, electronic circuits, and energy storage. Due to its wide application

fields, increasing performance of capacitors is a very intensely studied topic. In this study,

effect of composition ratio of components in nanocrystalline TiO2-Al2O3 ceramic

composites (TAO) on capacitance, break down voltage, and leakage current is

investigated. By using nanocrystalline structure flexible capacitors are obtained which can

extend applications of ceramic capacitors. Capacitor structure is formed by sandwiching

TAO between readily cleaned aluminum plates. TiO2/Al2O3 ratio is arranged from 100%

to 0% by weight. Ceramic mixture is grinded in agate mortar and coating paste is prepared

by introducing a few drops of acetic acid – water – dish soap solution. As prepared paste is

spread onto surface as a thick film by using doctor’s blade method. Coated plates are

sintered in muffle furnace at 500°C for 30minutes to remove organics in film structure.

Crystalline structure of the coatings examined by X-ray diffraction (XRD) and surface

morphology is examined by SEM. To form nanocrystalline ceramic capacitor two thick

film coated plates are sandwiched. Besides ceramics, effect of polymer introduction to

system is investigated by introduction of polyethylene (PE) sheets between coated plates.

Break down voltage of TAO ceramic capacitors have been investigated via high voltage

power supply. Leakage currents of capacitors tested with high resistance precision

electrometer alongside the DC capacitance measurements. The AC characterization of the

capacitors have been conducted with an impedance analyzer by means of analyzing

dielectric losses at room temperature.

KEYWORDS - Voltage breakdown, leakage current, TiO2-Al2O3, nano-composite

capacitor



REACTION OF ACETOPHENONE FORMALDEHYDE RESIN WITH S

TRIAZONE COMPOUNDS FOR FIRE RESISTANCE POLYURETHANE

MINE MUTLU1, NILGUN KIZILCAN1, AHMET AKAR1, NESRIN KOKEN1

1 Istanbul Technical University, Turkey

ABSTRACT

In this study, acetophenone formaldehyde resin (AFR) was synthesized primarily in basic

medium. Then, AFR was modified via their hydroxyl groups by reaction with s-triazone

that was produced by the reaction of urea with the mixture of formaldehyde and alendronic

acid. S-triazone reacts with formaldehyde in the presence of acid catalyst to give methylol

derivatives. Methylol derivatives of s-triazone give condensation products in an acid

media with resin. The mol ratio of s-triazone / resin was about 1/10. Insertion of functional

s-triazone cyclic structure to the AF-Resin effected its physical properties considerably.

The products were characterized by FTIR, 1H-NMR, 13C-NMR and 31P-NMR

spectroscopic analysis.

KEYWORDS - Resin, acetophenone, s-triazone, formaldehyde.



USING A NEW METHOD BASED ON FINSLER GEOMETRY FOR WIND

SPEED MODELLING

EMRAH DOKUR1, SALIM CEYHAN1, MEHMET KURBAN1


ABSTRACT

Accurately modelling of wind speed is very important for the assessment of wind energy

potential of a certain region. Before the installation of a wind energy conversion system in

a region, the wind speed potential of that region needs to be determined and modelled. For

this reason, different distribution functions such as two-parameter Weibull, Gamma,

Lognormal, Rayleigh etc. are proposed for accurately modeling wind speed in the

literature. In this paper, new probability and cumulative probability density functions

based on Finsler geometry are proposed for wind speed modelling. Two-dimensional

Finsler space metric function is obtained for Weibull distribution. Monthly analysis for

Yalova, Turkey is realized using a new method based on Finsler geometry and two-

parameter Weibull distribution. Wind data, consisting of hourly wind speed records

between October 2015-September 2016 were obtained from the Yalova station of Turkish

State Meteorological Service. The performances of the models are given comparatively by

using root mean square error (RMSE).

KEYWORDS - Finsler Geometry, Wind Speed, Modelling, Weibull Distribution,

Renewable Energy.



EFFECT OF INDIUM INCORPORATION ON THE STRUCTURAL AND

OPTICAL PROPERTIES OF ZNS NANOSTRUCTURED FILMS USED AS

WINDOW LAYER IN SOLAR CELLS

TULAY HURMA1

1 Anadolu University, Turkey

ABSTRACT

ZnS and ZnS:In films have been deposited by ultrasonic spray pyrolysis method. Three

different atomic ratios of indium chloride were used as a dopant element. The effect of

indium incorporation on structural and optical properties of ZnS films has been

investigated. XRD spectra show that the crystalline quality decreased with incorporation

of small amount of lithium in the spraying solution. The crystallite sizes were calculated to

be in the range of 4-5 nm. All the films exhibit the Raman hump-like feature around 131,

350 and 559 cm-1 and the hump profiles are almost symmetric. The optical band gap of

these films were determined. The optical properties of In doped ZnS films changed

compared to undoped ZnS film.

KEYWORDS - In doped ZnS film, XRD, FTIR, Raman, Optical properties



SYNTHESIS OF GLYCOLURIL DERIVATIVES FOR FIRE RESISTANCE

POLYURETHANE

NILGUN KIZILCAN1, ESRA TUNCEL1, AHMET AKAR1


ABSTRACT

Glycolurils have been received a great attention due to their applications as fertilizers (1),

psychotropic agents, stabilizers of organic compounds against photodegradation (2),

explosives (3), polymer crosslinking agents (4, 5), catalysts, bleaching activators (6- 8),

and their use in combinatorial chemistry (2, 9). Glycolurils are also important building

blocks for both molecular and supramolecular chemistry (10). Particular attention has been

turned to cucurbiturils (11) (CB[n]) which are intriguing macrocyclic compounds, whose

skeleton is constituted by glycolurils ring moieties (12-15). Glycoluril oligomers were also

synthesized before (16-17). In this study, condensation product of glycoluril with

formaldehyde and ethanol amine was produced and used as polyol in the production of

rigid polyurethane form in order to improve its fire resistance. In this study, condensation

product of glycoluril with formaldehyde and ethanol amine was produced and used as

polyol in the production of rigid polyurethane form.

KEYWORDS - Glycoluril, glycoluril dimer, ethanol amine, urea, formaldehyde.



EXPERIMENTAL INVESTIGATION OF HEAT STORAGE BEHAVIOUR IN A

PHASE CHANGE MATERIAL UNDER SOLAR RADIATION

MEHMET EMIN CANLI1, ADEM ACIR1


ABSTRACT

This study was aimed to investigate heat storage behaviour in a phase change material

(PCM) under solar radiation as experimentally. For the purpose of this, an experimental

set-up was designed and paraffin wax supplied from AGS Paraffin Company was used.

Two different types of PCM storage media were produced as without fin (Type-I) and with

fin (Type-II). Three different constant solar radiation fluxes were applied on PCM storage

boxes by using infrared lamb in laboratory conditions. The melting time and

characteristics of designed Type-I and Type-II were investigated. The results showed that

fin appliance under solar radiation absorber surface fastened paraffin wax melting,

moreover, it improved the internal heat transfer uniformity of paraffin wax as it is

expected. This study was accepted the first step for future studies about solar energy

storage systems with trying to understand melting characteristics of paraffin wax.

KEYWORDS - Solar energy, Phase change material (PCM), Thermal energy storage

(TES), Paraffin wax, Heat transfer



INVESTIGATION OF NONLINEAR PATHS OF FLUIDS FLOW DETERMINED

BY FINSLER GEOMETRY THROUGH INHOMOGENEOUS POROUS MEDIA

DERYA ULUG1, SALIM CEYHAN1


ABSTRACT

In this paper, nonlinear paths of the fluids flow determined by the Finsler geometry

conforming to Darcy's law have been investigated through inhomogeneous porous media.

Geodesics in a Finsler space are used to determine the non-linear paths of a fluid flow in

inhomogeneous anisotropic porous medium.Significant differences occur between the

Riemannian and Finslerian geodesics due to the directional dependence of Darcy’s flow of

fluids. In an optimum path (a geodesic) in physical space, it guarantees the maximum flux

or shortest transition time of the fluid through inhomogeneous porous medium. Fermat’s

variational principle is used to minimizing the total resistance through inhomogeneous

porous medium. The fluid streamlines move between areas exposed to minimal resistance.

Therefore, the path of Darcy’s flows of fluid can be defined by geodesics in Finsler

geometry [1]. The various functional properties of the hydraulic conductivity of porous

media are examined and given graphically. In this study, various functional structures of

hydraulic resistance of the inhomogeneous porous media are examined and the results

obtained are given graphically.

KEYWORDS - Darcy’s flow of fluids through inhomogeneous media, Fermat’s variational

principle, Finsler geometry, Prous media, Hydraulic head, Kropina metric



THE EFFECTS OF INCENTIVES ON RENEWABLE ENERGY RESOURCES

FOR HOME USERS

FATIH BULUT1 , MURAT LUY1 , ERTUGRUL CAM1


ABSTRACT

In this study, economical and emission analysis for meeting the electricity demand of a

randomly selected house which has renewable energy sources was carried out. The house

chosen as an example is in Balikesir city. As a household load, the average daily

electricity needs of a family of four were accepted and different scenarios were planned

using the Hybrid Optimization Model for Electric Renewable (HOMER) program to meet

this need. Several scenarios were dealt with as to whether the systems designed for grid-

connected photovoltaic, wind turbine and hybrid systems were indigenous manufacturing

incentives. In addition, energy unit costs and total investment costs were determined and

economic analysis was carried out. When the results of the study are examined, grid-

connected photovoltaic, wind turbines and hybrid systems were found to be significantly

less than the costs incurred by net present costs when they were established within the

scope of domestic manufacturing incentives.

KEYWORDS - incentive, investment, grid-connected,Renewable, domestic



FREQUENCY MODULATED IMPULSE RESPONSE PHOTOTHERMAL

REFLECTANCE THEORY OF JUNCTION VERTICAL PARALLEL SILICON

SOLAR CELL

GOKHAN SAHIN1

1 Igdir Universitesi, Turkey

ABSTRACT

This study investigates the study of the photo-thermal and temperature response of a

junction vertical parallel silicon solar cell illuminated under multispectral illumination

with a constant modulated frequency. Solving the continuity equation for minority carriers

in the base of the solar cell through the use of phenomenological parameters such as

recombination velocities at the junction and the back of the base and the thermal behaviour

of the solar cell, we analysed and studied the density of minority charge carriers, the

temperature amplitude and the density of heat flow. Representations of Nyquist and Bode

plots of the thermal dynamic impedance resulted in an equivalent electrical circuit of the

photocell.

KEYWORDS - Vertical parallel Junction silicon solar cell, Temperature, Heat flux, Bode

and Nyquist diagram.



TRANSMISSION EXPANSION PLANNING INCLUDING POWER LOSS COST

USING LINEARIZED AC MODEL

MEHMET FATIH CANKURTARAN1, FARUK UGRANLI2, ENGIN KARATEPE1

1 Dokuz Eylul University, Turkey; 2 Bartin University, Turkey

ABSTRACT

Transmission expansion planning (TEP) is a large-scale and mixed integer nonlinear

problem. Due to its nonlinear nature and large scale problem, it is hard to solve this

problem without any linearization. The motivation of this work is to present the

importance of power loss in the TEP problem. In this study, a linearized-AC network

model is used to solve this problem because of the lack of DC-model. In that manner, the

cost turnover time is evaluated by including both line investment and power loss costs in

the objective function. In the problem, bi-level programming is used to solve both

economic dispatch and TEP problem with together, where TEP problem is solved at upper

level and economic dispatch is solved at lower level. The linearized-AC and DC-lossy

models are applied to the IEEE 24-bus test system. Simulation results can be interpreted in

terms of cost turnover time when considering power loss in the TEP problem. Moreover,

the expanded networks obtained from the linearized-AC and DC-lossy models are tested

with full AC representation of transmission network for network constraints and

presenting trade-offs between two models.

KEYWORDS - Transmission expansion planning, AC model, power loss, mixed-integer

linear programming



EFFECT OF TRISO FUEL OF THE BURNING OF REACTOR GRADE

PLUTONIUM IN CANDU REACTOR

ADEM ACIR1, TAHA HUSEYIN MERMER1


ABSTRACT

The paper investigates the effect of the TRISO fuel of the burning of reactor grade

plutonium in CANDU reactors. TRISO fuels particles are imbedded random packing (RP)

in a graphite matrix with a volume fraction of 61%. Calculations have been performed

SCALE-XSDRNPM code. The three different mixed fuel were performed for CANDU

reactor containing 10 % RG-PuO2 + 90 % UO2; 20 % RG-PuO2 + 80 % UO2 and

50 % RG-PuO2 + 50 % UO2. The initial criticality (k,0 values) were obtained as 1.612,

1.638 and 1.681 for each modes, respectively. The operation times computed ~ 1.51, 3.02

and 6.86 years whereas; the corresponding burn ups values are ~ 77 000, 161 000 and 447

000 MW.d/tonne, respectively. TRISO fuel of CANDU reactor has been provided the

higher burn ups according to original fuel.

KEYWORDS - CANDU reactors, TRISO fuel, Uranium, Plutonium, High burn up



SELECTION OF A SOLAR POWER PLANT LOCATION BY USING AHP

METHOD

OZGE PINAR AKKAS1, YAGMUR ARIKAN1, ERTUGRUL CAM1


ABSTRACT

In recent years, investments of Solar Power Plant (SPP) have increased since the cost of

photovoltaic has decreased. Nevertheless, feasibility studies remain important as the cost

of installation is still relatively high. One of the most important parameters in the

feasibility studies is the appropriate location selection. In this study, the appropriate

location selection for the SPP was studied with the Analytical Hierarchy Process (AHP)

which is one of the Multi Criteria Decision Making (MCDM) methods. For the

application, Elazığ, Kahramanmaraş and Malatya cities, which are located on the same

plane, were selected as examples. As criteria, solar energy potential, earth slopes and

feeder capacities, which are important in feasibility studies, were considered. In the results

of the study, Kahramanmaraş was found as the most suitable city for establishment of SPP.

It was followed by the cities of Malatya and Elazığ. The results were approved by

considering the data of cities' sunshine duration and radiation value in Solar Energy

Potential Atlas (GEPA). It was also been proven that the method may be recommended for

SPP installations.

KEYWORDS - Renewable energy, solar power plant, Multi Criteria Decision Making,

Analytic Hierarchy Process



DETERMINATION OF SOME CHARACTERISTIC PARAMETERS OF

PHOTOVOLTAICS USING DIFFERENT ANALYTICAL METHODS

AYSEGUL TOPRAK1, HAMDI SUKUR KILIC1


ABSTRACT

Renewable energy is generated from natural resources such as wind, water, geothermal

and solar energy (Photovoltaics, PV). Among the various renewable energy sources, solar

energy is one of the most promising source and gaining attraction for a wide usage as a

natural source of energy. It is foreseeable that solar power generation can contribute

prominently to overcome the indisputable problems of global warming and climate

change. Solar energy is the ultimate source of power. The operation principles of all solar

cells of present days is essentially similar and based on the photovoltaic effect. In general,

the photovoltaic effect means the generation of a potential difference at the junction of two

different materials in response to visible or other radiation. Photovoltaic solar cells can

convert sunlight directly into electricity. The photovoltaic panels is defined by an

equivalent circuit model. Four parameters model is one of the commonly used in literature.

Knowing the photovoltaic panel parameters is of importance for the design and estimates

the performance of solar cells. Therefore, the parameters are needed to be calculated by a

proper method. The determination of some characteristic parameters for the four parameter

model using simplified method and slope method is the main purpose of the present work.

The results obtained with different methods were analysed and compared with those of the

manufacturer datasheet used. Further it has been determined that the different radiation

and the different temperature is effective on the photovoltaic parameters.

KEYWORDS - Photovoltaic, Four Parameters Model



SERVO MOTOR DRIVEN HUMAN GYROSCOPE STRUCTURE FLIGHT

SIMULATOR

KURSAD GOV1, EYUP YETER1, MEHMET HANIFI DOGRU1, IBRAHIM

GOV1, IBRAHIM HALIL GUZELBEY1


ABSTRACT

In this study, a flight simulator was designed as a human gyroscope structure and its

prototype was manufactured. Normally, human gyroscopes are used for entertainment

application in funfair. They do not have any driven system; their movements are actuated

by the rider movement. Human gyroscope is used for circulatory workout and balance

training for pilots and astronauts. They can also be used as flight simulator. In this study, a

prototype human gyroscope type flight simulator was designed and manufactured. In the

design stage of the human gyroscope, motor driven systems are the most crucial stage. So,

the motors’ selection and calculations are very important. Also, the structure of the flight

simulator is important, it must be light in weight and strong for rapid movement. The

servo motors, servo motor drivers and gearboxes are chosen as an actuator of the flight

simulator and the calculations have been done according to the servo motor actuators.

KEYWORDS - Flight simulator, human gyroscope, servo motor, actuators, prototype



SHORT TERM FORECASTING OF RENEWABLE ENERGY PRODUCTION

ACCORDING TO DAILY DATA USING ARTIFICIAL NEURAL NETWORK

BAYRAM AKDEMIR1, FATIH MEHMET SENALP1


ABSTRACT

Renewable energy is getting common day by day to improve government acts and

capability. Moreover, diversified power plants lead to reduce primary source risks. On the

contrary, renewable energy sources are limited to use up and wait for second cycle for

regeneration. In this study, solar energy plant for 1MVA is observed and compared with

simulations and real values. Solar power generation is related to environmental conditions

to achieve maximum power point. Temperature, wind speed, humidity and irradiation

parameters are scrutinized and sourced for artificial neural network inputs to forecast

possible energy production for daily using. As first, backward data structure is created and

filtered in case of solar power system failure. Arranged data is used to feed artificial neural

network inputs to obtain forecasted solar energy production. Artificial neural network is

widely used to solve nonlinear problems in many areas such as aerospace, construction,

medical and etc. Data structure has nonlinear characteristics and relations among the

variables effecting the output. Artificial neural network tries to find best possible solution.

Target is to find possible energy production using daily recorded data for the next day.

Proposed method finds possible energy production capability for the next day using

backward data. In order to evaluate the results, mean square error and R2 statistical

evaluation methods were used. R2 is a way to present for expected and obtained values

cover each other or not. In addition to R2, mean square error is used to evaluate system

success. In this study, artificial neural network is used to forecast short term energy

production for real 1 MVA solar plant. Obtained success is calculated as 94.2% and R2

value is 0.993.

KEYWORDS - Artificial neural network, solar energy, short term forecasting, data set,

mean absolute error, R2 validation.



COMPARISON BETWEEN HORIZONTAL AND VERTICAL PARALLEL

SILICON SOLAR CELL OF A PHOTOCURRENT DENSITY UNDER

MULTISPECTRAL ILLUMINATION

GOKHAN SAHIN1

1 Igdir Universitesi, Turkey

ABSTRACT

In this work, a theoretical study of a horizontal and junction vertical parallel silicon solar

cell of a photocurrent density under multispectral illumination is presented. The

photocurrent density behavior versus modulation frequency and junction recombination

velocity is presented. The study of this photocurrent, according to the junction

recombination velocity, allows us to determine the junction recombination velocity

limiting the short circuit. The photocurrent density is studied as a function of the junction

recombination velocity for different modulation frequencies. From the variation profile of

the photocurrent density as a function of the junction recombination velocity, a technique

for determining junction recombination velocity is presented. We also pointed out the

influence of the incidence angle on the photocurrent density of the solar cell. The aim of

this paper is to make a comparison of the photocurrent produced by the horizontal junction

solar cell and those produced by junction vertical parallel solar cell. A theoretical

discussion about the junction recombination velocity and angular frequency dependence of

the conversion efficiency is also reported. I used the different data, the datas are frequency

(ω) =103rad, junction recombination velocity (Sf) =3.103cm/s, thickness of photopile in

base (H)=0.03cm, length of diffusion of excess minority carrier’s in the base

(L0)=0.0001cm, coefficient of diffusion of charge carriers in the base of the photovoltaic

cell (D0)=26cm2/s, depth in base (z)꞊0.0001cm.

KEYWORDS - Horizontal Junction Silicon Solar Cell, Vertical Parallel Junction Silicon

Solar Cell, Photocurrent density, Conversion efficiency



FATIGUE ANALYSIS OF LANDING GEAR OF A LIGHTWEIGHT AIRPLANE

IBRAHIM GOV1, MEHMET HANIFI DOGRU1, EYUP YETER1, KURSAD

GOV1 , IBRAHIM HALIL GUZELBEY1


ABSTRACT

In this study, the landing gear of a lightweight airplane is analysed to investigate fatigue

performance. For this investigation, finite element analyses (FEA) program ANSYS is

used. Landing gears are very critical structural part of the airplanes. When designing a

landing gear, static, fatigue, dynamic and aerodynamic analyses must be performed. Static

analysis is the starting point of the analysis steps. The fatigue analysis, which is second

step of the analysis, must be carried out to see long life working performance. Hence in a

previous study, the static analysis of landing gear was performed in terms of stress and

deformation. And in this study, fatigue analysis is performed in terms of fatigue life and

safety factor. Landing gear assembly is divided into two parts and these parts are analysed

separately. Most critical part is improved to increase the fatigue life value. After this

improvement, fatigue life value is increased nearly 135% and minimum safety factor value

is increased approximately 35%.

KEYWORDS - landing gear, finite element analysis, fatigue analysis, fatigue life, safety

factor



STRESS ANALYSES OF DIFFERENT MECHANIC CONNECTION SHAPES

USED TO CONNECT COMPOSITE STRUCTURES

EYUP YETER1, KURSAD GOV1, IBRAHIM GOV1 , MEHMET HANIFI

DOGRU1 , IBRAHIM HALIL GUZELBEY1


ABSTRACT

Connection of composite structures is so crucial to prevent failure of this structures. In this

study, effects of different mechanical connection shapes on the structural strength

performances have been investigated. And Glass/Epoxy composite structures are used in

this investigation. Stress, deformation, fatigue life, and safety factor distribution of the

structure, which are connected using different connection shapes, are researched.

Butterfly, T-shape, and circular shape mechanical connections are used. Since the main

aim of the study is to understand the behavior of composite structures that are connected

without any bonded surface, the connections are only connected mechanically. All

analyses are performed using commercial finite element analysis program ANSYS.

Comparisons of stress, deformation, fatigue life, and safety factor distribution of

composite plate connected with different connection shapes have been performed.

KEYWORDS - Composite structures, finite element analysis, fatigue analysis, fatigue life,

safety factor



ENHANCEMENT OF PYROPLASTIC DEFORMATION BEHAVIOUR IN

CERAMIC SANITARYWARE FFC SLIPS

NAZIM KUNDURACI1, IREM NUR GAMZE SIMSEK1 , ENGIN KOCAMAN1

1 Bulent Ecevit University, Turkey

ABSTRACT

Ceramic sanitarywares are the products that are obtained by preparing a slurry of inorganic

ceramic raw materials such as clay, kaolin and feldspar in certain ratios, shaping by

convenient techniques and then firing at 1200–1250 °C, They have water absorption

values lower than 0.5%. The term “sanitaryware” is the general name of the glazed and

non–glazed white products such as sinks, toilets and shower stalls. There are two different

type of slips used in ceramic sanitaryware production. Vitrified slip is generally used for

production of closet group products and the fireclay slips are used for the big size and flat

sinks in which the deformation is highly important. Pyroplastic deformation is the

situation that fireclay products undergo a shape change during firing. In this study,

enhancement of pyroplastic deformation in fireclay slips and improvement of product

quality is aimed. Hence, the recipes are reformulated by increasing the chamotte amount

by considering the control of rheological properties. Water absorption, firing deformation,

semi–product strength and shrinkage values of the samples were measured. Results show

that reformulated 3 new recipes has improved deformation proportionally depending on

the chamotte amount.

KEYWORDS - pyroplastic deformation, ffc, ceramics, chamotte



DESIGN AND IMPLEMENTATION OF 3D VISUAL LOCALIZATION SYSTEM

FOR QUADROTORS

TOLGA YUKSEL1


ABSTRACT

Although quadrotors can measure roll-pitch-yaw angle with on-board inertial

measurement unit, they need devices like GPS to define their locations according to

predefined reference world frame. GPS is a localization system that is available

everywhere in the world but its error for small scale unmanned air vehicles (UAVs) are far

beyond acceptable levels. Thus, a much more precise localization measurement and

feedback should be obtained for indoor applications and experimental setups of

quadrotors. Because of the reasons mentioned above, a visual localization system for

quadrotor type UAVs is designed and implemented in this study to provide tracking of

quadrotor motion in x-y-z space by processing stereo images taken from a 3D camera with

double lens. The system consists of three stages. As the first stage, the stereo camera is

calibrated and intrinsic parameters of both cameras are obtained to avoid the effects of

fisheye lenses. These parameters are used to rectify video frames obtained from stereo

camera. Then, three balls in different colors are placed on the quadrotor as 3D markers. In

the following, disparity map is generated using each pixel of each frame. This disparity

map provides depth information. Then, 3D reprojection is obtained using this map. Only

points of red-green-blue markers are taken into consideration in the obtained reprojection.

The center of the quadrotor is obtained using the centers of these markers, the center is

transformed from camera reference frame to world reference frame and 3D localization is

implemented. All the software for the proposed 3D localization system is implemented

using MATLAB Image Processing and Computer Vision Toolboxes. The experimental

results verify that the proposed system is capable of 3D localization of quadrotors for only

small 3D spaces.

KEYWORDS - quadrotor, 3D localization,stereo camera



A NEW BROADBAND REFLECTARRAY ANTENNA WITH MODIFIED

CIRCULAR PATCH

HANDE BODUR1 , SIBEL UNALDI2 , SIBEL CIMEN1 , GONCA CAKIR1

1 Kocaeli University, Turkey; 2 Bilecik Seyh Edebali University, Turkey

ABSTRACT

This paper is composed of a novel broadband single layer reflectarray antenna design for

X-Band applications. We obtained nearly 323 degrees reflection phase range with slow

linear phase variation by using variable size of modified circular patch. A 12 x 12

unitcells, 144 elements reflectarray covers a square area 120 x 120 mm2 is designed on

0.054λ0 thin dielectric substrate (ɛr=3, h=1.524) and simulated with CST Microwave

Solver. The pyramidal horn antenna is used for illuminating the each unitcell to get main

beam in a desired direction (ϴ=0). In order to enhance the gain bandwidth of this proposed

reflectarray further, variable patch sizes are used. The simulation results demonstrate that

max gain of reflectarray antenna is 19.467 dB and it has %17 1-dB gain bandwidth

(between 9.76 GHz-11.66 GHz) and %22 3-dB gain bandwidth (between 9.53 GHz-12

GHz).

KEYWORDS - Unit cell, broadband, single layer, reflectarray antenna, pyramidal horn

antenna



A LOW PROFILE MINIATURIZED BAND PASS FREQUENCY SELECTIVE

SURFACE WITH ANGULAR STABILITY

AYBIKE KOCAKAYA1, GONCA CAKIR1


ABSTRACT

— In this paper, a novel low profile miniaturized element band-pass frequency selective

surface (FSS) is proposed. The unit cell is constructed by rotated cross dipole aperatures

on the metallic surface and it is printed on single side of dielectric substrate. The FSS is

designed on the dielectric substrate with a relative permittivity of 6.15, loss tangent 0.002

and a thickness of 0,508 mm. The dimensions of the unit cell are 3.7×3.7 mm. The pass-

band is between 8.08GHz and 12.128 GHz under the -10 dB on the transmission curve.

Respect to 10.32 GHz center frequency, the unit cell dimension corresponds to

corresponds to 0.12λ_0where λ_0 is wavelength of free space. It is achieved a stable

transmission curve up to 45° incident angles both perpendicular (TE) and parallel (TM)

polarization. The simulations of FSS are implemented in CST MWS.

KEYWORDS - Frequency selective surface, Band-pass, Miniaturization, Angular

Stability, X band



DESIGN AND OPTIMIZATION OF FRACTIONAL SLOT PERMANENT

MAGNET ASSISTED SYNCHRONOUS RELUCTANCE MOTOR FOR

WASHING MACHINE

OMER FARUK PAYZA1, GAMZE ODABAS1, YASEMIN ERTUGRUL1

1 Arcelik A. S., Turkey

ABSTRACT

In general, universal motors or permanent magnet synchronous motors are used in

washing machines. In this study, different from conventional motor applications, the

fractional slot permanent magnet assisted synchronous reluctance motor is designed and

optimized for the washing machine applications. The wide speed range and the noise level

of the washing machine have to be considered as significant design criteria. The output

torque quality affects the noise level of permanent magnet assisted synchronous reluctance

motor. The parameters, which affect torque quality, are considered while designing

permanent magnet assisted synchronous reluctance motor. There are two main parameters

that affect the output torque quality. These parameters are torque ripple and cogging

torque. In this study, the optimization is made for reducing cogging torque and also torque

ripples. The optimized motor design is given and compared with initial design in this

study. The motors, used for washing machine application, must have the ability to operate

at wide speed range. The designed permanent magnet assisted synchronous reluctance

motor has the ability to operate at a wide speed range between 1000-15000 rpm. The

motor is designed by using the Flux 2D FEA software. The designed permanent magnet

assisted synchronous reluctance motor is analyzed at both no-load and under-load

conditions using Flux 2D. Optimization studies and results are given for synchronous

reluctance motor operate at wide speed ranges with low torque ripple.

KEYWORDS - Fractional-slot machine, permanent magnet (PM) motor, PM assisted

synchronous reluctance motor, torque ripple. washing machine,FEA



POSITION CONTROL OF A BALL BEAM EXPERIMENTAL SETUP BASED ON

SLIDING MODE CONTROLLER

KAAN CAN1, ABDULLAH BASCI1

1 Ataturk University, Turkey

ABSTRACT

In this paper, a sliding mode control (SMC) method is introduced to design a control

methodology for the ball and beam experimental setup (BBS) that consists of a servo

motor, beam and ball. The proposed control method is realized in two cascaded control

structures such that primary and secondary, respectively. In the primary part, called outer

loop, the position of the ball is controlled by changing the angle of the beam. In the

secondary part, called inner loop, the needed voltage is generated to determine appropriate

position angle of the servo motor to adjust the position of the beam. Furthermore, a well-

tuned conventional PI controller is also applied to the system to indicate the priority and

effectiveness of the SMC. The results obtained in real-time show that SMC is better than

PI controller in the aspect of reference tracking, fast response to the changes and accuracy

as well.

KEYWORDS - Nonlinear control, sliding mode control, ball and beam system



COMPARISON OF FRACTIONAL PI AND CLASSICAL PI CONTROLLER

BASED POSITION CONTROL OF BALL AND BEAM SYSTEM

KAAN CAN1, ABDULLAH BASCI1, BIROL SOYSAL1, ADNAN DERDIYOK2

1 Ataturk University, Turkey; 2 Sakarya University, Turkey

ABSTRACT

In this paper, fractional-order PI (FOPI) controller is used to control the ball and beam

laboratory setup (BBS). The system consists of a servo motor, beam and ball. The position

of the ball is controlled via beam’s angle. Also, to adjust beam’s position, servo motor

must be controlled through controller to generate needed angle for the beam. The aim of

the proposed controller is to track reference trajectory very well, deal with parameter

uncertainties and provide more flexibility for the nonlinear BBS. The experimental results

show that the FOPI controller gives fast response, has lower rise time, settling time and the

successes of reference position tracking when compared with classical integer-order PI

controller.

KEYWORDS - Fractional PI, ball and beam system, Nonlinear control



MEANWHILE ANATOMY OF GASES WITH STUBBORN GAS SENSORS TO

INCREASE THE DETECTION EFFICIENCY

ALI MARDAN HAMEED QUTUB1, ISMAIL SARITAS1


ABSTRACT

The purpose of this material is to monitor the leakage of liquid natural gas (LPG) to

prevent incidents of fire that are an important issue. The system detects the LPG leak with

gas to report the gas leak. The system warns the system immediately if it detects that the

LPG Concentration in the air exceed a certain level.

KEYWORDS - LPG, safety feature, gas detection and prevention



WIND GENERATION ESTIMATION BASED ON WEATHER FORECASTING IN

KIRIKKALE TURKEY

MUSTAFA YASIN ERTEN1, HUSEYIN AYDILEK1, NIHAT INANC1


ABSTRACT

Increasing world population and with the transformation of human power employment

into machine power increases the energy need rapidly. Given the limited availability of

fossil-based energy sources and the harmful effects it has on the environment, wind power

is likely to be the most important source of clean energy in the near future for both Turkey

and the world. However, estimating how much power wind turbines generate is a

challenging and difficult problem due to the dependence of the wind power on

meteorological events. The increase in the share of wind turbines in the electricity grid and

the intermittent and stochastic nature of the wind power make it difficult to integrate into

the electricity grid. Estimating the speed of the wind has a vital role in terms of ensuring

the safety of wind turbines and avoiding power quality and voltage surge problems is

critical to the grid. In this study, the wind speed was estimated by using ARIMA method

and the wind speed, wind direction and wind intensity parameters were used while the

wind speed was estimated. In addition, the maximum speed used in the method is

calculated by using the extreme value theorem. The estimation of wind power has been

experimented on the data obtained through the turbine and air station located in Kirikkale

University campus area and successful results have been obtained.

KEYWORDS - Wind Speed Forecasting, ARIMA, Extreme Value Theorem, Gumbel

Distribution, Wind Power Estimation



THE INTERACTION OF THEORETICAL COURSES WITH PRACTICALS AND

THE EFFECTS ON SUCCESS CASE OF AN ELECTRONIC LECTURE

PIRIL TEKIN1, RIZVAN EROL1

1 Cukurova University, Turkey

ABSTRACT

Electrical and electronic engineers are trained for the purpose of acquiring more

information on this energy used so commonly today and ensuring performance of all tasks

in the process by the hands of experts. The training of this department seeks to make its

students graduate with all necessary information starting from fundamentals of electricity,

to the use and operation of machinery and equipment’s. That is why the trainings and the

courses given theoretically and practically in the department are of great importance. The

students are often asked to transform the theoretical information obtained into practice in

the laboratories of the courses. Thus, a small working environment is created for the

students who are expected to simulate the problems they may encounter in real life and the

related operations/mechanism to the class environment. The electric machinery training

sets have modular design so that the students can learn and use basic electricity laws with

their measurements, learn about, design, and analyse the circuit elements through hands-on

training. Modular design of the sets made the laboratory environment resemble an

industrial implementation setting. The motors and remote control switches used in the

training sets have the same design with the models used in the industrial and daily life. In

this paper, one of the sample of electrical and electronic engineering department course

covers also many laboratory courses is analysed. Each of courses which entail different

application models and process are examined for students’ and universities point of view.

In addition, the effects of theoretical courses on practical lessons are explained by using

students’ success rate.

KEYWORDS - System Modelling, Work Flow Diagram, Electrical and Electronic

Engineering, Performance Measurement, System Analyzing



IMAGE PROCESSING BASED WEATHER PREDICTION AND OUTDOOR

ACTIVITY DECISION FRAMEWORK WITH SUPPORT VECTOR MACHINES

HUSEYIN AYDILEK1, MUSTAFA YASIN ERTEN1, NIHAT INANC1


ABSTRACT

Weather conditions have an important effect on our daily life. Even choosing clothes,

travel planning, and outdoor activities; weather forecast gives valuable information to us.

According to weather information, we can prevent many unwanted situations and plan to

our day safely. Nowadays, there are many ways to predict weather forecast. In this work, it

is developed a framework to predict weather forecast from outdoor scenes and decide

availability to outdoor activity. To analyze the weather conditions availability, a machine

learning algorithm Support Vector Machines (SVM) used as a classifier, Histogram of

Gradients (HoG) and Locally Binary Patterns (LBP) approaches are used as a descriptor

and successful results have been obtained.

KEYWORDS - Weather Prediction, Support Vector Machines, Histogram of Gradients,

Locally Binary Patterns, Image Processing



EFFECTS OF INVERTER TOPOLOGIES AND MODULATION TECHNIQUES

ON POWER QUALITY ISSUES ON GRID CONNECTED PHOTOVOLTAIC

SYSTEMS

MEHMET CAN ALPHAN1, GURKAN SOYKAN1, LUTFI ARDA1


ABSTRACT

The usage of renewable energy resources on the power grid is increasing day by day due

to both environmental constraints and limited fossil fuels resources. The grid-connected

inverters are essential part of the renewable energy generation systems especially for a

photovoltaic power system. The photovoltaic system generates the power as DC output. In

order to supply this power to the AC power grid, the DC-AC inverter is inevitable

component in the system. Increasing the utilization of renewable energy generation

systems due to provide increasing load consumptions causes different power quality

problems at the grid connection point because of the intermittent nature of renewable

energy sources. The main power quality problems on grid-connected photovoltaic systems

are harmonics and voltage variations. In this study, effects of inverter topologies and

modulation techniques in terms of power quality issues are investigated on Matlab -

Simulink environment. The comprehensive analysis on the grid-connected photovoltaic

systems is achieved considering not only power quality issues but also efficiency of

inverters. Particularly, different inverter topologies have been analyzed and those

topologies have been tested with sinusoidal pulse width modulation and vector space

modulation techniques in terms of total harmonic distortion, voltage changes and

efficiency of inverters. Depending on the analysis results, the effects of inverter topologies

and modulation techniques on power quality are presented in the grid-connected

photovoltaic system.

KEYWORDS - power quality, Inverter Topologies, Modulation Techniques, Photovoltaic

cells



ECONOMIC EVALUATION OF SOLAR PHOTOVOLTAIC SYSTEMS FOR A

SMART HOME

AHMET NUR1, ABDULCELIL BUGUTEKIN1


ABSTRACT

Solar energy is the technology that finds the greatest development and application area

among the renewable energy technologies. There has been a significant increase in the use

of solar energy in recent years due to the decrease in the prices of photovoltaic panels.

Various financial incentives, including tax and production incentives are given to ensure

widespread use in many countries, especially in the European Union. Photovoltaic systems

can be used either alone or depending on the electrical grid. In this study, informations

about the components that should be used in the photovoltaic systems in a smart home

electricity are given and economic analyses are made.

KEYWORDS - Economic Evaluation, Photovoltaic Systems, Smart Home



HISTOGRAM BASED VEHICLE LICENSE PLATE RECOGNITION WITH KNN

METHOD

SECKIN KARASU1, AYTAC ALTAN1, ZEHRA SARAC1, RIFAT HACIOGLU1


ABSTRACT

In this study, 35 different plate characters obtained by character segmentation algorithm

are classified by k Nearest Neighbor (KNN) method. The characters are separated from the

background of the scene by a threshold method on the gray scale image. The histogram

components of the obtained characters are examined and the features representing the

characters are calculated. The performances of classification methods for different

classifier parameters at different noise levels (20 dB, 10 dB and 5 dB) are examined. In

this study, it is seen that the overall performance is 96.67% and the performance in the

noisy environment (5 dB) is 93.81%. The performance of the KNN classifier method

under noise is investigated and compared with the vehicle license plate recognition

applications in the literature.

KEYWORDS - Image Processing, License Plate Recognition, Histogram Based, KNN

Method,Machine Learning



ESTIMATION OF FAST VARIED WIND SPEED BASED ON NARX NEURAL

NETWORK BY USING CURVE FITTING

SECKIN KARASU1, AYTAC ALTAN1, ZEHRA SARAC1, RIFAT HACIOGLU1


ABSTRACT

In this study, a Nonlinear AutoRegressive eXogenous (NARX) neural network is used to

estimate the wind speed on three monthly data sets taken from the wind central in

Zonguldak province in Turkey. In the estimation study, the first and second order curve

fitting coefficients of the measured temperature, pressure, humidity and solar radiation

parameters together with the wind speed are used. In the estimation process, before these

coefficients are applied directly to the NARX network structure, the most suitable features

are selected with ReliefF method to minimize the MSE value. The number of delay steps

in the NARX network structure is varied from 3 to 15 and the number of hidden neurons is

varied from 3 to 15 to obtain model parameters that give the least estimation error. In

order to determine the performance of the obtained model, the model is evaluated in terms

of statistical error criteria such as MAE, MSE and RMSE. The model parameters and

features matrix giving the least estimation error for the wind speed estimation of the

NARX network structure are determined. It has been observed that this approach provides

a high performance for estimating the wind speed with related to other measured

parameters.

KEYWORDS - Wind Speed , Estimation, NARX, Curve Fitting, ReliefF Method



SIMULATION OF BROKEN ROTOR BAR FAULT IN INDUCTION MOTORS

CEMIL KOZKURT1, MEHMET AKAR1

1 Gaziosmanpasa University, Turkey

ABSTRACT

Induction motors are workhorses of the industry, and it is important to detect the faults at

the beginning stage. In this proposed study a simulation software has been developed in

order for the students to better understand the rotor faults in the induction motors. Within

this interface it is possible to analyze the current/speed/electromagnetic torque signals by

operating the motor at different speeds and to detect the faults with the help of the

frequency transforms of those signals.

KEYWORDS - Induction motors, rotor fault, computer aided educational tool



RELIABILITY ANALYSIS OF ATAKUM SAMSUN DISTRUBUTION SYSTEM

BY USING MONTE CARLO SIMULATION

FATIH DURMUS1, CAGRI KOCAMAN1


ABSTRACT

Reliability in energy systems is necessary to minimize the impact of system failures on

consumers. Therefore the analysis of the reliability of an energy system is very important.

In order to be able to perform reliability analysis, it is necessary to be known frequency of

faults (λ), mean time between failures (m) and average failure elimination time (r) in the

previous working period of the system. These data can be obtained on a yearly basis by the

compilation of the fault data. Reliability analysis can be done by using the analytical

method. So mathematical model is used. However, it is difficult to solve the reliability

problems of great and complex systems analytically. The Monte Carlo Simulation, written

in Matlab programming language, makes it possible to analyze great and complex systems

easily. Internationally accepted reliability indexes (SAIDI, SAIFI, CAIDI, ASAI) are

obtained as the output data of the program. The frequency of faults (λ), mean time between

failures (m), average failure elimination time (r) and information such as consumer

number, line length, transformer number and drawed power are inputs. When the output

datas compared with the internationally accepted values, information about the reliability

of the system is obtained. In this paper, reliability analysis is performed by using Monte

Carlo Simulation by utilizing real time data in Atakum Samsun Distribution System.

SAIDI, SAIFI, CAIDI, ASAI reliability indexes are obtained and they are compared with

internationally accepted values. It is seen that this system is safe and sustainable.

KEYWORDS - Reliability, Monte Carlo Simulation, CAIDI, SAIDI, ASAI, SAIFI,

Distribution Systems



DESIGN OF A 400 HZ CURRENT SOURCE 1 PH INVERTER TOPOLOGY FOR

AVIONIC SYSTEMS

ERALP SENER1, GURHAN ERTASGIN1


ABSTRACT

Inverters are widely used in aviation industry. There are ground type inverters and plane

type inverters for supplying the avionics. The inverters which are used in the planes are

called static inverters and they work at a 400hz frequency. 400hz frequency is used in aero

planes basically for reducing the size of the electronic elements. Also using a higher

frequency may require thicker and naturally heavier wiring. So 400hz is the general

frequency used in planes. There are negative effects of using high frequencies for energy

transmission The disadvantage of working at 400 Hz is that high frequency systems will

probably endure voltage drops. The most critical of these losses comes about because of

reactive losses. Reactive drops are brought on by the inductive properties of the leading

links or wires through which the electrical current is transmitted. This kind of drawback is

influenced both by the length of the conductor and in addition the frequency of the

influence coursing through it. As frequency expands, the bigger the voltage drop gets to

be. At a high frequency of 400 Hz, reactive drops can be as much as seven circumstances

bigger than at a low frequency of 60 Hz. Also reactive loads may cause PF problems for

the inverter used. PF problems can be solved easily by using a current source inverter.

Current source inverters are not widely used because of their bulky DC Link inductor.

However, the DC link inductor can be minimized thanks to the 400Hz system. Without the

bulky inductor, Current Source Inverter can be very small, stable, redundant and efficient

for the avionics. In this paper, a 400hz current source inverter with a airplane battery input

and 115V output will be simulated and presented.

KEYWORDS - 400hz, current source inverter, avionics



DECOMPOSING LIGHTING COMPONENT OF AN IMAGE A REVIEW

UMIT CIGDEM TURHAL1, YAVUZ BUYUKKOCAK1


ABSTRACT

In machine vision and image processing applications, illumination differences cause lots

of problemlems that are crucial for many visual tasks. This is because, different

illuminations cause very large changes in the image. Therefore, in many applications,

illumination differences are tried to be removed. There can be found many studies, related

to removing the illumination component by decomposing it from image in the literature.

Some of these studies are related to image segmentation, extracting accurate colour of

objects, object colour changing, image watermarking, illumination transferring between

images, image enhancement, intrinsic image decomposition, illumination normalization

and etc. In this paper we tried to gather current studies relevant to decompose illumination

component of an image. Thus researchers are given the opportunity to compare and see the

differences between different methods.

KEYWORDS - Image processing, machine vision, image decomposition, illumination

component



DESIGN AND ACTUALISATION OF EOG BASED INTERACTIVE

COMMUNICATIONS FOR ALS PATIENTS

UGUR FIDAN1 , NESE OZKAN1 , ALI ISIK1

1 Biomedical Engineering, Turkey

ABSTRACT

Amyotrophic Lateral Sclerosis (ALS) is a motor neuron disease caused by loss of function

of the spinal cord and brain stem nerve cells. The loss of function in the nerve cells leads

to weakness and eruption (atrophy) in the muscles. This situation eliminates the movement

functions so patients need the assistance of a companion. Although there are limitations of

the movement, the consciousness levels of ALS patients are not different from healthy

individuals, and studies are carried out on the methods that will increase the daily life

quality of ALS patients. In this study, Ag / AgCl electrodes were placed around the

orbicularis oculi muscles. Electrooculogram (EOG) signals were taken from an

instrumentation amplifier with 48dB gain and 107dB CMMR ratio. A 16Hz Low Pass

Filter and a 50Hz Notch Filter were used to increase the signal to noise ratio. The EOG

signals are digitized with 10bit resolution ADC and applied to ATmega328

microcontroller. So, the software of microcontroller determines the horizontal and vertical

movements of the eye. The movement information of the eye was sent to the computer

with NRF24L01 transceiver at 2.4GHz operating frequency and 9.6Kb/s data transmission

speed. In this way, the interactive PC software was controlled by the EOG signals. An

EOG-based interactive software developed using the C # programming language has

provided patients' daily requirements, and social media accounts, etc. The system has been

tested on healthy subjects and it has been seen that people can control the software by eye

movements. As a result, system will be useful not only for ALS patients, but also for

permanent or partially bedridden patient (MS, Hemiplegia, etc.) groups. Moreover, while

the quality of life in the patients' own environment is being raised, they will also be able to

benefit from health services within the scope of home health services.

KEYWORDS - Electrooculogram (EOG), Amyotrophic Lateral Sclerosis (ALS), Atrophy,

İnteractive, Home Health Care



EFFECT OF THICKNESS ON THE STRUCTURAL AND OPTICAL

PROPERTIES OF SILAR DEPOSITED CUS THIN FILMS

TUBA CAYIR1, YUNUS AKALTUN2

1 Koaceli University, Turkey; 2 Erzincan University, Turkey

ABSTRACT

Cupper sulfide (CuS) thin films of different thickness were prepared on glass substrates by

successive ionic layer adsorption and reaction (SILAR) method at room temperature. CuS

thin films were deposited 30, 40, 50 and 60 cycles respectively. The effect of film

thickness on characteristic parameters such as the structural, morphological and optical

properties of the films was investigated. The X-ray diffraction (XRD) and scanning

electron microscopy (SEM) studies showed that all the films exhibit polycrystalline

structure are covered well with glass substrates. The crystalline and surface properties of

the films improved with increasing film thickness. The energy band gap values were

decreased from 1.92 to 2.06 eV depending on the film thickness.

KEYWORDS - Thin Films, CuS, SILAR, XRD, SEM, AFM, RAMAN.



BARRIER FREE DOOR FOR INDIVIDUALS WHO HAVE A PHYSICAL AND

VISUAL INSUFFICIENCY

FECIR DURAN1 , TURKAN TEKE2 , MUSTAFA TEKE3 , MEHMET GUNES3

1 Gazi University, Turkey; 2 Guidance Research Center, Turkey; 3 Cankiri Karatekin

University, Turkey

ABSTRACT

As in the past, there are a large number of disabled individuals on Earth now. The numbers

of individuals with disabilities is increasing every day due to several reasons. To simplify

the lives of individuals with disabilities, android-based Smartphone software has been

implemented to make it easier to open and close locks on the entrance doors in homes and

work areas where they live. The door control circuit which controls the door lock is also

designed to be accessibly by telephone. Android software communicates with the control

circuit on the door by the help of bluetooth. The MAC address of the mobile phone and the

MAC address of the bluetooth module on the door control circuit are matched to provide a

secure entry. After the initial installation of the software and the match is completed, the

user will be able to open the mobile phone and control the door lock by pressing the button

on the secreen. The barrier-free door entry system has been implemented for individuals

with disabilities who are unable to perform precise motoring operations that are difficult to

handle for unlocking and locking the door lock with the key.

KEYWORDS - disabled, Embedded system, Arduino Board, Bluetooth, Control



DESIGN OF DIFFERENTIAL RELAY APPLICATION WITH MATLAB

SIMULINK FOR PROTECTION OF TRANSFORMER IN ELECTRIC ARC

FURNACE EAF LOADS

MUSTAFA SEKER1, MURAT CIKAN1, ARIF MEMMEDOV2, SERDAR

KOCKANAT1, HULYA DOGAN1

1 Cumhuriyet University, Turkey; 2 Inonu University, Turkey

ABSTRACT

EAFs are nonlinear loads with the highest power consumption in electrical systems. In

addition to high power consumption, due to the rapidly changing nonlinear characteristic

of the EAF loads in the transmission line supply transformers, step-down transformers are

necessary to design them in accordance with the pulse operating conditions. Although

special designs are made for these transformers, these designs are sometimes insufficient

to protect the transformers. The three-phases system model of the transmission line

feeding the Sivas Iron and Steel Works CO. (SIDEMIR) plant was modeled with the help

of Matlab / Simulink. In this model, electrical arc has been determined as mathematically

by used the exponential-hyperbolic model and real time measurement data with HIOKI

3197 PQ analyzer. Because of the failure occurring in 2012, Step-Down transformer has

failed and becomes unusable. For this reason, Transformer protection system have been

developed for step-down transformer in case of failure with designed differential relay in

three phases simulation model for the transformer is not damaged in case of a similar fault

and the same problems are not experienced in similar plants to prevent damage to the

transformer in case of a similar fault. In the studies presented in the literature, differential

relays have not been evaluation of their performance in high-power nonlinear loads such

as arc furnaces. With these simulation model, the performance of the differential relay

designed to protect the transformer was evaluated by examining the variations in one

phase ground fault, phase-phase fault and three phase ground fault. Simulation results are

shown that the applied differential relay can be used to protect the Step-Down transformer

feeding the EAF loads. In addition, with this designed differential relay, the tripping times

in case of failure can also be adjusted and the tripping times can be lowered to less than 10

ms.

KEYWORDS - Electrical Arc Furnace(EAF), Nonlinear Loads, Transformer Protection,

Differential Relay.



SYNCHRONIZATION OF THE CHAOTIC SYSTEM WITH GOLDEN

PROPORTION EQUILIBRIA ACTIVE CONTROL VERSUS SLIDING MODE

CONTROL

YILMAZ UYAROGLU1, UGUR ERKIN KOCAMAZ2

1 Sakarya University, Turkey; 2 Uludag University, Turkey

ABSTRACT

In this study, the synchronization of the chaotic system with golden proportion equilibria

is investigated by using active control and sliding mode control methods. A Lyapunov

function is used for achieving the active controllers ensure the global asymptotic stability

of the error system. An appropriate sliding surface is designed on the basis of the sliding

mode control theory. Numerical simulations are demonstrated to present the

synchronization performances. In all cases the synchronization is obtained in a proper

amount of time. However, the active control approach gives more efficient results.

KEYWORDS - Chaos synchronization, Golden proportion equilibria, Active control,

Sliding mode control



LYAPUNOV BASED NONLINEAR CONTROL OF THE LOTKA VOLTERRA

SYSTEM WITH FOUR COMPETITIVE SPECIES

GULTEKIN CAGIL1, UGUR ERKIN KOCAMAZ2, YILMAZ UYAROGLU1

1 Sakarya University, Turkey; 2 Uludag University, Turkey

ABSTRACT

In this paper, the control of four-dimensional predator-prey Lotka–Volterra system is

investigated. The Lotka–Volterra system with four competitive species is briefly

described, its equations are presented and hyperchaotic figures are shown. Based on the

Lyapunov stability theory, the appropriate nonlinear control signals are obtained for the

control. Simulation results are demonstrated to validate the feasibility of the proposed

control technique. Numerical simulations also show that the controlled Lotka–Volterra

system effectively stabilizes towards its equilibrium points in the state space.

KEYWORDS - Lotka–Volterra system, Four competitive species, Predator-prey model,

Nonlinear control, Hyperchaos control



MULTIPLE TARGET TRACKING SIMULATION AT AN INTERSECTION

AYKUT GORKEM GELEN1, AYTEN ATASOY1


ABSTRACT

In this study, multiple target tracking problem is investigated for vehicles at an

intersection. Collecting data for this problem is expensive and time consuming. At this

point, simulation is a very powerful tool. The target data can be generated by means of

simulation with system models under some assumptions. By destroying some

measurements as missing measurements and adding modeled extra clutter measurements,

the measurement data can be obtained. Then, a feasible tracker can be developed for this

problem. This tracker will probably work smoothly with real data. In this paper, the aim

was to study maneuvering target tracking by simulated motion data of multiple vehicles on

an intersection. Different situations such as straight acceleration, turning and waiting at the

traffic lights have been considered. Simulated vehicles are tracked by a tracking algorithm

that combines Interacting Multiple Model for maneuvers and Global Nearest Neighbor

Filter for data association. A tracker is successfully developed for this problem by using

those algorithms.

KEYWORDS - Multiple Target Tracking, Traffic Tracking, Interacting Multiple Model,

Global Nearest Neighbor Filter, Intersection Tracking



APPLICATION OF MULTI AGENT SYSTEMS IN MICROGRID CONTROL

ZIYODULLA YUSUPOV1, ANIS AMHARIB1, NURI ALMARGANI1

1 Karabuk University, Turkey

ABSTRACT

Last decades with the rapidly penetration of distributed energy resources to the power

system, the interest on microgrid is growing. Microgrids as a key component of smart grid

are intended to improve energy efficiency, reliability of power system and decrease carbon

dioxide emissions. In recent year multi-agent systems as a branch of distributed artificial

intelligence method have been employed in the several areas of power systems, such as

electric grid control and protection, monitoring and optimization of distributed systems,

power system restoration. In this paper considered application of multi-agent systems in

microgrid control.

KEYWORDS -



DEMAND SIDE MANAGEMENT APPLICATION FOR THE KIRIKKALE

UNIVERSITY TURKEY

ERTUGRUL CAM1, VOLKAN ATES1, AYSEGUL AKSAKAL1, YAGMUR ARIKAN1


ABSTRACT

Today, as energy generation has become important as well as the consumption of energy,

many studies have been conducted on consumer habits. One of these studies is Demand

Side Management (DSM). First, energy consumption of a load center has been monitored

by DSM equipment and software, various decisions are made about consumption hours

and / or consumption devices. In this study, consumption curves were obtained from 3

different transformers which have energy analysers in Kırıkkale University Engineering

Faculty. The curves were entered to the Fuzzy Logic (FL) system which was realized on

MATLAB for meaning and clustering. Outputs of the FL system was then entered the

model of Artificial Neural Networks, realized in MATLAB, to analyse the state of energy

consumption of the faculty. Thus, the relationship between the energy consumption of the

faculty and the curriculum was analysed and suggestions were made to reduce or control

of consumption. The results show that the system's success is more than 70% and the

energy consumption of the faculty was also reduced with the proposed DSM system.

KEYWORDS - Demand side management, Fuzzy Logic, Artificial Neural Networks,

Energy saving.



INVESTIGATION OF CURRENT AND VOLTAGE FLUCTUATIONS IN

ELECTRIC ARC FURNACES WITH THE HELP OF PROBABILITY THEORY

MUSTAFA SEKER1, ARIF MEMMEDOV2, MURAT CIKAN1, HULYA

DOGAN1, SERDAR KOCKANAT1

1 Cumhuriyet University, Turkey; 2 Inonu University, Turkey

ABSTRACT

Electric Arc Furnaces (EAF) are non-linear loads which have the greatest power, cause

electric power quality problems in electrical networks. Due to the nonlinear nature of the

electric arc furnace, they are characterized by rapid changes in the current and owing to

these rapid changes, the voltage and frequency are also variable. Determination of the

Mathematical modeling of this variability is very difficult. Analysis of stochastic processes

with probability methods are allowed that essential information to be indicate with a small

number of characteristics separately about physical phenomena using experimental data

without considering their physical process. These characteristics have been arbitrary load

variation and allow analysis of EAF loads. In this study, Power quality problems of

Electrical Arc Furnace power which have 60 MVA power at the Sivas Iron and Steel

Company (SIDEMIR) facility were recorded with the help of the 8332B power quality

analyzer. Using obtained experimental data, the propagation of current fluctuations is

determined by the probability theory with the determination of the statistical and

correlation function of current and voltage in the melting processes of electric arc

furnaces. The results show that determination dispersion of current fluctuations of electric

arc furnaces for whole boundary value limits in EAF which gets its maximum value of

current,heat and efficiency will provide when its optimized value are chosen.Probability

results are shown that nominal high voltage short circuit current can reach 2.12 times the

nominal current. This rate must be taken into account for transformer protection.

KEYWORDS - Probability Theory, Power Quality, Electric arc furnace, current and

voltage fluctuations



SERIAL IN PARALLEL OUT DISCRETE FOURIER TRANSFORM

EROL SEKE1, ZEYNEP KAYA2

1 Eskisehir Osmangazi University, Turkey; 2 Bilecik Seyh Edebali University, Turkey

ABSTRACT

Discrete Fourier Transform (DFT) and its fast implementations (FFT) have wide use in

signal processing applications. As the requirements of the applications change, DFT

implementations evolved accordingly to meet them. This paper presents a serial-data-in-

parallel-coefficients-out approach where input data is a stream and transform results are

generated all in parallel fashion. Such an implementation may find use in spectrum

analysis applications. One may argue that employing a serial-to-parallel converter

structure at the input of a parallel-to-parallel algorithm would solve the problem with a

small price. On the other hand, when the data is presented in a stream, FFT building-

blocks (butterflies) become redundant and their utilization is very low. There exists a

number of serial-to-serial (pipelining) FFT implementations with high hardware

utilization, but they output transform data in serial form. We solved serial-to-parallel

problem within FFT structure by eliminating redundant butterflies and switching

intermediate data to registers, generating output data in parallel with some latency which is

lower than that of pipelining implementations.

KEYWORDS - DFT, FFT, Fourier Transform, Embedded Processing, pipeline FFT



DESIGN OF X BAND SINGLE LAYER REFLECTARRAY WITH NOVEL

ELEMENT

SIBEL UNALDI1, HANDE BODUR2, SIBEL CIMEN2, GONCA CAKIR2

1 Bilecik Seyh Edebali University, Turkey; 2 Kocaeli University, Turkey

ABSTRACT

This paper presents a novel single-layer reflectarray with rotational element. A X band

reflectarray designed and simulated by using CST Microwave Solver. The designed unit

cell structure of the reflectarray provides more than 420 degrees reflection phase range. A

9 x 9 element reflectarray antenna covers 126 x 126 mm2 square plane. Linearly polarized

pyramidal horn antenna is used for centre-fed configuration. The simulation results show

that 3-dB gain bandwidth of 27 % and 1.5 - dB gain bandwidth of 16 % and 18.7 dB max

gain level are achieved at X band.

KEYWORDS - Reflectarray, X band, wide phase range, microstrip, single layer



TIME RESOLVED FLUORESCENCE TECHNIQUE FOR STUDYING

SWELLING PROCESS IN HYDROGELS

DEMET KAYA AKTAS1, FILIZ OZTEKIN1


ABSTRACT

Fast Transient fluorescence (FTRF) spectrometer was employed during the swelling of

polyacrylamide (PAAm) hydrogels in water. Disc-shaped acrylamide gels were prepared

from acrylamide (AAm) with various N, N’ methylenebis (acrylamide) (Bis) contents as

crosslinker in the presence of ammonium persulfate (APS) as an initiator by free-radical

crosslinking copolymerization (FCC) in water. Polyacrylamide (PAAm) gels dried before

using for swelling experiments. Pyrene was introduced as a fluorescence probe.

Fluorescence lifetime of pyrene from various BIS content gel samples was measured

during in situ swelling process. It was observed that fluorescence lifetime decreased as

swelling has proceeded. Gravimetrical and volumetrical experiments were also performed.

Li-Tanaka equation was used to determine the swelling time constants, T and cooperative

diffusion coefficients, Dc from intensity, weight and volume variations during the swelling

processes. It was observed that swelling time constants, T increased and diffusion

coefficients, Dc decreased as the crosslinker content was increased. We also developed a

computer programme for the calculation of cooperative diffusion coefficient, Dc

automatically for any choosen BIS content, based on the experimental outcomes.

KEYWORDS - hydrogels, swelling, fluorescence, diffusion



BACKGROUND SUBTRACTION TECHNIQUES TO DETECT HUMAN VITAL

SIGNS USING UWB GPR

CANSU BUYUKHAN1, SAEID KARAMZADEH2, ALPER CALISKAN3, MELEK

ORHAN1, INCI CILESIZ1

1 Istanbul Technical University, Turkey; 2 Istanbul Aydin University, Turkey; 3 Yildiz

Technical University, Turkey

ABSTRACT

In this work, Ultra-wide band Ground Penetrating Radar system is purposed for detection

of human respiratory movement through obstacle. As background subtraction methods,

Principal Component Analysis, Singular Value Decomposition and Frame Differencing

Techniques are applied for reducing clutter effect. After background subtraction, human

breathing is detected by using Fast Fourier Transform and results of all steps will be

presented.

KEYWORDS - Ground Penetrating Radar; Background Subtraction Tecniques; Ultra-wide

Band; radar signal processing



DETECTION OF LESION BOUNDARIES IN BREAST MAGNETIC

RESONANCE IMAGING WITH OTSU THRESHOLDING AND FUZZY C

MEANS

GOKCEN CETINEL1, SEVDA GUL1

1 Sakarya University, Turkey

ABSTRACT

Abstract— Today, breast cancer is among the leading causes of mortality for women. So,

there is a great number of investigations about screening, diagnosing and treating the

breast cancer. Among these studies, determining the boundaries of lesion is of great

importance for radiologists. Automatic detection of small lesions that can’t be detected

with the naked eye and marking the borders of lesions provide supportive information to

the radiologists in the diagnosis and treatment process of cancer. In this study, it is aimed

to determine the boundaries of lesions using Otsu thresholding and Fuzzy c-means (FCM)

methods in breast magnetic resonance (MR) images taken after contrast agent uptake. As a

measure of success, the overlap between the lesion boundaries obtained by manual

segmentation and automatic segmentation is used. In the first experiment, % 87 success

was achieved by applying the Otsu thresholding method. Furthermore, in the second

experiment FCM method which is widely used for similar purposes in the literature, is

utilized. The performance measure for this method is obtained as % 91. According to the

comparative simulation results the performance attained by adjusting the number of

clusters in the FCM method are very close to the results obtained by manual segmentation.

KEYWORDS - Keywords— Lesion Segmentation, Breast MRI, Otsu thresholding, FCM



PERFORMANCE COMPARISON OF REGULAR AND IRREGULAR PILOT

PLACEMENTS IN OFDM SYTEMS

CETIN KURNAZ1, BEGUM KORUNUR ENGIZ1, FATIH AYVENLI1


ABSTRACT

In this study, the effect of the positions of the pilots on the Bit Error Rate (BER)

performance is investigated for ITU Indoor Office, Pedestrian-A, and Vehicular-A

channels. BPSK modulation scheme, 1024 subcarriers, 512 OFDM symbols, Nc/4 cycling

extended guard interval, 100 MHz transmission bandwidth were used in simulations. The

pilot separation was varied from 4th, 8th, 16th for regular placement, while the number

and the positions of pilots determined according to local maxima and minima of the

channels’ transfer function for irregular placement. Channels state information was

obtained through linear regression for all placements. The loss in Signal to Noise ratio

(SNR) due to pilot symbol usage was considered while assessing the BER performances. It

is seen from the results that, irregular pilot placement shows similar or better performance

with regular placement considering both BER and bandwidth efficiency. With the use of

irregular pilot placement bandwidth efficiency can be improved up to 99%.

KEYWORDS - OFDM; Channel estimation; Regular/Irregular pilot placement; Frequency

selective channel.



AN AUTOMATIC SYSTEM FOR SEGMENTATION OF AGGREGATE AND

BITUMEN IN ASPHALT CONCRETE IMAGES

ONDER CIVELEK1, MEHMET OZTURK1


ABSTRACT

Along with the rapid development of computers today, digital image analysis has begun to

be widely used in many areas. Difficult and time-consuming tasks for people can now be

accomplished quickly and precisely with digital image processing techniques and

computer vision applications. In this study, a system was developed to classify structures

such as bitumen and aggregate in asphalt concrete materials with the help of digital image

processing techniques. First, images of asphalt material samples were obtained under

controlled lighting conditions with a standard RGB camera. Imaging process was realized

in a closed environment that blocks the ambient light and has its own lighting system. The

RGB images are converted to grayscale with a set of special conversion coefficients. The

coefficients required for this transformation are calculated with utilizing the Particle

Swarm optimization algorithm under a condition that the average contrast level between

bitumen and aggregate is maximized. Then, the aggregate and the bitumen are segmented

by a simple thresholding algorithm from the gray level images obtained as a result of this

conversion. The threshold level required for the segmentation is calculated by the Otsu

algorithm. Obtained classification results are encouraging such that the accuracy,

sensitivity, and specificity values are over 90%, 70%, and 90%, respectively.

KEYWORDS - Aggregate, Bitumen, Segmentation, PSO, Otsu, Thresholding



A WIRED SENSOR NETWORK FOR THE PURPOSE OF MULTI DETECTION

SYSTEMS

AYHAN YAZGAN1, OGUZHAN CAKIR1, HUSEYIN ATASOY1


ABSTRACT

In this study, a wired sensor network and a current based coding algorithm is presented.

Proposed structure composed of fire and gas sensor boards with a base controller who

address and read all sensors in a very short time. Every sensor board has a predefined

address and therefore in case of a fire or gas leak, proposed system reacts automatically

and sends the necessary signals to both near the operation area and remote unit by the base

station.

KEYWORDS - fire sensor, gas sensor, wired sensor network



CURRENT VOLTAGE CURVE PREDICTION OF ANNEALED NI N GAAS SI IN

SCHOTTKY BARRIER DIODE USING ANFIS

SERDAR KOCKANAT1, HULYA DOGAN1, YAVUZ TURKAY1, MUSTAFA

SEKER1, DERYA BETUL UNSAL CELIMLI1, MURAT CIKAN1

1 Cumhuriyet University, Turkey

ABSTRACT

Measurement of the current-voltage characteristic of Schottky barrier diode under various

annealing temperature like as 200 C and 600 C requires intensive labor and it is costly

procedure. Thus, this paper proposed a method of using an adaptive-network-based fuzzy

inference system (ANFIS) to model and predict current-voltage curve of Schottky barrier

diodes (SBDs) for thermal annealing process. The Ni/n-GaAs/In Schottky barrier diodes

was produced in the laboratory and the current-voltage characteristic of the Schottky diode

were measured from 200 C to 600 C with steps of 100 C. In this approach, ANFIS model

includes three input parameters. These parameters are temperature, voltage and current

and they were used for training of the ANFIS model. A hybrid grid partitioning ANFIS by

Generalized Bell membership function was employed for the training of the experimental

data. The current-voltage data from 200 C to 500 C were used to train ANFIS model. In

order to test the accuracy of the ANFIS model, estimated and experimental current-voltage

data for 600 C were graphically compared. Also, from scatter diagram, the fitting curve of

the data is y=0,995x-0,091, so the rate of slope and intercept are very close to one and

zero. That is to say, the predicted values are very near to the experimental values. The

results shows that the ANFIS model is an efficient model and can be used to predict for

current-voltage curve of the Schottky barrier diode for different annealing temperature.

Acknowledgement: This work was supported by Scientific Research Projects Coordination

Unit of the Cumhuriyet University (CUBAP) (Project Number: M-649). Corresponding

Author: Hülya Dogan ([email protected])

KEYWORDS - ANFIS, schottky barrier diode, thermal annealing, artificial bee colony



DESIGN AND SIMULATION OF MINIATURE MICROSTRIP ANTENNA USING

CIRCULAR MUSHROOM EBG STRUCTURE FOR WLAN PAN APPLICATION

BELABBAS KHEDEDJA1, KHEDROUCHE DJAMEL1, HOCINI

ABDESSELAM1, BOUGOUTAIA TAHAR1

1 Universite De M Sila, Algeria

ABSTRACT

The popularity of personal wireless communication devices has grown extremely and it

requires a very low profile and efficiency. The microstrip patch antenna achieves this

requirement, but the main problem of these antennas is how to design a miniature antenna

for wireless communication with high quality service. For this aim, the artificial materials;

such as metamaterials and electromagnetic band gap (EBG) structures, are used to design

and fabricated a miniature microstrip antenna with high performances. In this work, a

microstrip patch antenna printed on FR4-Epoxy dielectric and excited by a probe feed is

designed and simulated in the first part. Secondly we applied a circular mushroom-EBG

structure on the dielectric substrate of this antenna, which contributed to obtain a much

miniaturized antenna, (the size of the antenna decreased intensively about 60%). The

comparison between the microstrip antenna with and without EBG structure is made on

the basis of different characteristics, such as bandwidth, return loss and the size. Due to

presence of EBG structure in the dielectric substrate, the electromagnetic band gap form

that reduced the surface waves and increases the overall dielectric constant. As a result an

improved band width antenna, better impedance matching and reduced size are achieved.

KEYWORDS - artificial materials, micrstrip patch antenna, EBG structure,

miniaturization, band width



RAPID PROTOTYPING LOW COST SOLAR SIMULATOR FROM OF THE

SELF COMPONENTS

LUTFI ARDA1, M C ALPHAN1, D AKCAN1, ALI GUNGOR1


ABSTRACT

This paper proposes a configuration of a low cost solar simulator for basic PV studies.

Solar simulator utilizes tungsten halogen lamp as light source and laboratory bench power

supply and DMM for I-V characterization system. Light intensity controllable by means of

input power of tungsten lamp or utilized Fresnel lens assembly. The solar simulator

capable of generating 100W/m2 of irradiation with exceeding expectations of BAA class

solar simulator without employing any spectral correction technique. Uniformity of

irradiance and spectra of light source mapped with developed prototype hardware and

software. Spectral uniformity measurements carried out at multiple point in range of

200nm -1000nm with alongside the total irradiance measurement. To test the capability of

proposed solar simulator a case study designed for characterization of quantum dot solar

cells.

KEYWORDS - Solar Simulator, I-V measurement, characterization



PID TYPE STR BASED ON SVR FOR NONLINEAR SYSTEMS

KEMAL UCAK1, GULAY OKE GUNEL2

1 Mugla Sitki Kocman University, Turkey; 2 Istanbul Technical University, Turkey

ABSTRACT

Generalized self-tuning regulator (STR) based on support vector regression (SVR) which

was previously proposed in [1] is used with a PID controller in the STR block to control a

continuously stirred tank reactor (CSTR) system. Two separate SVR blocks are utilized to

tune PID controller parameters and estimate the system model. Simulation results are

presented to assess the performance of the controller.

KEYWORDS - Adaptive Control, PID Type STR, Self-Tuning Regulators, Support Vector

Regression, System Identification



THE EFFECT OF DISTURBANCES ON GRID TIED POWER CONVERTERS

AND DSOGI BASED POWER CONTROLLER SOLUTION

DOGAN CELIK1, M EMIN MERAL1

1 Yuzuncu Yil University Van, Turkey

ABSTRACT

The grid disturbances have various impact on control of power converter devices

interfaced with hybrid renewable energy sources (HRES) such as solar cell, fuel cell and

wind turbine. Control is one of significant parameter for the deployment of renewable

energy systems. The various control strategies are implemented for the HRES applications

under unbalanced conditions in the literature. The grid disturbances lead to power quality

problems such as voltage sag/swell, voltage unbalanced and fluctuations, long interruption

and harmonics. In this paper, proportional resonant (PR) based power controller is applied

to the solar power system and dual second order generalized integrator-phase locked loop

(DSOGI-PLL) based power controller (PR controller) in stationary reference frame

provides solutions to overcome these problems. The PR controller based on DSOGI-PLL

is examined and analysed under various cases such as unbalanced grid voltage, phase

faults conditions. The effectiveness and correctness of the controller is tested and verified

by using PSCAD/EMTDC software package.

KEYWORDS - Renewable Energy Sources, PR controller, Phase Locked Loop, DSOGI,

Grid Disturbances, Power Quality



PREVENTIVE PROPERTIES OF METAL COMPOSITE FABRICS AGAINST

ELECTROMAGNETIC RADIATION

RUSLAN ABDULLA1

1 Karabuk University, Turkey

ABSTRACT

The physical properties of the metal blended composite fabrics contribute to the deduction

of how preventive the fabric is, against electromagnetic radiation. Likewise, the metal

contained in fabric texture behaves differently under the exposure of electromagnetic field.

Accordingly, the metal type and the incident frequency, fabric type and its geometry,

thickness, density has been analyzed along with the correlations among themselves.

Shielding measuring techniques were specified in accordance of feasibility, stability and

reliability. The factors effecting the shielding and the accuracy of measurement methods

were revealed in brief expressions.

KEYWORDS - Electromagnetic radiation, shielding, composite fabric, preventive

properties, measurement techniques



AN EXPERIMENTAL COMPARISON AND PERFORMANCE ASSESSMENT OF

NONLINEAR CONTROL TECHNIQUES FOR AN INDUSTRIAL AIR HEATING

SYSTEM

NECDET SINAN OZBEK1, ILYAS EKER1

1 Cukurova University, Turkey

ABSTRACT

In this study, a set of nonlinear control technique has been instigated experimentally.

Different structures of sliding mode control technique is compared experimentally, further

backstepping control strategy is designed and tested in real-time. The control schemes

have been successfully applied to a time-delay air heating process. The contributions of

this paper are; i) describing a set of nonlinear control techniques, ii) analysing the effect of

the delay approximation technique on nonlinear controller, iii) determining the advantages

and drawbacks of the proposed feedback strategy according to control effort and a number

of performance metrics, iv) conducting an experimental implementation to demonstrate

the effectiveness and validity of the elaborated control methods. A comprehensive

performance evaluation is demonstrated via results of real-time experimental applications

through a set of performance indices. Particular attention is paid to the tracking precision,

quality of the transient response, robustness against disturbances and energy efficiency of

the each control strategy with respect to some performance criteria such as integral

absolute error (IAE), integral squared error (ISE), maximum absolute error (MAE), error

variance (EV), integral time squared error (ITSE), and integral squared control input

(ISCI). Moreover, complementary comments are highlighted on the advantages and

drawbacks of each control structure. Due to the popularity of the problem, contribution of

the current study is to provide an in-depth guide to the control designers of nonlinear

systems.

KEYWORDS - Time Delay Systems, Sliding mode control, Backstepping control, Real-

time experiment, nonlinear systems



THD COMPARISON OF VENTURINI AND SVM ALGORITHMS OF MATRIX

CONVERTERS FOR VARYING SWITCHING FREQUENCY

TOLGA SEN1, ATA SEVINC1


ABSTRACT

In this research, total harmonic distortions (THDs) of matrix converters (MCs) are

compared as a function of switching frequency for Optimum Venturini and space vector

modulation (SVM) algorithms. The purpose is to determine a small change in the desired

switching frequency yielding a lower THD. In the simulations, the load fed by the MC

consists of a resistor and an inductive reactance. For a fair comparison with constant

current amplitude, the reactance is chosen constant. Each simulation is carried on changing

the switching frequency with a small step through a wide frequency range. It is discovered

that the SVM method usually ends up with less THDs at irregular switching frequencies

whilst the Optimum Venturini method yields less THDs at regular steps in the switching

frequency. Therefore the Optimum Venturini method can be safely preferred for lower

THDs as described here.

KEYWORDS - AC-AC power conversion, matrix converter, total harmonic distortion,

Venturini algorithm, SVM algorithm



GENETIC ALGORITHM BASED FUZZY LOGIC SYSTEM FOR SHORT TERM

LOAD FORECASTING

MURAT LUY1, VOLKAN ATES1, NECAATTIN BARISCI1, ERTUGRUL CAM1


ABSTRACT

Because of the rising energy prices and the fact that the energy generation-consumption

balance is important, a lot of work has been done on forecasting of electrical load. In this

study, in parallel, load estimation was studied. For this purpose, a Genetic Algorithm

based Fuzzy Logic (GA-FL) system for load estimation was modelled by considering

meteorological and calendar effects. Thus, Turkey's short-term electrical load requirement

was tried to be determined. In the model, GA optimization approach was used to

determine the most appropriate rule base for increasing the stability of the FL model. In

the training and test phases of the proposed model, hourly air temperature and electrical

consumption data for the years 2011-2013 were used. The results were shown that natural-

inspired optimization models, such as genetic algorithms, have an effect on, and even

increase, predictability and accuracy in such systems.

KEYWORDS - Short-Term Load Forecasting, Fuzzy Logic, Genatic Algorithm



DESIGN AND MOTION EQUATIONS OF A ROBOT DEVELOPED FOR USE IN

LAPAROSCOPIC SURGERY

SERHAT AKSUNGUR1, OGUZ YAKUT2

1 Inonu University, Turkey; 2 Firat University, Turkey

ABSTRACT

Nowadays, robot technology is improved high and its usage is widespread. Robots that

have been encountered in many sectors have also been used in surgical operations in

recent years. RCM (Remote Center of Motion) mechanisms that occupy less space and are

relatively more functional, take the place of these robots. In this study, RCM mechanisms

used in surgical operations will be mentioned. In this context, a original surgical robot

design and motion equations which are not in the literature belongs to surgical robots will

be described. All parts of the designed robot and the solid model assembly of the robot

have been created with 3D design program. Motion equations have been obtained by

Lagrange mechanics using kinematic equations of robot.

KEYWORDS - Surgical robots, RCM mechanisms, Lagrange motion equations,

Laparoscopic surgery



INDUCTANCE AND LOSS BEHAVIORS OF MEDIUM FREQUENCY HIGH

POWER GAPPED CORE INDUCTORS

SELAMI BALCI1, M BAHA BAYRAM1, NECMI ALTIN2, IBRAHIM SEFA2

1 Meb, Turkey; 2 Gazi University, Turkey

ABSTRACT

Three phase power factor correction (PFC) circuits consist of an inductor on each phase,

which is a key component of the design. When the design power reaches hundreds of kW

then the core material selection will be more important. Gapped-core structure is the key

to provide inductance stability when the inductor is designed with alloy core materials.

Besides the total air gap length, determining the number of gaps is also important to fulfil

the requirement of minimizing the fringing effects caused by the two or multiple

(distributed) air gaps. In this study, the losses and the inductance variation of the PFC

inductor designed at a certain power and inductance values are analyzed comparatively

with respect to different gapped-core structures. Different inductor models have been

developed with different air gap configurations and electromagnetic modelling and

analysis of them have been performed with the finite element analysis software, and the

obtained simulation results have been reported. It is seen from the simulation results that

distributed air-gap provides better results in terms of inductance stability, roll-off values

and total power losses.

KEYWORDS - Inductor design, FEA, gapped-core, inductance stability, co-simulation.



PARTIAL GAIN INCREASE AT DESIRED FREQUENCY RANGE FOR UWB

MICROSTRIP ANTENNAS USING BAND STOP FSS

KOMRON HOTAMI1, CENGIZHAN MUSTAFA DIKMEN1, GONCA

CAKIR1, AYBIKE KOCAKAYA1, SIBEL CIMEN1


ABSTRACT

Many studies have been done on gain enhancement in ultra-wideband (UWB) microstrip

antennas. Microstrip antennas are preferred in many applications because of their small

size, easy production and low cost, but they have low gain characteristics. Various

techniques are used to increase gain in UWB microstrip antennas. Some of them are

concentrated on attaching a reflective metal layer, frequency selective surface behind the

antenna, and directive structures in front of the antenna. With these techniques, the gain

can be increased, but high gain cannot be obtained along the radiation frequency band, and

the gain at low frequencies will be low. In some applications where UWB antennas are

used (e.g., underground imaging radars), constant gain is desired over the frequency band.

Therefore, instead of increasing the gain at whole band, gain enhancement at frequency

intervals where the gain is low is more appropriate. In order to achieve the aim, band-stop

Frequency Selective Surface (FSS) structures that operate at the frequency range at which

we want to increase the gain of the antenna can be used. In this study, as a first step, UWB

microstrip antenna is designed to operate in the frequency range of 3.1-10.6 GHz. Then, to

show the partial gain increase using FSS, a higher order band-stop FSS structure operating

at 8-12 GHz band is designed and located to the antenna. Finally, a partial gain increase

using high-order FSS is obtained.

KEYWORDS - FSS, UWB Antenna, Microstrip Antenna



FRACTIONAL ORDER DISCRETE TIME MODEL ORDER REDUCTION

BASED ON SVD APPROACH

KARIMA GHARBI1, DAOUD BERKANI1

1 Ecole Nationale Polytechnique D Alger, Algeria

ABSTRACT

This paper deal with the approximation of linear discrete-time invariant (L.T.I) fractional

order system. The presented method is based on the left and right eigen spaces and the

Schur decomposition of the expanded state space matrices of the original fractional order

system. The advantage of our method is the fact that it operates in the general case, i.e. for

the minimal and non-minimal systems. A simulation on fractional order systems is given

which proves that the reduced order models are always stable, and faithfully follow the

initial model in the time and the frequency domains.

KEYWORDS - Fractional order systems, Model order reduction, State space matrices

expansion, Hankel singular values, singular values decomposition (SVD), Schur

decomposition, controllability and observability gramians



ADAPTIVE CRUISE CONTROL SYSTEM FOR ELECTRIC VEHICLES USING

FUZZY LOGIC

FATIH KARAGOZ1 , MUZEYYEN SARITAS2

1 Ministry of Science Industry and Technology, Turkey; 2 Gazi University, Turkey

ABSTRACT

In this study, an adaptive cruise control system for an electric vehicle is designed using

fuzzy logic controller. The simulation of the system has been realized on a computer

software. According to simulation results, fuzzy logic controller has maintained to cruise

at desired speed of electric vehicle if there is safety distance between host vehicle and

leading vehicle. In case of any safety distance between the vehicles, the safety distance is

ensured by prevention of acceleration to desired speed. The safety distance has ensured the

time gap rule in ISO 15622 which is a standard for adaptive cruise control systems.

KEYWORDS - Electric vehicle, Fuzzy logic, Adaptive cruise control system



MULTIMODE LASER DIODE DRIVER DESIGN FOR PHOTODYNAMIC

THERAPY

MUHAMMED GULER1, ALI FURKAN KAMANLI1, MUSTAFA ZAHID

YILDIZ1, AYLA EREN OZDEMIR1, HYUN SOO LIM1


ABSTRACT

Photodynamic therapy (PDT) is a minimally invasive photochemotherapy method based

on the principle that photosensitizer induced target tissue is stimulated by the light sources

in the appropriate wavelength and sufficent energy. Excitation of the Photosensitisier leads

cells to the necrosis. After the cell necrosis healthy cells produced by the body and

Phtodynamic therapy has a high success rate in soft tissue cancers. In this study, LASER

diode based light source system is designed which can radiate at 635nm to excite 5-

Aminolevulinic acid (5-ALA) which is used as PDT synthesizer. Since the current /

voltage (I / V) relationship of the laser diodes is exponential and the output power is

related to the current passing through the diode, the LASER diodes are driven by the

constant current source. In order to improve the performance of the PDT and minimize the

thermal damage to the tissue, the light is switched in different modes instead of only in the

continuous mode. Thus, the light can reach the instantaneous higher powers and intervene

in cancerous cells in deeper regions of the tissue. In pulse and burst-pulse modes based on

the switching principle of light, spike voltages are observed in the transient state at the

rising edges of the LASER voltage. These high amplitude voltages are damped to prevent

damage to the laser diode. In spite of the damping of the spike voltages, a LASER diodes

parallel capacitor connection has been tried. However, since this causes the delay of the

LASER diode to be delayed, the parallel MOSFET is connected to the LASER diode and

the switching time of this MOSFET is being controlled by the microcontroller

(DsPic33WP128MC506). Finally, the laser diode voltage is fed back to the analogue

digital converter of the microcontroller to provide PID control of the output power.

KEYWORDS - Photodynamic therapy, high power laser driver, laser driver design



REVIEW OF EXTREMELY LOW FREQUENCY (ELF) ELECTROMAGNETIC

FIELD MITIGATION TECHNIQUES

KENAN OZEL1, ILHAN KOSALAY1

1 Ankara University, Turkey

ABSTRACT

Due to the increased energy demand, the electromagnetic field level in our environment is

constantly increasing during the 60 years we have passed. On the rise of the

electromagnetic field, energy transmission and distribution lines, switching points,

transformer centers, electrical installations, panels and devices with high power

consumption are effective. The effects of power frequency electromagnetic fields emitted

from these sources on the living things have always been controversial. While some

scientists argue that extremely low frequency (ELF) magnetic field are not harmful to

health, some scientists have argued that those who work on electrical power systems and

the people living near those systems have negative health effects by suggesting discomfort

such as headache, fatigue, dizziness and sleeplessness. When we look at the history of this

subject, until the 1960s, 50-60 Hz magnetic fields were considered to be non-hazardous. In

1979, epidemiological study of Wertheimer and Leeper on the possible negative effects to

human health of magnetic fields with extremely low frequencies caught the attention of

those working in the field. In that study, it has been reported that power frequency

magnetic fields arising from energy transmission lines significantly increase the incidence

of blood cancer in children living near the line. Up to now, many epidemiological and

biological studies have been conducted on this subject. In addition, community and

occupational exposure limit values for harmful electromagnetic emission were determined.

These limit values have been adopted by many countries including Turkey. Limit values

are determined by the International Commission on Non-Ionizing Radiation Protection

(ICNIRP). ICNIRP publishes guidelines for low and high frequency bands, examining

possible negative effects of electric and magnetic fields on human health. Non-ionizing

electromagnetic radiation at the power frequency passes through the human skin and

weakens inside the body. It does not cause the formation of charged ions in the passing

part because of the low particle energy. However, where they pass, they cause some

effects such as electrical overload, nervous disorders, muscle reactions, dizziness, nausea

and cellular malfunction. For this reason, it is necessary to reduce the electromagnetic

emission at the power frequency and to draw it below the limit values determined by the

international commission. For this, there are many methods such as repositioning of the

sources which emit the power frequency electromagnetic field, shielding with conductive

and ferromagnetic materials, using active and passive mitigation loop. In this study, a

detailed examination will be made on these reduction methods.

KEYWORDS - Magnetic field, Mitigation, Shielding, Passive loop, Active loop



GRAPHICAL STABILIZATION APPROACH FOR UNSTABLE FIRST ORDER

LINEAR SYSTEMS WITH TIME DELAY

ABDELKADER MADDI1, ABDERREZAK GUESSOUM1, DAOUD BERKANI1

1 University of Blida 1, Algeria; 2 Polytechnic School of Algiers, Algeria

ABSTRACT

In this paper, a graphical stabilization approach is proposed and analyzed for a class of

unstable first order linear systems with time delay. We first show that the control designs

based on time invariant models are unable to guarantee stability and asymptotic tracking

for unstable first order linear systems in general case. So, the condition stability is

analysed graphically by computing the first derivative and plotting the graph of a function

with precision; the first derivative allows us to determine the critical points and several

conditions of stability. Therefore, it’s important to note that the method can guarantee the

existence of a proportional gain to ensure the stability of the closed-loop system such that

the time delay is small relatively to the time constant. Finally, a numerical example

illustrates the efficiency and performances of the proposed approach.

KEYWORDS - Linear system; time delay; proportional gain; asymptotic stability; output

feedback



THROUGHPUT ENHANCEMENT OF MULTIPATH TCP BY SUBFLOW

BANDWIDTH MEASUREMENT IN WIRELESS NETWORKS

MIN SUB KIM1, KI MOON HAN1, JAE YOUG LEE1, BYUNG CHUL KIM1

1 Chungnam National University, Korea Republic Of

ABSTRACT

Multipath TCP (MPTCP), which is an extenstion of TCP, is a protocol that supports

multipath transmission. MPTCP has advantages in compatibility with original TCP

because it uses option field in TCP protocol. On the other hand, it has the same problem as

TCP in wireless network because TCP misinterprets the packet loss due to channel noise

or handoff as a congestion, resulting in congestion control and degradation in

performance. Thus, if MPTCP protocol is applied in wireless network, loss in a subflow

make congestion window decrease and aggregation effect of MPTCP does not occur. In

this paper, we propose a congestion control method that measures the bandwidth on each

subflow of MPTCP and reduces the size of congestion window by the measured

bandwidth when packet loss occurs in order to solve MPTCP performance degradation due

to frequent congestion control from high bit loss rate in wireless environment. In order to

analyze the performance of the proposed congestion control, we implemented the

proposed congestion control in the Linux kernel and compared it with the existing MPTCP

in the wireless testbed and the actual wireless network. Experimental results show that the

proposed congestion control has better throughput performance than the conventional

MPTCP congestion control in wireless environments and furthermore it meets MPTCP

design.

KEYWORDS - MPTCP, TCP, congestion control, wireless network, Multipath TCP



A MOBILE APPLICATION DESIGN FOR INTRODUCTION OF DEVICES OF

THE PHYSICS LABORATORY

NIGMET KOKLU1, DUNDAR YENER2

1 Selcuk University, Turkey; 2Abant Izzet Baysal University, Turkey

ABSTRACT

Mobile learning is a powerful environment that combines mobile application and e-

learning. Mobile application development, which is one of the new technologies in today's

our modern world, continues to rapidly increase its popularity. Many experimental sets,

measuring instruments and high cost equipment are used in physics laboratories. Students

need to use these laboratory equipment well because all responsibilities of these

instruments were largely given to the students. If the laboratory equipment is damaged,

education will have a hitch, and this case will also bring high cost to education institutions.

Operating the devices in accordance with the user guide is necessary to prevent such

undesirable situations. For this reason, students need to be informed about the use of the

equipment, and how to do experiments. Considering these situations; a mobile application

containing video images, brochures, and animations describing the use of devices found in

the physics laboratory. Considering all these situations; a mobile application has been

designed that explains the use of devices found in the physics laboratory and includes their

video image, brochures, and animations. In this designed mobile application, it is

considered that the images, graphics, shapes and texts used are understandable and that the

colors used are integrated in the screen. It is also important that the animations and images

used are not eye strain and that the view is clear.

KEYWORDS - Mobile Application, Physics Laboratory, Introduction to Laboratory

Devices



DESIGN OF INTERACTIVE E BOOK APPLICATION FOR PHYSICS COURSE

NIGMET KOKLU1, DUNDAR YENER2

1 Selcuk University, Turkey; 2 Abant Izzet Baysal University, Turkey

ABSTRACT

Differentiations are seen in the book format along with the rapid development of

technology. Thanks to the possibilities provided by the electronic environment, e-books

are supported by electronic contents such audio, music and images. Also, getting feedback

in the e-book is easy. In addition, e-books have high storage capacity and it encourages

people to read them. Another advantage of the book is that it is accessible for special

needs individuals. E-books are usually read by using tools such as tablets and computers

which have the necessary software for it. Interaction is defined as the communication

between two or more people/ objects, or interaction of two or more persons or objects with

each other, or two or more people/objects reacting to each other as a result of a stimulus.

The interactive e-book offers much more to learners than a book. The interactive e-book,

which is enriched with multimedia elements such as audio, video, animation and has an

entirely interactive learning environment, enables more effective and productive learning

by drawing learners to the center of the work process. Considering the philosophy of” less

is more” that, the most important and brief information about the subject is taught via

video. It is thought that understanding of difficult subjects in the field of physics will be

understood and persisted through the interactive e-book. Especially, it is aimed to

minimize the misconceptions related to electricity. Considering of all this, interactive e-

book applications are designed for physics lessons so that students understand the

complicated subjects more easily and to help eliminate the misconceptions about them.

Adobe InDesign software is used for this design. In this designed mobile application, it is

considered that the images, graphics, shapes and texts used are understandable and that the

colors used are integrated in the screen. It is important that the images used are not eye-

straining and the view is clear. It has taken care of the font is legible in terms of style,

color and size. Also, very long texts are not preferred in the designed e-book. Thus, it is

ensured that the information to be given becomes more remarkable.

KEYWORDS - Interactive E-Book, Application Design, Physics Course



INVESTIGATION OF ATMOSPHERIC ATTENUATION FACTOR FOR HIGH

SPEED TERAHERTZ WIRELESS COMMUNICATION LINKS AT 0 275 THZ

AYHAN YAZGAN1


ABSTRACT

Atmospheric attenuation factor limits high speed wireless communication links due to the

high absorption of gases in the communication medium. In this study, several kinds of

scenarios have been implemented using MATLAB platform. In order to see the negative

effect of atmosphere attenuation clearly, a link budget has also been conducted at 0.275

THz according to the good and bad conditions of the communication medium.

KEYWORDS - Atmospheric effects, attenuation, high speed wireless communication, link

budget, terahertz communication



ANNEALING TEMPERATURE DEPENDENT CHARACTERIZATION OF NI AU

N GAN SCHOTTKY DIODE USING ARTIFICIAL BEE COLONY ALGORITHM

HULYA DOGAN1, SERDAR KOCKANAT1, YAVUZ TURKAY1, MUSTAFA

SEKER1, MURAT CIKAN1, DERYA BETUL UNSAL CELIMLI1


ABSTRACT

In this work, we have analyzed the effect of the thermal annealing process on Ni-Au/ n-

GaN Schottky diodes. Also, the diode characteristics were presented for pre- and post-

annealing process using two important electrical parameters ideality factor (n) and barrier

height (ϕb). For this purpose, in the first stage, n-GaN films used and were epitaxially

grown by metal organic chemical vapor deposition (MOCVD) on c-plane Al2O3 (Saphire)

substrates. This crystal grown on by us an undoped GaN layer thickness of 1250 nm was

grown followed by the growth of 1350 nm thick n-GaN: Si (Nd = 1.01x1018 cm-3) layer.

On the enlarged structure, Schottky contacts were prepared by evaporation deposition of

Ni and Au metals. At last, Ni-Au/ n-GaN Schottky diodes were produced. For pre-

annealing process, the I-V characteristics of unannealed prepared devices were measured

using a Keithley 487 Picoammeter/Voltage Source in the room temperature. After

formation and measurement of the unannealed Schottky diodes, they were annealed at for

1 min in N2 atmosphere. For post-annealing process, the I-V characteristics of annealed

Schottky diodes were measured with same measuring instrument. In the second stage, in

order to analyze the beneficial effect of thermal annealing process on Ni-Au/ n-GaN

Schottky diode, the modified artificial bee colony algorithm has been used in extraction of

ideality factor (n) and barrier height (ϕb). For pre-annealing process, ideality factor (n) of

3.5888 and barrier height (ϕb) of 0.5390 eV of were achieved. For post-annealing process,

ideality factor (n) of 1.2898 and barrier height (ϕb) of 0.6410 eV of were obtained. The

results show that ideality factor decreases while barrier height increases. So, the ideality

factor, which is equal to one for ideal diode, moves closer to one and the characteristics of

the diodes were improved.

KEYWORDS - Schottky diode, modified artificial bee colony, annealing, metal-

semiconductor contact



PERFORMANCE COMPARISON BETWEEN IP AND MPLS NETWORKS

USING OPNET MODELER

ALI GEZER1, MARWA KHALEEL1


ABSTRACT

Multi-Protocol Label Switching (MPLS) is an up to date technology used in packet routing

in ISP’s backbones due to its superiorities in many performance criteria including VPN,

QoS and traffic engineering supports. It allows routing traffic efficiently in the network,

reduces network congestion and improves performance and efficiency of the Internet.

Since MPLS routing supports quality of service, it provides smooth and steady voice

transfer. Delivering real time traffic such as video conference doesn’t affect the

performance of other applications and provides high security with its data encryption

which helps to connect all the branches together in a safely manner. In this paper, we

conduct a performance comparison study of two networks which use MPLS and IP routing

for voice, video conferencing and web applications using OPNET simulator environment.

Jitter, end to end delay, send and received traffic performance criteria are taken into

account to compare the performances of two routing strategy. In terms of dropped packets,

throughput and jitter, MPLS network provides better results in comparison to IP networks.

KEYWORDS - Multiprotocol label switching (MPLS), Quality of service (QoS), Routing

Internet Protocol (RIP), Background traffic, Opnet modeler



CONTROL OF TRIGLIDE ROBOT WITH ARTIFICIAL NEURAL NETWORKS

MUHAMMET AYDIN1, OGUZ YAKUT1


ABSTRACT

In this paper, artificial neural networks control of a three degrees of freedom triglide

parallel robot has been realized via inverse-forward kinematics equations and dynamic

equations of it. The coefficients of the artificial neural networks controller were calculated

by using genetic algorithm optimization method. The controls have been performed for

references of x, y and z coordinates of the triglide parallel robot to demonstrate the success

of artificial neural networks control method. Matlab package program has been used for

numerical solutions. The system responses of the triglide parallel robot have been obtained

graphically and the results have been examined.

KEYWORDS - Dynamics, Kinematics, Artificial Neural Networks, Triglide, Genetic

Algoritm



CENTRE OF MASS MOTION AND THE MOTT TRANSITION IN LIGHT

NUCLEI

AREEJ ABDELRAHMAN1, MAHMOUD STATY1, HENRRY JAQAMAN1

1 Birzeit University, Palestinian Territory

ABSTRACT

We calculate the effect of the inclusion of the centre of mass (CM) motion on the survival

of light nuclear clusters in a vapour of nucleons at very low densities and oderate

temperatures. The Mott transition refers to the dissolution of these clusters above a certain

vapour density (the Mott density). The cluster CM motion has so far been included

approximately or at fixed CM momentum in calculations of the Mott density. Often it is

neglected altogether due to technical difficulties. In the present work we include the CM

motion exactly for clusters up to A = 4 and calculate its effect by averaging over all values

of the CM momentum at a given temperature rather than at a fixed value of the CM

momentum. We demonstrate that the inclusion of the CM momentum shifts the Mott

density to higher values. The shift, however, decreases as the cluster mass increases.

KEYWORDS - Low density nuclear vapour, cluster formation, Mott transition, centre of

mass effects



CLASSIFICATION OF COMMONLY HEARD LUNG SOUNDS RECORDED IN

SINGLE CHANNEL

SELIM ARAS1 , MEHMET CINEL1 , ALI GANGAL1


ABSTRACT

Auscultation is one of the first methods used to diagnose lung diseases. The success of

auscultation depends on the hearing ability and experience of the physician who

performing the auscultation. In the past, studies were carried out on the automatic

recognition of lung sounds to make people independent of their abilities. However, most

of these studies have been performed in multi-channel. Auscultation is essentially

performed as a single channel. Therefore, in this study, commonly heard lung sounds

consisting of healthy, rhonchi, crackles (consist of fine, coarse, and Velcro crackles) were

recorded as the single channel. An electronic stethoscope (20-2000Hz) capable of

recording in the frequency range suitable for lung sounds and digital sound recorder were

used in the recording. All records were taken and labelled by two physicians. The dataset

was randomly selected and then equally divided into training and test clusters. This

process has been repeated 100 times. The Leave-One-Out Cross-Validation method was

applied to the obtained training and test clusters in order to get rid of the random selection

problem. Unlike previous studies, time and frequency domain attributes and attributes

derived from Mel Frequency Cepstrum Coefficients (MFCC) were used as features in this

study. Support Vector Machine, Naive Bayes, and k-Nearest Neighbour (k-NN)

algorithms were used in the test phase of the classification. The mean and standard

deviation of classification results from randomly selected training and test clusters were

presented in the results section. After examining the results of each feature, the Sequential

Forward Selection method was applied to investigate the success of the appropriate

combinations of features. The mean and standard deviations of the MFCC and mean of

features obtained from frequency domain is the most successful combination. When this

combination was classified by the k-NN algorithm, it gave the most successful result with

96,47%1,94.

KEYWORDS - Common lung sounds, Frequency domain features, Time domain features,

Mel frequency cepstrum coefficients, classification



POSITION ESTIMATION OF SENSORS USING PARAMETRIC EQUATION OF

A HYPERBOLA WITH STATIC AND DYNAMIC COMMUNICATION RANGE

DOGAN YILDIZ1, SERAP KARAGOL1


ABSTRACT

A Wireless Sensor Network (WSN) represents a wireless network of autonomous devices

that are spatially distributed in the application domain. Sensors are used in a WSN to

monitor physical or environmental conditions. A WSN system incorporates a gateway that

provides wireless connectivity back to the wired world and distributed nodes. Accurate

and low cost localization is a crucial requirement for wireless sensor networks to be used

in a wide variety of applications. Time Difference of Arrival (TDOA) localization is one

of the widespread methods by virtue of its high estimation accuracy. The TDOA algorithm

is guaranteed to find all possible relevant solutions, even when implemented on a

computational node with limited capability. In our recent study, Parametric Equation

based Hyperbolic Localization Algorithm in 2D space were proposed. In the Parametric

Equation based Hyperbolic Localization Algorithm in 2D, two hyperbolic intersection

points were found and hyperbolas were rotated clockwise at a certain angle. Uniformly

distributed network were used to localize the target nodes while aforementioned algorithm

is being run. In this paper, Normal, Rician, Burr and Rayleigh distributed networks is used

for localization and the localization performance of the networks are evaluated and

compared using MATLAB simulations.

KEYWORDS - Localization, Parametric-Hyperbolic Based Algorithm, Statistical

Distributions, Time Difference of Arrival, Wireless Sensor Networks.



SELJUK STAR PATCH ANTENNA DESIGN ON WOOLENFELT MATERIAL

FOR WEARABLE 2400 MHZ ISM BAND APPLICATIONS

DILEK UZER1, SEYFETTIN SINAN GULTEKIN1, RABIA TOP1


ABSTRACT

In this study, a Seljuk star geometry microstrip patch antenna design, which operates at

2400 MHz frequency, is achieved on Woolenfelt material for using in Industrial, Scientific

and Medical (ISM) band applications. Woolenfelt is a natural material that can be

produced at desired thicknesses with low cost and has a lower dielectric constant as 1,45.

Owing to easy fabrication and being adaptable to wearable applications, this material is

preferred for the design. In addition to be cheaper than other dielectric substrates

andfabricated in our country are others reasons for preferring the material as square-shape

substrate of Seljuk star antenna. All designs are fed by using coaxial feedline. The designs

are improved by HFSS. While simulations are repeated for patch dimensions, ground

plane and substrate dimensions, the feed point and the thickness of the dielectric substrate

one by one for each parameter at a time, so the effects of these parameters on antenna

performance can be exhibited clearly. At the end of the study more than one design are

obtained that can be used for wearable ISM band applications. These designs resonate near

and at 2400 MHz with a bandwidth over % 35. For the best design, thanks to its under -50

dB S11 result, it can be used in practical applications on human body. Ground plane and

substrate thickness dimensions are reduced by 60% and 49%, respectively. The

empedance bandwidth is enhanced to 47,03% with optimizations. The design is simple and

easy fabricated with low cost so it can find place in nowadays mobile and wearable

applications.

KEYWORDS - Seljuk star, patch antenna, HFSS, ISM, wearable applications.



A COMPARATIVE STUDY FOR COOPERATIVE MISSION PLANNING OF

UAV SWARM

EBRU KARAKOSE1


ABSTRACT

Today, Unmanned Aerial Vehicle (UAV) are used in many military and civilian

applications. In recent times, it is necessary to use UAV swarm in more complex tasks

such as multi-target tracking and tracking, multi-purpose task coordination and strategic

objectives identification. Coordination of UAV swarm requires very complex control,

planning, communication, and computation. In this study, a comparison study is presented

for the optimization and control algorithms that can be used for coordinated task planning

of UAV swarm. Basically, the coordination process of UAV swarm consists of an

optimization algorithm for task planning and a decision-making approach for cooperation.

For this purpose, the performance and the performance of the optimization algorithms

were first compared in the study. Then, comparative results for different scenarios are

given for task planning in the study. From the optimization algorithms that can be used for

task planning of UAV swarm, the performance of particle swarm optimization, genetic

algorithm, artificial immune systems and hybrid algorithms are tested for the targeted

study and the results are presented. The performance comparisons aimed at the study are

presented with simulation data and the performance of the algorithms used is evaluated

according to these results. In addition to these, inferences are given to the selection and

design of optimization algorithms that can be used for task planning of UAV swarm in

operation.

KEYWORDS - UAV swarm, optimization, mission planning



DETERMINING TWO COMMON POWER QUALITY DISTURBANCES

VOLTAGE SAG AND VOLTAGE SWELL

CAGRI KOCAMAN1, MUAMMMER OZDEMIR1


ABSTRACT

In this paper, wavelet based feature extraction method are used for determining power

quality disturbances which are important for power systems. These power quality

disturbances used in this study are voltage sag and swell. They are two common type

power quality disturbances. 50Hz pure sine as a reference, voltage sag and voltage swell

are generated at MATLAB in order to extract feature vector of pure sine, voltage sag and

voltage swell. Sampling period is 25.6 kHz. Firstly, discrete wavelet transform is applied

to these power quality disturbances and 50Hz pure sine. 12 level Daubechies-4 wavelet

function are used. Parseval theorem is applied to wavelet coefficients for obtaining energy

distribution of detail coefficients of these relevant disturbances in different resolution

levels. Energy distrubution of detail coefficients of constituted power quality disturbances

and pure sine are obtained. Pure sine coefficients are considered as a reference. It is seen

that voltage sag, voltage swell and pure sine can be distinguished when examineing energy

distrubution of detail coefficients. Also data size can be reduced in small size. Then

amplitudes of voltage swell and sag are incresed. It is observed increased in detail

coefficients which are important for these power quality disturbances. Energy coefficients

of power quality disturbances and amplitude of power quality disturbances are in direct

propotion.

KEYWORDS - power quality disturbances, voltaga sag, voltage swell, wavelet transform,

parseval theorem, discrete wavelet filter



TIME FREQUENCY GRAY LEVEL CO OCCURRENCE MATRIX

DESCRIPTORS FOR DECEPTION DETECTION

ABDULKADIR SENGUR1, SAMI EKICI1, YAMAN AKBULUT1, TURGUT KAVAS1


ABSTRACT

In this paper, time-frequency domain gray level co-occurrence matrix (GLCM) descriptors

are used to discriminate between lies and truth in speech signals. The proposed method has

previously been successfully applied to EEG classification. This encouraged the authors to

adapt this scheme in order to discriminate between lies and truth in speech signals. To the

authors' knowledge, this is the first time that such a methodology has been applied in

deception detection. The proposed methodology initially applies the spectrogram method

to convert the input speech signal to a time-frequency image; after which, GLCM

descriptors are extracted. Liblinear SVM classifier is employed to classify the input speech

signals as lies or truth. The efficiency of the proposed method is tested with a real-life

dataset. The obtained results are encouraging, with classification accuracy reported at

almost 10% higher than previously reported results.

KEYWORDS - Deception detection, speech signals, GLCM



CHARACTERIZATION OF ZNO THIN FILMS DEPOSITED BY PULSED

LASER DEPOSITION PLD TECHNIQUE

AHMET TOPRAK1, SERAP YIGIT GEZGIN1, HAMDI SUKUR KILIC1

1 Selcuk University Konya, Turkey

ABSTRACT

Zinc oxide (ZnO) is a wide band gap (Eg=3.30 eV) n-type semiconductor material and

presents a high transparency for the visible region of the spectrum. Because of this optical

and electrical properties, ZnO thin films were preferable for electronics, optoelectronics,

sensors, semiconductor devices, electronic circuits and devices, especially in solar cells

and thin film transistor (TFT) applications. In this study, we present the optical and

structural properties of ZnO thin films, which we have deposited on glass substrates in our

laboratory by using PLD technique at room temperature under oxygen (O2) ambient gas

environment. An Nd:YAG laser was used for ablation process connected to a homemade

PLD vacuum chamber which was design by our group and produced in local industrial

area in Konya. The optical, electrical and surface properties of the ZnO thin films

produced were characterized by atomic force microscopy (AFM), X-ray diffraction (XRD)

and ultraviolet-visible (UV-VIS) absorption spectroscopy. As a result, PLD system has a

great flexibility to adjust its parameters, therefore, we used these parameters to fabricate

ZnO thin film. This parameters will be referred to further studies like the solar cell and the

TFT applications.

KEYWORDS - PLD, ZnO, TFT, Thin Film.



IMPACT OF BACKGROUND LANGUAGE ON TURKISH TEXT DEPENDENT

SPEAKER VERIFICATION

HAVVA CELIKTAS1, CEMAL HANILCI2

1 Antalya International University, Turkey; 2 Bursa Technical University, Turkey

ABSTRACT

Speaker recognition is a pattern recognition task which has long been studied, but the

accuracies are still far from the desired levels. The majority of the studies on speaker

recognition demonstrates the results obtained from databases consisting of English voice.

Since there are very few studies on Turkish speech, the speaker recognition performance

of the known successful methods on Turkish spoken language are uncertain. Therefore, in

this study, the performance of the Turkish text - dependent speaker verification system is

investigated using well-known Gaussian Mixture Model - Universal Background Model

(GMM - UBM) method. A Gaussian mixture model (GMM) for each speaker is derived

from the Universal Background Model (UBM) with maximum a Posteriori (MAP)

adaptation. The impact of the language used to train the UBM is considered in this study.

To this end, UBMs are trained using Turkish and English speech databases and their

effects on the performance of Turkish speaker verification system are compared.

Experiments are conducted on Turkish text-dependent speaker verification database using

two well-known feature extraction techniques, Mel-frequency cepstral coefficients

(MFCC) and Modified Group Delay cepstral coefficients (MGDC). Experimental results

reveal that training the UBM model using Turkish data considerably improves the equal

error rate (EER) of text-dependent speaker recognition in comparison to English data and

magnitude spectrum based MFCC features outperforms the phase spectrum based MGDC

features.

KEYWORDS - Turkish speaker recognition, Gaussian mixture model, Mel-frequency

cepstral coefficients, modified group delay features, universal background model.



A NEW MODIFICATION ON FORMULATION OF BROADBAND IMPEDANCE

MATCHING

METIN SENGUL1

1 Kadir Has University, Turkey

ABSTRACT

In broadband matching network design, one of the matching network parameters is

optimized until reaching the desired transducer power gain in the interested frequency

band. In transducer power gain calculations, the input or output impedance of the

matching network is employed. So it is necessary to find the values of the input or output

impedance. In this work, a new formulation is given to calculate these impedance values.

The proposed approach can be used to design both single and double broadband matching

networks with lumped, distributed or mixed elements. The utilization of the new approach

is illustrated via the given examples.

KEYWORDS - Broadband, matching, lossless networks, passive networks, real frequency

techniques.



PROPERTIES OF MULTILAYERED MG BASED GUIDED MODE RESONANCE

FILTERS GMRFS

DUAA ABUMURRA1, EDWARD SADER1

1 Birzeit University, Palestinian Territory

ABSTRACT

Due to the importance of guided mode resonance filters (GMR filter) in versatile

applications in integrated optics, spectroscopy, and holography. Such as; fiber optic

communication systems, laser resonator frequency selective mirrors, electro optic

modulators, and electro optic tunable filters. This work targets towards optimizing the

properties and performance of multilayered Mg – based guided mode resonance filter in

terms of resonance efficiency, sidebands effects, and resonance bandwidths, while

additional layers are introduced. The effects of varying the parameters of the grating such

as its thickness, period, and duty cycle, also the effects of changing the number, location

and thicknesses of the additional layers were investigated and optimized by means of

rigorous – coupled wave analysis (RCWA). The most interesting things about the

multilayered Mg – based guided mode resonance filter are its simplicity, compactness,

integration of other sensing systems, and possibility of growth of all layers in one process.

The spectral sensitivity of the filter for both the transverse electric (TE) and transverse

magnetic (TM) polarization modes were estimated and compared for various regions in the

spectrum. A spectral response characterized with narrow bandwidth and low reflection

sidebands was realized by carrying out extensive parameter search and optimization. The

most interesting thing in our results is that one can design a GMR filter with a specific

wavelength and linewidth, also the existence of the buffer layer gives us more ability to

control the thickness of the grating and the waveguide.

KEYWORDS - Diffractive optics, guided waves, filters, spectral sensitivity



OBTAINING THE EFFECTS OF THE VOLTAGE SAG DURATION ON

FLUORESCENT LAMPS

MEHLIKA SENGUL1


ABSTRACT

Most of the studies about the effects of the voltage sags on sensitive loads are about

getting tolerance curves of such equipments and comparing them with the standards. This

study helps to fill information gap about the effects of voltage recovery after the voltage

sags on the fluorescent lamps by providing detailed analysis. Varied voltage sag

magnitudes and durations were applied to the test circuits by using California Instruments

4500LX programmable power source. Tests were performed on selected equipments such

that voltage sags tend not to damage them. Data sets were recorded by Scopemeter 199-C.

Obtained data sets were analysed by using MATLAB programme. The test results showed

that at the instant of the voltage recovery, there may be a current spike in the illuminating

system. Evaluating the test results, some special conditions have been identified that cause

the inrush current to reach its maximum level.

KEYWORDS - Power quality, Voltage sag, fluorescent lamp, Inrush current



A MICROSTRIP PATCH ANTENNA DESIGN TO DETERMINE THE LEVEL OF

OCCLUSION IN HEART VEIN

RABIA TOP1, SEYFETTIN SINAN GULTEKIN1, DILEK UZER1


ABSTRACT

To detect heart failures, several solutions based on antenna are produced. In this work, a

microstrip patch antenna structure is designed towards detecting occlusion rates of the

heart vessel. As operating frequency and substrate, 5.8 GHz and RT/Duroid 5880 are

selected. Proposed antenna and modelled heart vein and its occlusion are simulated in

Ansoft HFSS program. When modelled occlusion radius is changed, different

electromagnetic field values are obtained as based on each radius. Differences in these

values is a basic study on determining level of occlusion in heart vein. Accordingly, It is

thought that the proposed antenna structure can be used to determine the occlusion rates of

the heart vessel.

KEYWORDS - Microstrip antenna, modeling heart vein, occlusion, HFSS



IMPROVEMENT OF A SOLENOID FOR ELECTROMECHANICAL VALVE

EMV SYSTEMS

NIDA BIRGUL1

1 Baskent University, Turkey

ABSTRACT

Electromechanical valve (EMV) systems perform valve timing entirely independently of

the crankshaft position. It provides multiple control of the valves in one cycle. It has a

great potential to improve the engine performance on one hand and to provide optimum

fuel consumption and to minimize emissions on the other hand. In this study, it was aimed

to improve the solenoid used in an EMV system with 12V operating potential and to

perform static and dynamic analysis of it. In the initial design, the coil resistance is 1.2 Ω,

and the maximum solenoid force at 10 A current value is measured as 76 N.

KEYWORDS - Valve, Electromechanical Valve, Variable, Mechatronics, Internal

Combustion Engines. Solenoid.



UNDERSTANDING OF BIPOLAR DISORDER ON RAT BEHAVIORS VIA

NONLINEAR STATE ESTIMATION

AYSEN CETIN KARDESLER1, MERIC CETIN1, MEHMET DOGAN ELBI1

1 Pamukkale University, Turkey

ABSTRACT

Bipolar disorder is a disease that is mostly encountered in medical literature. This

psychiatric illness with hypomanic and depressive mood can be modelled as limit cycle

oscillator. Artificial sweetener like Aspartame (ASP), which has neurologic and behavioral

effects on animals, induces the bipolar disorder-like behavioral properties on rats. In this

study, mood changes of rats exposed to intraperitoneal ASP injections in 8 different

sessions at following dosages: 50 mg/kg/day, 100 mg/kg/day and 200 mg/kg/day have

been investigated. Open Field Test (OFT) is implemented to determine the bipolar

disorder behaviors of animals. In order to interpret the chaotic moods of bipolar disorder a

sliding-mode observer (SMO) has been designed. The advantage of the estimation

unmeasurable moods is that provides us the knowledge of which bipolar disorder mood

the rat is in and helps to understand the future moods of rats. According to the observation

results after the ASP injection, considerable increases were observed in grooming values

as OFT results were monitored, while considerable decreases were observed in line

crossing and rearing activities after ASP injections.

KEYWORDS - Rat, bipolar disorder, animal behaviors, nonlinear state estimation, limit

cycle oscillator, sliding mode observer



PETRI NETS MODELLING OF SMART AGRICULTURAL SYSTEM

HANIFE APAYDIN OZKAN1


ABSTRACT

Technological developments that affect human life make people's life easier. Recently, we

have started use smart systems in our homes and factories. In this work, a smart

agricultural scheme (CPS-cyber physical system) is proposed and its adaptation to farmer

life is presented. Firstly, working principles of sensors for agricultural applications and the

proposed smart farming system are explained and compared. Then, Petri net model of the

sensors is developed in order to enable a good general view of the operation of these

sensors and their adaptation to smart farming system. Moreover, for monitoring and

controlling the sensors and actuators, a simulation program is prepared. Simulation results

show that the proposed scheme makes it easier for farmers to understand smart farming

systems.

KEYWORDS - Petri nets, smart agriculture, sensors, simulation



PHOTOELECTRICAL PARAMETER CALCULATION OF METAL ORGANIC

INORGANIC PHOTO DIODE USING AN APPROACH BASED MABC

ALGORITHM

HULYA DOGAN1, SERDAR KOCKANAT1, YAVUZ TURKAY1, MURAT

CIKAN1, DERYA BETUL UNSAL CELIMLI1


ABSTRACT

In this study, we have proposed a metaheuristic technique based on modified artificial bee

colony (MABC) algorithm to determine the electrical parameters of photodiode device.

Firstly, the deposition of tris (2,2’-bipyridine) ruthenium (II)-complex thin film on p-Si

substrate by spin coating technique was presented. The Ru (II) was directly formed by

casting 5 mL of the 0.001 M Ru (II) solution in alcohol on the front surface of the p-Si

substrate and then spin coating the solution onto the Si substrate, which was rotated at

1000 rpm/min for 30 s and then dried for 60 min at room temperature in order to remove

the solvent. The electrical and photovoltaic properties of the Al /tris (2,2’-bipyridine) Ru

(II) complex/p-Si were investigated both under dark and under 100 mW/cm2 illumination

level, using forward and reverse bias current-voltage (I-V) room temperature. Secondly,

modified artificial bee colony (MABC) algorithm has been employed to calculate the

electrical parameters and series resistance (Rs), ideality factor (n) and barrier height (ϕb)

are identified 13.80 Ω, 1.44 and 0.84 eV, respectively. Compared to other conventional

methods that find Rs of 14.43 Ω, n of 1.42 and ϕb of 0.84 eV, proposed method based

MABC algorithm is very efficient and quickly technique to estimate the electrical

parameters of the fabricated photodiode. Also, the fabricated device shows a photovoltaic

behavior with the maximum open circuit voltage (Voc) of 0.419 V and short circuit

current (Isc) of 32.9 μA under 100 mW/cm2 light illumination. **Acknowledgement: This

work was supported by Scientific Research Projects Coordination Unit of the Cumhuriyet

University (CUBAP) (Project Number: M-649). **Corresponding Author (Serdar

KOCKANAT, [email protected])

KEYWORDS - Photodiode, modified artificial bee colony, optimization, electrical

parameter.



ELECTRICAL PARAMETER DETERMINATION OF AL P SI SCHOTTKY

DIODE FROM I V AND C V CURVES USING MABC BASED METHOD

SERDAR KOCKANAT1, HULYA DOGAN1, YAVUZ TURKAY1, MURAT

CIKAN1, DERYA BETUL UNSAL CELIMLI1


ABSTRACT

In previous works, we have successively proposed a calculation method based on modified

artificial bee colony algorithm (MABC) to perform characteristic parameters of the

fabricated electronic device from its current-voltage (I-V) characteristics. Different from

these studies, now we have tried the parameter extraction method based artificial bee

colony algorithm on the capacitance-voltage (C-V) characteristics. For this purpose, a

metal-semiconductor-Al/p-Si Schottky diode has been produced and measured its

capacitance-voltage (C-V) and current-voltage (I-V) characteristics under dark and room

temperature conditions. From the I-V characteristic of metal-semiconductor-Al/p-Si

Schottky diode, series resistance (Rs) of 8.94 Ω, ideality factor (n) of 1.24 and barrier

height (ϕb) of 0.84 eV are calculated using MABC method. Also, from the C-V

characteristic of metal-semiconductor-Al/p-Si Schottky diode, carrier concentration (Na)

of 2.65x1013 cm-3 and diffusion potential (VD) of 0.61 V have been extracted using

MABC method. These results show that MABC method works successively and flexible

in parameter extraction problem of semiconductor device for both I-V and C-V

characteristics. **Acknowledgement: This work was supported by Scientific Research

Projects Coordination Unit of the Cumhuriyet University (CUBAP) (Project Number: M-

649). **Corresponding Author: Hülya DOĞAN ([email protected])

KEYWORDS - Schottky diode, artificial bee colony, parameter extraction, metal-

semiconductor.



LEACH AND AHP BASED CLUSTERING ALGORITHM FOR WIRELESS

SENSOR NETWORKS

GOKCE HACIOGLU1, ERHAN SESLI1, VOLKAN GANGAL1, VOLKAN AYDIN1

1 Karadeniz Teknik Universitesi, Turkey

ABSTRACT

— LEACH is a popular clustering algorithm for Wireless Sensor Networks (WSN).

LEACH can be applied with minimum system resources in a distributed manner. The

LEACH algorithm selects cluster heads randomly with a certain probability. The

probability of being cluster head for a node depends on how many times it is selected as a

cluster head. Therefore, remaining energy of nodes in network may not be homogeneous

across the network. The proposed method uses Analytic Hierarchy Process (AHP) to

determine the probability of being elected as cluster head in a distributed manner.

According to simulation results the proposed method can extend network lifetime.

KEYWORDS - Wireless Sensor Networks, LEACH, AHP, Clustering, WSN



IMPROVEMENT OF AN ELECTROMECHANICAL VALVE EMV SYSTEM

DYNAMIC ANALYSIS AND DETERMINATION OF WORKING LIMITS AT

DIFRENT LIFTINGS

NIDA BIRGUL1

1 Baskent University, Turkey

ABSTRACT

In four-cycle internal combustion engines, the valves control the gas inlet and outlet

events. Electromechanical valve (EMV) systems perform the required valve timing

independently of the crankshaft position. With this feature, EMV systems have a great

potential for increasing engine performance, ensuring optimum fuel consumption and

minimizing emissions. The intent of this study was to improve an EMV system, which has

12V supply potential, and to investigate dynamic performance at different lifting valve

operations and then to determine the support limits of an internal combustion engine.

KEYWORDS - Valve, Electromechanical Valve, Variable, Mechatronics, Internal

Combustion Engines, Solenoid.



A COMPARISON OF PSO GA AND NELDER MEAD OPTIMIZATION

ALGORITHMS ON BUSBAR TRANSMISSION LINE PARAMETER

EXTRACTION

ZEYNEP HASIRCI1, MEHMET OZTURK1, RIFAT BENVENISTE2, HAYDAR KAYA1,

ISMAIL HAKKI CAVDAR1

1 Karadeniz Technical University, Turkey; 2 Avrasya University, Turkey

ABSTRACT

In the smart grid concept, monitoring energy from the point of production to consumption,

loss-efficiency analysis, and increasing energy quality are some of the main targets. All of

these can exist by integrating communications infrastructure to the traditional grid which

enable to provide data acquisition, protection, and control qualifications. Due to the power

grid is already distributed over a large area, utilization of the existing infrastructure as a

communication medium - power line communication - will be a wise choice. Therefore, it

is necessary to know the behavior of each element in the grid as a communication

medium. Transmission line characteristics of cables are a well-established subject with the

help of recent power line communication studies. On the other hand, busbar distribution

systems which are used to carry electrical energy as a part of low voltage grid have

differed from cables at installation requirements, physical and electrical properties. Thus,

modeling busbar distribution system as a transmission line is a necessity for the sake of

completeness in this research area. At this point, the modeling approach is a powerful

method to obtain the primary line parameters of a transmission line by using an

optimization algorithm according to some objective functions that utilize measurements

such as S-parameters. In this study, Nelder-Mead, genetic and particle swarm optimization

algorithms which are commonly preferred methods in the literature are compared for the

aspect of obtaining accurate busbar transmission line parameters in 1-50 MHz frequency

band. The comparisons are conducted under same conditions such as parameter search

range and objective function definition at the point of convergence rate, runtime, number

of iterations, and the number of function evaluations. According to the numerical results,

particle swarm optimization method is the most efficient algorithm for the extraction of

busbar transmission line parameters.

KEYWORDS - Busbar, genetic algorithm, Nelder-Mead simplex, particle swarm

optimization, parameter extraction, power line communication, smart grid, S-parameters,

transmission line modelling



COMPARATIVE PERFORMANCE EVALUATION OF MULTI SOGI PLL FOR

MINIMUM SETTLING TIME AND LOW UNIT VECTOR DISTORTION UNDER

DISTORTED CONDITIONS

TUGCE DEMIRDELEN1, TAYLAN OZDEMIR2, MEHMET TUMAY1

1 Adana Science and Technology University, Turkey; 2 Cukurova University, Turkey

ABSTRACT

Phase-locked loops (PLLs) are necessary in grid connected systems to obtain information

about the frequency, amplitude and phase of the grid voltage. Inverter systems rely on

accurate and fast detection of the phase angle, the amplitude and the frequency of the

utility voltage to guarantee the correct generation of the reference signals. This is also

required by the relevant industrial standards affecting distributed generation systems.

Recently, several advanced phase-locked loop (PLL) techniques have been proposed for

single-phase applications. This paper presents performance investigation a multi phase-

locked-loop (PLL) method which generates the orthogonal voltage system using a

structure based on a second order generalized integrator (SOGI). The advantages of this

PLL include simple implementation independent of the grid frequency and avoidance of

filtering delays due to its resonance at the fundamental frequency. The solutions for the

discrete implementation of the structure generating the orthogonal system are presented.

Simulation results validate the effectiveness of the proposed method under distorted

conditions by comparing Dual-SOGI and SOGI PLLs performances.

KEYWORDS - Phase lock loop, SOGI-PLL, PSCAD, second order generalized integrator,

simulation



SLOW LIGHT IN PHOTONIC CRYSTAL WAVEGUIDE

MAACHE MOHSSIN1, HOCINI ABDESSELAM1, KHEDROUCHE DJAMEL1

1 Mohamed Boudiaf University M Sila, Algeria

ABSTRACT

Recently, slow light with low group velocity attracted wide attention because it is regarded

as a promising technology for future all-optical communication networks. It can be used

for the enhancement of light-matter interaction it also offers the possibility for

miniaturization sensors with high sensitivity, slow light occurs due to large first order

dispersion arising from the resonance of light with a material or structure. Photonic crystal

waveguide is especially attractive for generating slow light in wideband with high group

index and low group velocity dispersion. Slow light in photonic crystal waveguides can be

used for a wide range of applications, such as delay lines or buffersand enhanced light-

matter interaction. In our work, we had manipulated the geometry of a modified photonic

crystal waveguide for generating a slow light focusing on reducing the unwanted

dispersion by engineering the dispersion curve with the aim of achieving a constant group

index over a broad wavelength range, the modified photonic crystal is optimized through

simulation performed by a 2-D PWE method of the RSoft software.

KEYWORDS - Slow light, Photonic crystal, Waveguide



MOVE YOUR WHEELCHAIR WITH YOUR EYES

GOKCEN CETINEL1, ZAFER TIRYAKI1, ENES KUZU1, MELTEM

MIILIGUNEY1, SEVDA GUL1


ABSTRACT

In the proposed study, our goal is to move paralyzed people with their eyes. For this

purpose, we use their Electrooculogram (EOG) signals obtained from EOG goggles

completely designed by the authors. Through designed EOG goggles, vertical-horizontal

eye movements and voluntary blink detection are verified by using 5 Ag-AgCl electrodes

located around the eyes. EOG signals utilized to control wheelchair motion by applying

signal processing techniques. The main steps of signal processing phase are preprocessing,

maximum-minimum value detection and classification, respectively. At first,

preprocessing step is used to amplify and smooth EOG signals. In maximum-minimum

value detection we obtain maximum and minimum voltage levels of the eye movements.

Furthermore, we determine the peak time of blink to distinguish voluntary blinks from

involuntary blinks. Finally, at classification step k-Nearest Neighboring (k-NN) technique

is applied to separate eye movement signals from each other. Several computer

simulations are performed to show the effectiveness of the proposed EOG based

wheelchair control system. According to the results, proposed system can communicate

paralyzed people with their wheelchair and by this way they will be able to move by their

selves.

KEYWORDS - Electrooculugram, Signal Processing, Classification, Control



ELECTROMAGNETIC DESIGN OF LINE START AXIAL FLUX PERMANENT

MAGNET SYNCHRONOUS MOTORS

MUSTAFA EKER1, MEHMET AKAR1

1 Gaziosmapasa University, Turkey

ABSTRACT

Compared to other motors, the power, the moment, and the efficiency produced by the unit

volume in the axial-flux permanent magnet synchronous motors is higher. Additionally it

also has the advantages such as using different stator and rotor topologies and the air gap

is planar and adjustable. For this reason, this issue is one of the most studied topics by the

motor manufacturers and scientists in recent years. However, these motors cannot be

connected directly to the network and high-performance drive is required for operating. In

the presented study, with the novel rotor structure the axial flux permanent magnet

synchronous motor which can be directly connected to the network is designed and the

obtained results are presented.

KEYWORDS - Line start, axial flux PMSM, electromagnetic design



WIRELESS POWER TRANSFER VIA AIR GAP TRANSFORMER

MUSTAFA OZSOY1, ZAFER DOGAN1

1 Gaziosmanpasa University, Turkey

ABSTRACT

Wireless power transfer systems have become more widespread in recent years. In this

study wireless power system structure is researched. And electromagnetic analysis of

fundamental system is analyzed by FEM. In wireless power transmission systems (WPT),

spiral shaped coils are commonly used. At this method two coils can be used. One of them

works as a transmitter and the other works as a receiver coil. In this study, unlike this

extensive method, a transformer approach is been implemented. Therefore transmitter and

receiver coils are applied to primary and secondary windings. Core structure is divided

into two parts. One of the parts is used as a transmitter side core and the other is

considered as a receiver side core. Steel laminations are used as a core material. In

addition to that, at WPT systems with a high frequency signals is needed for higher

efficiencies. So, to realize that ferrite core which provides higher frequency range than

steel core, is been applied to the design too. As a side effect of higher frequency, skin

effect that can be come across at primary and secondary windings. Therefore these

situations considered, and conducting materials has been chosen according to that. Primary

and secondary side cores’ geometric shapes determined and conductors applied to this

cores by entering number of turns. Analyzes are done at different distances and input

output powers are obtained at no-load and full-load states. According to that, efficiency up

to 20% is achieved. Distance arranged different values up to a 100 mm. At the beginning

an input signal arranged to a 220 V 50 Hz, after that this values arranged different values,

and results are obtained. When input signal arranged to Turkey’s network values, at 5mm

distance, about 145W transfer obtained. For more efficient transfer, at the external circuit

passive circuit elements can be applied. At the suggested method, proposed design is a

new approach for wireless power transmission.

KEYWORDS - Wireless power transfer, transformer, Air core



I V AND C V CHARACTERISTICS OF P SI NZNO NRS PHOTODIODES

DEPOSITED BY MW CBD

YASEMIN CAGLAR1, KAMURAN GORGUN2, MUJDAT CAGLAR1, SALIHA

ILICAN1

1 Anadolu University, Turkey; 2 Eskisehir Osmangazi Univeristy, Turkey

ABSTRACT

In modern materials science, nanostructured ZnO films are versatile material to make in

numerous novel applications, owing to their several unique features, including the wide

band gap (3.37 eV), good transparency and electrical conductivity, excellent electron

mobility, and low thermal expansion. So, they have been extensively studied and applied

in many applications, such as sensors, light-emitting diodes, thin film transistors,

photodetectors and solar cells. Microwave-assisted chemical bath deposition (MW-CBD)

was used to deposit zinc oxide nanorods (ZnO-NRs) films by using different microwave

irradiation time (5, 8 and 10 min). The films exhibit a good crystallinity having a

hexagonal wurtzite phase formation. An increase in the microwave irradiation time

improved the structural quality of ZnO-NRs. ZnO-NRs which grew perpendicular to the

surface were detected by field emission scanning electron microscope (FESEM) and it was

identified the effect of microwave irradiation time on the densities and orientations of

nanorods. The p-Si/n-ZnO-NRs heterojunction photodiodes were fabricated. The current-

voltage (I-V) and capacitance-voltage (C-V) measurements were taken with KEITHLEY

4200SCS and SIGNATONE Probe Station. The intensity of illumination measurements

was carried out using a solar simulator (SCIENCETECH). The I-V of these photodiodes

were investigated under dark and different illumination intensity. An increase in the

reverse current with increasing illumination intensity confirmed that the fabricated

photodiodes exhibited a photoconducting behavior. The diode parameters of the

photodiodes determined using different methods. In addition, the existence of interface

states and series resistance values of the photodiodes were determined from C-V

measurements. The obtained all results suggest that the p-Si/n-ZnO-NRs photodiodes can

be used as a sensor in the optoelectronic applications. Acknowledgement: This work was

supported by Anadolu University Commission of Scientific Research Project under Grant

No: 1402F055.

KEYWORDS - ZnO, nanorod, MW-CBD, photodiode, Norde function.



EFFECT OF MN DOPANT ON THE STRUCTURAL AND MORPHOLOGICAL

PROPERTIES OF ZNO FILMS BY SPIN COATING

SALIHA ILICAN1, YASEMIN CAGLAR1, MUJDAT CAGLAR1


ABSTRACT

ZnO films have attracted a great interest due to their unique properties and potential

applications in optical devices such as blue, violet and UV- light emitting diodes (LEDs)

and laser diodes (LDs). The pure or doped ZnO films can be deposited by several

methods. The sol gel method is one of the most commonly used method for preparation of

transparent and conducting oxides owing to its simplicity, safety, non-vacuum system of

deposition and hence inexpensive method for large area coatings. In this work, undoped

and manganese doped ZnO films have been prepared onto n-Si substrate by facile sol gel

method using spin coating technique. The precursor sol was spin coated with a speed of

3000 rpm for 30 s onto the n-Si. Then the films were dried at 300 oC for 10 min. This

coating/drying procedure was repeated for ten times before the film was inserted into a

tube furnace and annealed at 700 oC in air for 2 h. The effects of Mn doping level on the

structural, morphological and optical properties of Mn doped ZnO films were investigated

by means of XRD, FESEM and UV spectrophotemeter, respectively. XRD analysis

revealed that the films are polycrystalline in nature and consist of single phase ZnO

(zincite phase) with wurtzite structure. The crystallinity was changed with increasing Mn

concentration. The structural parameters are determined. The diffuse reflectance spectra of

the Mn doped ZnO films deposited on n-Si substrate were measured and the optical band

gaps of them were determined using the reflectance spectra by means of Kubelka-Munk

function.

KEYWORDS - Mn doped ZnO, Sol gel, Microstructure, Structural parameters, Kubelka-

Munk function



COMPARATIVE STUDY ON THE STRUCTURAL MICROSTRUCTURAL AND

OPTICAL CHARACTERISTICS OF DIFFERENT DOPED ZNO

NANOPOWDERS BY MW HTS

MUJDAT CAGLAR1, KAMURAN GORGUN2, SEVAL AKSOY3, SALIHA

ILICAN1, YASEMIN CAGLAR1

1 Anadolu University, Turkey; 2 Eskisehir Osmangazi University, Turkey; 3 Sinop

University, Turkey

ABSTRACT

Nanosized particles of semiconductor materials have gained much more interest in recent

years due to their desirable properties and applications in different areas. ZnO is n-type

semiconductor with wide band gap of 3.37 eV and large exciton binding energy of 60 meV

at room temperature. Undoped and doped ZnO can be deposited using various techniques.

The microwave assisted hydrothermal synthesis (MW-HTS) is one of the most important

methods to prepare the nanopowder due to simple and cost effective technique. Compared

with conventional methods, microwave irradiation requires very short reaction times,

produces small inorganic particles with narrow particle size distributions, and consumes

low energy. In this study, undoped and different doped (Iron (Fe), Copper (Cu) and

Magnesium (Mg); 1%) ZnO nanopowders were synthesized via MW-HTS. The effect of

Fe, Cu and Mg content on the structural, morphological and optical properties of ZnO

nanopowders was investigated. To investigate the crystalline structure and the orientation

of the ZnO nanopowders, XRD patterns were used. All of the nanopowders have

polycrystalline in nature with (101) orientation. The lattice parameters and texture

coefficient values of the nanopowders were determined. Field emission scanning electron

microscope (FESEM) was used to analyze the surface morphology of all the ZnO

nanopowders. The molecular bonding characterization of these nanopowders was also

determined by Fourier transform infrared spectroscopy. Acknowledgement: This work was

supported by Anadolu University Commission of Scientific Research Project under Grant

No. 1605F411.

KEYWORDS - Doped ZnO, Nanopowder, MW-HTS, FTIR.



MEASUREMENT OF ENERGY CONSUMPTION OF WSN FOR INDOOR

ENVIRONMENT

SEYMA AYMAZ1, GOKCE HACIOGLU1


ABSTRACT

The energy consumption of indoor Wireless Sensor Networks (WSN) were measured. Ti

CC2538 modules were used as WSN nodes and they were programmed via Contiki OS.

Measurements were taken directly from the module via Labview and DAQ card. The

network used Low Energy Adaptive Clustering Hierarchy (Leach) protocol which is a very

popular networking protocol for WSN. The nodes in the network were prevented to go in

sleep mode. Therefore measurements showed transmit and receive mode energy

consumption. On the other hand cluster head election period and transmission period of

sensed data selected as short as possible to make measurements in a short time. The

proposed measurement system can be used evaluate any new networking protocol more

realistically.

KEYWORDS - ENERGY CONSUMPTION, WSN, CONTIKI, COOJA, LEACH



MULTILAYER PLASMONIC ABSORBER BASED METAMATERIAL FOR

REFRACTIVE INDEX SENSING AND SURFACE ENHANCED SPECTROSCOPY

APPLICATIONS

ERDEM ASLAN1, EKIN ASLAN1, OMER GALIP SARACOGLU1, MUSTAFA

TURKMEN1


ABSTRACT

Metamaterial plasmonic absorbers have promising properties to be used in biomolecular

sensing applications such as refractometric biosensing and surface enhanced spectroscopy.

Although their inherent advantages, novel plasmonic perfect absorbers with additional

features are highly essential for real-world sensing applications. In this context, we present

the design and characterization of a subwavelength plasmonic absorber array with crucial

properties including the multi-band absorption, high near-field enhancement, large number

of hot-spots and high sensitivity. We reveal the optical characteristics of the structure via

finite difference time domain simulations. In order to investigate the physical origin of

absorption resonances we investigate near-field properties of the structure. For the purpose

of exploring the sensor-potential of the proposed structure, we study on the refractive

index sensing performance. Finally, we show the utilization of the proposed device in

surface enhanced spectroscopy applications through simulations and present the results for

the detection of the vibrational mode of a thin absorbing film. The results show that the

proposed device can be utilized in the plasmonic sensing applications.

KEYWORDS - Plasmonics, sensors, subwavelength structures, nanostructures, surface

enhanced spectroscopy.



REVIEW OF REDUCTION AND SIMPLIFICATION METHODS FOR

TRANSMISSION NETWORKS

ALI AJDER1, KADIR DOGANSAHIN1, SELIM AY1


ABSTRACT

The increasing penetration of renewable energy sources, integration of decentralized

generation and emerging different consumption profiles on demand-side expand the

complexity of operation and control of power systems. Besides, the raising economic

concerns and competition in the market have begun play an important role in the

development of power systems, as efficiency trend, generation portfolio management. For

these reasons, the analysis used in system design and planning become more complex,

detailed and precise than ever. Transmission systems are the backbone of the power

systems, where essential analysis are performed, such as economic dispatch, power flow

analysis, contingency analysis. These analysis come with very high computational efforts

for large power systems. There are proposed approaches in the literature for simplifying

large scale transmission systems into reduced systems, so the analysis as well. These

approaches are titled as reduction and simplification methods. These methods differentiate

from each other, according to their accuracies and computational efforts, in the scope of

applicability to the studies, which can be dynamic, steady-state or economic analysis. In

this study, these methods are reviewed, discussed and compared with each other. The

purpose of the study is to present a brief knowledge and guidance about reduction and

simplification methods.

KEYWORDS - Model reduction, System modeling, Transmission lines



DESIGN AND APPLICATION OF PRINTED CIRCUIT TRANSFER MACHINE

MEHMET RAHMI CANAL1, MEVLUT AHMET HAS2, FECIR DURAN1, MERAL

OZARSLAN YATAK1

1 Gazi University, Turkey; 2 Selcuk University, Turkey

ABSTRACT

In this study, an application of a microcontroller controlled printed circuit transfer

machine is presented. This study aimed to speed, simplify and lower faults while

preparation of printed circuit. This machine can be used in laboratory of vocational

schools, public training courses, people try with hobb, people try with hobby electronics

and firms study on research and application. The production of the circuit is easy and fast

with minimum cost and minimum fault in a short duration. The system control was

realized with PIC16F877 for easy programming and low cost. The software was done with

PICC. This machine contains two stages as mechanical and electrical stages. As a result, a

user friendly transfer machine was implemented with a low cost usable microcontroller

with debugged process and high speed.

KEYWORDS - PRINTED CIRCUIT TRANSFER MACHINE, microcontroller



DETERMINATION AND ANALYSIS OF ELECTRIC FIELD STRENGTH

LEVELS IN SAMSUN CITY CENTER

CETIN KURNAZ1, BEGUM KORUNUR ENGIZ1


ABSTRACT

In this study, in order to examine and evaluate the change of electromagnetic radiation

(EMR), electric field strengths (E) were measured at eight different times in Samsun,

Turkey. The short term measurements were collected from 67 different locations using

PMM-8053 EMR meter. In the measurements, the maximum electric field strength (Emax)

and the average electric field strength (Eavg) were recorded. The short term measurement

results show that the E originating from the base stations change significantly depending

on the measurement location and time, the highest Emax is 6.32 V/m and the highest Eavg

is 3.82 V/m. The mean Eavg in 2017 has increased by 24% compared to that of 2014. In

order to observe Es within a day; long term (24 hour) E measurement was conducted at the

location where the highest Eavg value was recorded. The maximum recorded E is 3.75

V/m, the mean E is 2.76 V/m while the standard deviation is 0.40 V/m for long term

measurement. Analysis of 24 hour measurement results for four specific time interval such

as morning, afternoon, evening and night show that the mean E value for morning is 2.53

V/m, while 3.03 V/m, 3.04 V/m and 2.42 V/m for afternoon, evening and night

respectively. Evening E level increases by 25.6 % compared to night E level. At the end of

the study daily variations of E value was examined and an empirical model were proposed

using curve fitting method. With the use of this model the E in the medium can be

predicted with an accuracy of 90%.

KEYWORDS - Electric field strength; Electromagnetic (EM) measurement; EM pollution;

Base station; Statistical analysis.



ILLUMINANT DIRECTION ESTIMATION METHODS

ABDULLAH TAHA ARSLAN1, EROL SEKE1


ABSTRACT

Estimation of light sources and illumination conditions in digital images is an essential

part of many image processing applications. This is particularly important in 3D face

reconstruction from one or more face images. The literature provides several established

methods to recover a single lighting vector. Problem is reduced into two parts as the tilt

and slant angles of the illuminant, in general. In this paper, we provide an experimental

analysis of the common methods and shown that given algorithms predict the former at a

high success rate while this is not the case for the latter. Outcomes of the conducted

experiments are discussed and methods for improvements are proposed.

KEYWORDS - illuminant direction estimation, shape from shading, surface reconstructing



DESIGNING A 2 D PAINTING ROBOT

HUSEYIN DEDE1, HUSEYIN EMRE GUNER1, KADIR ERGUNER1, BEKIR

KARATAS1, OMER AKBAS1, H METIN ERTUNC1


ABSTRACT

Today, 2-D painting robots are the new technological products in industry. Many

technological companies produced and developed these systems for industrial applications

and home users. The 2D painting robot is a computer-aided painting robot that paints a

two-dimensional surface due to the direction of the users. The aim of this project is that to

transfer real images to the computer and two dimensional painting on a A4 paper. Firstly,

painting surface is scanned using basic image processing algorithms and transferred to the

computer. Then, picture on the surface is analyzed using image processing algorithms.

Picture coordinates is obtained and a graphical user interface program shows the

coordinates of the pictures and movements of the motors to the users. After that picture

will be painted to paper using the H-bot mechanical system. Electronic system is designed

and controlled using Arduino environment. There are many industrial systems are in the

market. Many commercial products are so expensive and painting just one color. The

difference of this system than the others is that this system will be more accurate low cost

to manufacture and painting three separate color.

KEYWORDS - Painting robot, Arduino, 2-D



A NEW MODEL FOR MEASURING DIMENSIONAL ACCURACY BRIDGING

AND OVERHANG PERFORMANCE OF FDM 3D PRINTERS

SERDAR KUCUK1, NURETTIN GUNES1


ABSTRACT

In recent years, FDM-based 3D printers has been increasingly used in the industry. At the

beginning, 3D printer technology was used in prototype and mold production. However

ongoing years this technology has started to be used for final product manufacturing. With

the help of 3D printers, manufacturing cost of the some end-products significantly

decreased. The all production process for 3D printers is completed according to the files

created on the computer. This significantly decreased the manufacturing cost of the some

end-products. This new technology also results in emerging different models for

measuring the performance of raw materials used in FDM 3D printers. In this paper, a new

model is proposed to measure the dimensional accuracy, bridging and overhang

performance of 3D printers.

KEYWORDS - Benchmarking object; bridging performance; dimensional accuracy;

overhang performance; additive manufacturing; fused deposition modeling; 3d printer;



EVALUATING THE PERFORMANCE OF DELTA TYPE 3D PRINTER

CONSIDERING THE INTERNATIONAL TOLERANCE VALUE

SERDAR KUCUK1, NURETTIN GUNES1


ABSTRACT

Nowadays 3D Printers are extensively used in many fields of industry. It has become

popular as a new production model. Several different models of 3D printers have been

developed in a very short time. Today 80% of 3D printers are manufactured in Cartesian

type. The other 3D printers are manufactured in Delta and SCARA and Polar types. These

3D printers are more sophisticated and experimental than the Cartesian. In this study, a

Delta type is developed and the performance of this Delta type 3D printer is evaluated by

considering the IT (international tolerance) value. The developed 3D printer is also

compared with previous research results.

KEYWORDS - 3D Delta Printer; performance; international tolerance;



AN EASY TO USE JOYSTICK CONTROLLER DESIGN FOR INDUSTRIAL

ROBOTS

SINAN OZCAN1, MUSTAFA COBAN1, GOKHAN GELEN1

1 Bursa Technical University, Turkey;

ABSTRACT

Industrial robots are widely used in manufacturing systems to perform various tasks.

Robots have a control box named as teach pendant that is used to programming robot

motions and moving robot to desired positions. The robot can be set to teach mode and the

current position can be got by teach pendant. The movement of robot is provided by using

the keys on the teach pendant. It is not easy to perform a robot by using keys. It requires

expertness. This paper proposes to design a 3-axis joystick to manipulate the robot instead

of teach pendant. A 3-axis Joystick with button, an Arduino Uno and an ethernet shield are

used. The Ethernet protocol is used to connection between the joystick and the robot

controller. The obtained joystick controller manipulates the robot into three axes and

provides gripper control. The manipulation of a robot via 3D joystick is easier than by

using teach pendant.

KEYWORDS - Industrial Robots, Control, Teach pendant, joystick, Controller Design



ROBOT COLLISION AVOIDANCE CONTROL AND MANAGEMENT METHOD

FOR INDUSTRIAL ROBOTS IN AUTOMOTIVE INDUSTRY

CENGIZ DENIZ1, MUSTAFA CAKIR2

1 Ford Otosan, Turkey; 2 University of Kocaeli, Turkey

ABSTRACT

In this study, a newly developed robot collision avoidance control and management

method using kinematic equations is presented for industrial robots in the automotive

industry. Industrial robot collision control and management is one of the most significant

points in multi-robot production cell in all of the automotive industry. If robot collisions

are not configured with the right methodology, then they can impact production cycle-

times, cause robot crashes and breakdowns, spare part consumption and even result in

severe injuries. Typically, two types of methods are used for industrial robot collisions

programming; the first method is offline programming before robot installation and the

second method is online programming during robot commissioning. However, these two

methods are not the exact solution for robot collision control and management for multi-

robot production cells. We proposed a new collision management method for modelling

robot kinematic on the PLC (programming logic controller) by getting all axis positions

data from all robots in multi-robot production cell and then calculating the actual dynamic

TCP (tool center point) for each robot. This method makes it possible PLC to know all

robots’ actual and auxiliary TCP positions on the robot body and create dynamic work

space zones to manage all robot collisions in the same cell. Using proposed methodology,

it is possible to improve production cycle time, efficiency, breakdowns of robot and safety

functions in multi robot production cells.

KEYWORDS - Industrial Robot, Collision Avoidance, Multi-Robot Production Cell,

Robot Kinematic, Automotive Industry and PLC System.



INFLUENCES OF ROTOR SLOT DISTANCE ON THE PERFORMANCE OF

SINGLE PHASE INDUCTION MOTOR

AHMET FENERCIOGLU1, MERVE SEN KURT1

1 Amasya University, Turkey

ABSTRACT

In this paper, several single phase induction motors which have different rotor slot

distances are analyzed with 2-D Finite Element Method (FEM). All of the models have

same stator and rotor slot geometry only difference between models are their rotor slot

distance from the rotor core. In analysis, a single phase, 2 pole, 96 W, 220 V with

capacitor start-capacitor run induction motor was used. For each rotor slot model,

electrical performance characteristic included speed, efficiency, current, torque and power

factor (PF), the distribution of the magnetic field density on different location on motors

and current density on all of the rotor slot were examined.

KEYWORDS - Finite element method (FEM), Single phase induction motor (SPIM), rotor

slot distance



MULTI OBJECTIVE H2 HINFINITY CONTROL OF BLADE VIBRATIONS

USING PIEZOELECTRIC ACTUATOR

FEVZI CAKMAK BOLAT1, SELIM SIVRIOGLU1

1 Gebze Technical University, Turkey

ABSTRACT

The aim of the study is to suppress the vibration modes of a blade element with attached

piezoelectric actuator using robust multi-objective control. A modal analysis of the blade

is realized and natural frequencies of the blade are derived analytically, numerically and

experimentally. The analytical model of the blade element for control design is obtained

by using the Euler-Bernoulli beam theory. In a control system, different control

specifications should be satisfied at the same time. Norm based linear control approaches

have some distinct advantages. While Hinfinity control mainly enforces the robust

stability, H2 control improves the transient behavior of the control system. The multi-

objective H2/Hinfinity control combines both design objectives. It is an advantage that

both frequency and time domain specifications are performed in vibration control of the

distributed parameter systems. In this paper, three different H2/Hinfinity controllers are

designed by introducing a multiplicative uncertainty which represents unmodeled high-

frequency dynamics in the control system. The designed controllers are realized in

experiments and performances of the controllers are compared using frequency and time

domain responses.

KEYWORDS - Blade vibration, multi-objective control, robust control, piezoelectric

actuator



DETERMINATION OF WEIGHT COEFFICIENTS USING FUZZY LOGIC IN

WEIGHTED 3D DISCRETE WAVELET TRANSFORM

GUZIN OZMEN1, BURAK YILMAZ2, SERAL OZSEN1

1 Selcuk University, Turkey; 2 Konya Food And Agriculture University, Turkey

ABSTRACT

In this study, a new method for determining the coefficients of weighted 3D discrete

wavelet transform (w-3D-DWT) by using fuzzy logic is proposed. The method generates a

set of coefficients for each sub-band of the 3D-DWT. The fuzzy inference model uses

energy, entropy, and standard deviation values of each sub-band as inputs and generates

coefficients as output. Fuzzy membership functions and rules are determined by analysis

and experiments performed on Brainweb MRI image database. Experiments are evaluated

with MSE, PSNR and SSIM. It was observed that using fuzzy logic inference system

improved the results obtained by w-3D-DWT.

KEYWORDS - Weighted 3D DWT, Fuzzy Logic, 3D MRI, weighted -3D-DW



DUAL BAND PERFECT ABSORBER FOR BIO SENSING APPLICATIONS

WITH CARDINAL POINT STAR SHAPED NANOPARTICLES

AYTAC ONUR1, MUSTAFA TURKMEN1


ABSTRACT

In this study, a novel perfect absorber (PA) array based on cardinal point star-shaped gold

nanoparticles for infrared sensing applications is presented. Proposed PA array has a dual-

band spectral response and the locations of resonances can be adjusted by varying the

geometrical dimensions of the structure. Nearly unity absorbance is obtained from the PA

array for both resonances. Absorbance characteristics of PA array are analyzed by using

finite difference time domain (FDTD) method. High near-field enhancement is achieved

by the interaction of the sharp corners of triangular nanoparticles. Due to the nearly unity

absorbance and high near-field enhancements the proposed dual-band PA array with

adjustable spectral responses can be useful for bio-sensing applications in mid-infrared

regime.

KEYWORDS - Perfect absorber, bio-sensing, nanoparticle, plasmonics, Mid-infrared

rejime



DESIGN AND DEVELOPMENT OF A BIOTELEMETRY SYSTEM FOR

MONITORING BEDRIDDEN PATIENTS

ILKAY TASLI1, ILKER ALI OZKAN1


ABSTRACT

Today, the popularity of mobile and object internet sites has increased with the

development of technology. These developments have facilitated the use and construction

of biotelemetry systems. At this point, the patient can be traced from his home by the

biotelemetry system designed, instead of constantly going to and from the hospital. With

the help of the sensors on the system, the desired values are obtained about the patients

and transferred to the necessary individuals. It is important for nurses and doctors to have

access to patient-related information on an ongoing basis. In this study, an Arduino based

Patient Tracking System was developed for the online monitoring and storage of data

related to patients in need of care. With the developed system, the data received by the

sensors can be monitored in any environment and at any time through an internet

connection. The data belonging to the patient can be instantly evaluated and recorded with

the use of an Android based software. The designed system also has the feasibility to

report and graphically examine the recorded data for the purpose of analyzing the state of

the disease.

KEYWORDS - Patient monitoring, Biotelemetry, Remote monitoring, Healthcare at home



WEB BASED FUZZY EXPERT SYSTEM DESIGN FOR DETECTION OF

RHEUMATOID ARTHRITIS DISEASE SEVERITY

FATIH TARAKCI1, ILKER ALI OZKAN1, SEMA YILMAZ1


ABSTRACT

Rheumatoid Arthritis (RA) disease is an inflammatory disease that can be seen in many

joints at the same time as a chronic illness and of which etiology is unknown and clinical

course of the disease varies from patient to patient. Knowing the severity of RA disease is

very important for the treatment to be carried out. While determining the severity of

disease in RA, clinical laboratory findings are very important as well as anamnesis. In this

study, a fuzzy expert system has been designed to determine the severity of RA disease,

using the laboratory test results of CRP, Sedim, Rheumatoid Factor (RF), and Antinuclear

Antibodies (ANA). The input variables of the designed system, linguistic expressions, and

its rule base have been determined with the aid of a specialist, and a web based software

has been realized to determine the RA disease severity.

KEYWORDS - Rheumatoid Arthritis, Disease Severity, Fuzzy Expert System, Disease

Diagnosis



SEMI AUTOMATIC DETECTION OF STOMACH CANCER BY REGIONAL

SEGMENTATION METHODS

ALI YASAR1, ISMAIL SARITAS1


ABSTRACT

Stomach cancer is the third cancer type in females and second cancer type in males

according to mortality rates. Cancer is the leading cause of death worldwide, creating 8.8

million deaths in 2015. The four most common causes of cancer death are stomach cancer.

Approximately 754,000 people die from stomach cancer every year in the whole world.

The importance of image processing and its use is increasing day by day in today's

medical world. An important step in image processing is the segmentation process.

Diagnosis of stomach cancer can not be performed fully automatically because of the

complicated structure of the stomach. The aim of this study is to study locally for the

presence of the cancerous zone and to suggest a cancerous area using statistical methods in

the determined region. Segmentation will be done by Region Growing and Statistical

Region Merging methods in the region determined by the proposed model and the results

will be obtained by comparing the accuracy of the cancerous zone found in the study with

the zone determined by the specialist doctor.

KEYWORDS - Stamach Cancer, REGIONAL SEGMENTATION METHODS, SEMI-

AUTOMATIC DETECTION



MEASUREMENT OF THE OPTICAL PENETRATION DEPTH IN CHICKEN

BREAST TISSUE

YASAR BARIS DOLUGAN1 , HALIL ARSLAN1 , MUSTAFA ZAHID YILDIZ1 , AYLA

EREN OZDEMIR1 , ALI FURKAN KAMANLI1 , AYSE NUR AY1


ABSTRACT

Photodynamic therapy (PDT) is a cancer treatment requiring a photosensitizing molecule

(drug) and a light source to activate the drug. In PDT, the optical energy received by the

tumorous tissue is crucial for successful treatment. Therefore, penetration depth of the

light in the target tissue is an important parameter for dosimetry calculations. In this study,

the penetration depths of LED light with the wavelengths of 635 nm and 660 nm in

chicken breast tissue have been measured. For this purpose, 13 tissue samples with

different thicknesses between 2.2 mm and 8.2 mm have been prepared. Collimated

transmittances through the samples have been analyzed using Beer-Lambert law to obtain

the penetration depths. It has been found that the optical penetration depths for chicken

breast tissue at 635 nm and 660 nm are 7.1±0.4 mm and 7.6±0.5 mm, respectively.

KEYWORDS - PDT, Optical property, Penetration depth



THE EVALUATION OF TOTAL OXIDANT AND ANTIOXIDANT LEVELS IN

INACTIVE HEPATITIS B PATIENTS WITH DELTA POSITIVE AND

NEGATIVE

ILHAN SABANCILAR1, TUGBA GUR1

1 Health Sciences Institute, Turkey

ABSTRACT

The aim of the present study was to determine the levels of total oxidant and antioxidant in

inactive hepatitis B patients with delta positive and negative. The levels of TAS, TOS and

CRP were detected as the highest, and the level of PON-1 were detected as the lowest in

delta positive patients compared to healthy control and HBsAg positive groups. For the

delta positive and HBsAg positive groups although there were not found statistically

significant (p>0.05) in the activities of TOS, PON-1 and CRP, the levels of TAS, ALT and

AST were found statistically significant (p˂0.05). For the control group and hepatitis B

patients, whereas the levels of TAS and TOS were found statistically significant (p˂0.05);

ALT, AST, PON-1and CRP levels were not found statistically significant (p>0.05). The

patients with hepatitis D and the control group showed statistically value for TOS, ALT

and AST levels (p˂0.05). But the levels of TAS, PON-1 and CRP were not found

significant (p>0.05).

KEYWORDS - HBsAg, Delta positive, TAS, TOS, PON–1, ALT, AST, CRP



MACROMOLECULAR INVESTIGATION OF INTERMOLECULAR

INTERACTIONS DRIVING THE SELF ASSEMBLY OF COLLAGEN BY DLVO

THEORY

ESMA ERYILMAZ1


ABSTRACT

Collagen molecules self-assemble both in vitro and in vivo into structures ranging from

nanoscale to macroscopic length scale with D-periodic spacing. Its self-assembly into

highly organized different level of structures deeply influences a variety of physiological

processes such as cell migration, proliferation, motility, gene expression, and signaling [1,

2, 3]. Due to the comprehensive function of this important physiological event, collagen

self-assembly is extensively used in a variety of biomedical and biotechnological

applications such as tissue engineering [4], cellular processing [5], bio-compatibility

improving, generating templates for nonowire assembly and biofunctionalization of

surfaces [4]. Although both experimental and theoretical understanding of the interaction

forces driving the self-assembly is crucial, most of the studies are experimental. In this

study, we theoretically investigated the main interaction forces accompanying the self-

assembly of collagen molecules onto hydrophilic mica surface with the Derjaguin-Landau-

Verwey-Overbeek (DLVO) theory [6] as a qualitative model at macro-molecular scale.

Application of the theory to collagen-mica system provides insight regarding the

intervening factors and rationalizes important system parameters such as force, energy, or

adhesion. A pure attraction at long distances is calculated in addition to a comparative

adhesion at contact with the literature. To further investigate the nanoscale organization of

collagen fibrils, a more detailed sequence based computational analysis will be performed

at short distances for the future goal of the project.

KEYWORDS - Collagen, DLVO Theory, Intermolecular interactions, self-assembly,

hydrophobic interactions, hydrophilic interactions



ELECTRONIC HEALTH REMOTE EHR CONTROL OF PATIENTS BY

MEDICAL STUDENTS WITH MOBILE AND SENSOR TECHNOLOGY

PINAR KIRCI1

1 Istanbul University, Turkey

ABSTRACT

Today, with internet and mobile devices like mobile phones and computers, the people

communicate easily with each other. But this technology is not efficiently utilized by

patients, doctors and medical students yet. There are not enough literature researches and

studies about the considered technology utilization and their effect on the education of

medical students. Purpose: The study proposed that with the use of mobile and sensor

technology for remote monitoring, following and controling of patients, such as; their

behaviors after operations, given medicine effects on them and daily routine (fever, heart

rates, weight.etc) data can be collected by medical students, examined and reported

weekly to the professors. This will help medical students to improve their vision and skills

in patient surveillance.

KEYWORDS - Medical education of medical students, Medical teaching, Mobile

technology, Mobile health market.



AUTOMATIC DETECTION OF EPILEPSY USING EEG ENERGY AND

FREQUENCY BANDS

SAMEH BELLGDI1 , SAMER ARAFAT1

1 Kfupm, Saudi Arabia

ABSTRACT

This paper demonstrates the effectiveness of information fusion at the feature vectors level

for automatic detection of epilepsy. Experiments used different types of features ranging

from separate EEG filtered band waves to combinations of filtered band waves in addition

to two added feature vectors: signal power and signal energy. We used two classifiers with

the feature vectors: TreeBoost and support vector machines. We carried out experiments

using a real life EEG signals data set that is available from the University Hospital of

Bonn in Germany. The main contribution of our work is to show the effect of combining

together signal energy and power with different EEG band waves to classify epilepsy, and

that this combination has shown improvement (96.5% accuracy) over using feature vectors

with fewer band wave transformations (86-92.75% accuracy), using the TreeBoost

algorithm. This combination also computed 99% specificity and 94% sensitivity. The

results were evaluated using cross validation with 10 folds.

KEYWORDS - EEG, Machine Learning, Epilepsy, Signal Energy, Signal Power,

TreeBoost



SERVO MOTOR CONTROL USING PLC

SERHAT AKSUNGUR1, TARKAN KOCA1

1 Inonu University, Turkey

ABSTRACT

Today, with the rapid development of technology, importance of product quality and speed

of production are increasing. In parallel, the sensitivities of the actuators and the controller

used have gained importance. The servo motors and drives that are emerging in this

process are found more places for industrial applications. In this work, servo motor control

will be mentioned using PLC (Programmable Logic Controller). Different examples will

be emphasized and the benefits of the practice will be tried to be revealed.

KEYWORDS - Servo motors, servo drivers, PLC control

International Conference on AdvancedTechnology & Sciences

ICATScience@

ABSTRACTS - icatsconf.org · ABSTRACTS E-ISBN: 978-605-67535-0-3. International Conference on Advanced Technology & Sciences ... ELIF DOGU, MICHELE CEDOLIN, Y ESRA ALBAYRAK EVALUATION

Documents