Technology & Sciences ABSTRACT BOOKKemal Tutuncu, Selcuk Universitesi – Turkey . Alla Anohina Naumeca, Riga Technical University – Latvia ... International Conference on Advanced

September 12-15, 2017R�ga / LATVIA

ICAT 'Riga6 International Conference on Advanced

Technology & Sciences

th

Science@

LATVIA SECTION

Ed�torsOmer Faruk BAYIsma�l SARITAS

E-ISBN: 978-605-67535-2-7

ABSTRACT BOOK

ilker

Typewriter

ilker

Typewriter

International Conference on Advanced Technology & Sciences

6th International Conference, ICAT’Riga Riga, Latvia, September 12-15, 2017

Abstracts

Editors Ismail SARITAS

Omer Faruk BAY

International Conference on Advanced Technology & Sciences, ICAT’Riga 6th International Conference on Advanced Technology & Sciences

Riga, Latvia, September 12-15, 2017

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

International Conference on Advanced Technology & Sciences

6th International Conference, ICAT’Riga Riga, Latvia, September 12-15, 2017

Abstracts

Editors Ismail SARITAS

Omer Faruk BAY

E-ISBN: 978-605-67535-2-7

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

September – 2017


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

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 Selcuk University, Turkey Depertmant of Electric-Electronic Engineering, Faculty of Technology Alaeddin Keykubat Campus 42031 Konya, Turkey [email protected] Ilker Ali OZKAN Selcuk University, Turkey Depertmant of Computer Engineering, Faculty of Technology Alaeddin Keykubat Campus 42031 Konya, Turkey [email protected] Murat KOKLU Selcuk University, Turkey Depertmant of Computer Engineering, Faculty of Technology Alaeddin Keykubat Campus 42031 Konya, Turkey [email protected]


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


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

PREFACE

International Conference on Advanced Technology & Sciences (ICAT'17) has been organized in Riga, Latvia on 12-15 September, 2017.

The main objective of ICAT'Riga 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'Riga.

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 146 paper submissions from 21 different countries and 87 universities. Each paper proposal was evaluated by two reviewers. And finally, 83 papers will be presented at our conference.

In particular we would like to thank Prof. Dr. Mustafa SAHIN, Rector of Selcuk University; Prof.Dr. Leonids Ribickis, Rector of Riga Technical 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


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

PROGRAMME COMMITTEES

HONORARY CHAIR : Mustafa Sahin, Rector of Selcuk University, Turkey Leonids Ribickis, Rector of Riga Technical University – Latvia

GENERAL CHAIR :

Omer Faruk Bay, Gazi University, Turkey Ismail Saritas, Selcuk University, Turkey

CO-CHAIR :

Janis Grundspenkis, Riga Technical University – Latvia Ilker Ali Ozkan, Selcuk University – Turkey Kemal Tutuncu, Selcuk Universitesi – Turkey Alla Anohina Naumeca, Riga Technical University – Latvia Murat Koklu, Selcuk University – Turkey Jurgis Porins, Riga Technical University – Latvia

PUBLICATION CHAIR :

Mehmet Akif Sahman, Selcuk University, Turkey Ali Yasar, Selcuk University, Turkey Mustafa Buber, Selcuk University, Turkey

INTERNATIONAL ADVISORY BOARD

Agris Nikitenko, Latvia Aleksandrs Sisojevs Latvia Alexander Sudnitson, Estonia Alla Anohina Naumeca, Latvia Amar Ramdane Cherif, France Anca Loana Andreescu, Bulgaria Andis Supe, Latvia Anne Villems, Estonia Ansis Kluga, Latvia Antonella Reitano, Italy Artan Luma, Macedonia Biagio Lenzitti, Italy Binod Kumar, India Boris Akanaev, Kazakhstan Dimitris Dranidis, Greece Dmitrijs Pikulins, Latvia

Dmitrijs Bliznuks, Latvia Domenico Tegolo, Italy Egons Lavendelis, Latvia Eisha Akanksha, India Elinda Kajo Mece, Romania Erika Nazaruka, Latvia Essa Essa, Iraq Gabriel Luna Sandoval, Mexico Gints Jekabsons, Latvia Guntars Balodis, Latvia Halife Kodaz, Turkey Harun Uguz, Turkey Heinz Dietrich Wuttke, Germany Howard Duncan, Ireland Huse Fatkic, Bosnia and Herzegovina Ilker Ali Ozkan, Turkey


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

Ilmars Slaidins, Latvia Ingars Erins, Latvia Ismail Saritas, Turkey Ivan Jelinek, Czech Republic Jan Vom Brocke, Liechtenstein Katrina Bolochko, Latvia Kemal Tutuncu, Turkey Larisa Zaitseva, Latvia Mahdi Shahbakhti, United States Majida Ali Abed Meshari, Iraq Marco Porta, Italy

Marina Uhanova, Latvia Mehmet Cunkas, Turkey Mehmet Akif Sahman, Turkey Natasa Hoic Bozic, Croatia Oksana Nikiforova, Latvia Pavels Rusakovs, Latvia Saadetdin Herdem, Turkey Sergei Parshutin, Latvia Valerijs Zagurskis, Latvia

ORGANIZING COMMITTEE

Omer Faruk Bay, Turkey Alla Anohina Naumeca, Latvia Nafiz Arica, Turkey Saadetdin Herdem, Turkey Jurgis Porins, Latvia Ismail Saritas, Turkey Kemal Tutuncu, Turkey

Silyan Sibinov Arsov, Bulgaria Dimitris Dranidis, Greece Angel Smrikarov, Bulgaria Sakir Tasdemir, Turkey Mehmet Cunkas, Turkey Ilker Ali Ozkan, Turkey Murat Koklu, Turkey Marco Porta, Italy

TECHNICAL COMMITTEE

Mehmet Akif Sahman, Turkey Sintija Petrovica, Latvia Ali Yasar, Turkey Mustafa Buber, Turkey Sandis Spolitis, Latvia

Esra Kaya, Turkey Mara Pudane, Latvia Arturs Aboltins, Latvia Vjaceslavs Bobrows, Latvia Fatih Saday, Turkey Abdulkadir Saday, Turkey


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

COMPARISON OF HARMONICS AT PHASE SHIFT CONTROLLERS OF ISOLATED BIDIRECTIONAL DC DC CONVERTER FOR DC GRID SYSTEMS 1

RAHMI ILKER KAYAALP, TUGCE DEMIRDELEN, MEHMET UGRAS CUMA, KAMIL CAGATAY BAYINDIR, MEHMET TUMAY

HEART SOUND RECORDING DEVICE DESIGN AND AUTOMATIC S1 AND S2 WAVES DETECTION 2

MEHMET FEYZI AKSAHIN, BURCU OLTU, BUSRA KUBRA KARACA

IMPLEMENTATION OF CHEST DISEASE DIAGNOSIS OF ARTIFICIAL NEURAL NETWORKS 3

MEHMET BALCI, ERKAN ULKER, DILEK ERTAS, FIKRET KANAT

AFM CHALLENGES FOR CHARACTERIZATION OF SOFT BIOMOLECULAR NANOMATERIALS AND PRACTICAL INSIGHTS TO BE OVERCOME 4

ESMA ERYILMAZ

SUPERVISED SEGMENTATION OF RETINAL VESSEL STRUCTURES USING ANN 5

ESRA KAYA, ISMAIL SARITAS, ILKER ALI OZKAN

A NOVEL DATA PRE PROCESSING METHOD IN THE CLASSIFICATION OF DIABETIC RETINOPATHY DIFFERENCE BASED SUPERVISED WEIGHTING BY DBSCAN

6

KEMAL POLAT, KADIR SABANCI

INTERNET OF THINGS IN PRECISION AGRICULTURE FOR GREENHOUSE PARAMETER CONTROL 7

HIDAYET OGUZ, ABDULLAH ERDAL TUMER

WIND ENERGY SOURCE BASED NEW GENERATION ELECTRIC GENERATOR 8

FIRAT EKINCI, TUGCE DEMIRDELEN, ERHAN FIRAT, MEHMET BILGILI

COMPARISON OF SRF PLL AND E PLL TECHNIQUES FOR GRID CONNECTED INVERTER 9

FEHMI SEVILMIS, HULUSI KARACA

PERFORMANCE ANALYSIS OF DUAL SECOND ORDER GENERALIZED INTEGRATOR PHASE LOCKED LOOP FOR GRID INTERACTIVE INVERTER 10

FEHMI SEVILMIS, HULUSI KARACA


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

THE EFFETCS OF DIFFERENT PWM METHODS ON THE PERFORMANCE OF INDUCTION MOTOR DRIVE 11

M MUSTAFA ERTAY

BLACKOUT DETECTION ALGORITHM IN SMART GRID 12

OKAN OZGONENEL, UNAL KURT

IOT AND ARM BASED DISABLED PERSON OR WHEELCHAIR FALL DETECTION 13

ABDULKADIR SADAY, FATIH SADAY

BARRIER LYAPUNOV FUNCTION BASED CONTROL OF AN EXPERIMENTAL BALL AND BEAM SYSTEM 14

FARUK DEMIRKIRAN, ESREF BOGAR, SELAMI BEYHAN

STRIP MAPS PRODUCTION WITH UNMANNED AERIAL VEHICLE 15

OMER MUTLUOGLU, ALPEREN ERDOGAN

CUTTING FORCES AND TOOL GEOMETRY BASED TOOL CONDITION MONITORING USING ARTIFICIAL NEURAL NETWORK IN TURNING 16

HACI SAGLAM, SULEYMAN NESELI

REVIEW OF INDUSTRY 4 0 17

ABDURRAHMAN KARABULUT, HAKAN SAHMAN, MEHMET AKIF SAHMAN, ABDULLAH OKTAY DUNDAR

A COMPARISON STUDY ON INDUCTION MOTOR PARAMETER IDENTIFICATION 18

MEHMET CUNKAS

GIS BASED PROCESSING OF IOT DATA KONYA EXAMPLE 19

ABDULLAH OKTAY DUNDAR, MEHMET AKIF SAHMAN

PROGRAMMABLE AND MODULAR ELECTRONIC SYSTEM FOR FLUID NETWORKS 20

STEFANO NOLI, CARLA VACCHI

DESIGN OF A PROTOTYPE FOR MONITORING OF THE FAULTY EQUIPMENT VIA GPRS 21

OMER FARUK BAY, OSMAN OZKARACA


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

SHORT TERM WIND POWER FORECASTING USING ANFIS AND REGRESSION METHODS 22

HASAN HUSEYIN CEVIK, MEHMET CUNKAS

ANALYSIS THE SINGLE PHASE DISTRIBUTION TRANSFORMER USING NUMERICAL METHOD 23

ATABAK NAJAFI, ABDOLREZA GHAFFARLOU, OKAN OZGONENEL

JET PULSE FILTER AUTOMATION 24

MUCIZ OZCAN, NIHAT CANKAYA

THE USE OF ELECTRONIC CONTROL SYSTEMS IN ROAD TRAFFIC 25

AYHAN GOKTEPE, FATIH YIGIT

EXAMINATION AND COMPARISON OF THE COMMUNICATION PROTOCOLS ON THE APPLICATION LAYER IN IOT 26

CEM GULTUNCA

SLIMWEAR S SMART PATCH INTERFACE DESIGN FOR LIGHTWEIGHT MIDDLEWARE IN WEARABLE SYSTEM 27

TAE GYU LEE, YONG K KIM

EFFECTIVE HYPERLINK SETUP ALGORITHM IN WIRELESS SENSOR NETWORK 28

BYUN GON KIM, YONG K KIM

INTERNET CONTROLLED ELECTRONIC SCOREBOARD DESING 29

FUNDA DEMIR, BATIKAN ERDEM DEMIR

A REALISTIC APPROACH TO SOLVE QUAY CRANE SCHEDULING PROBLEM 30

YUSUF YILMAZ, MUSTAFA EGEMEN TANER

TECHNICAL ENVIRONMENTAL AND FINANCIAL FEASIBILITY OF WIND TURBINE FOR SHOEBURYNESS 31

RIZA BUYUKZEREN, ALI KAHRAMAN

A MODIFIED CARNEGIE MELLON ESTIMATION METHOD FOR WHITE GOODS E WASTE QUANTITY OF TURKEY 32

ZEYNEP OZSUT BOGAR, ASKINER GUNGOR NUMERICAL SOLUTION OF MAGNETOCONVECTION IN A FERROFLUID FILLED CAVITY 33

BENGISEN PEKMEN GERIDONMEZ


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

A COMPARATIVE STUDY BETWEEN COMPUTER AIDED STATISTICAL APPROACH AND ARTIFICIAL NEURAL NETWORK 34

MUSTAFA ALTIN, SAKIR TASDEMIR, SULEYMAN NESELI

ARTIFICIAL NEURAL NETWORK FOR DETERMINATION OF PRESSURE RESISTANCE OF CONSTRUCTION ELEMENTS 35

SAKIR TASDEMIR, MUSTAFA ALTIN, BURAK TEZCAN

NUMERICAL SIMULATION OF MIXED CONVECTION FLOW IN A CAVITY FILLED WITH FE3O4 WATER 36

BENGISEN PEKMEN GERIDONMEZ, SEVDI MERVE EROGLU

A FOURTH ORDER SWITCHED CAPACITOR LOW PASS FILTER WITH DYNAMIC SWITCHING BIAS OP AMPLIFIERS 37

HIROO WAKAUMI

DETERMINING THE OPTIMAL PORTFOLIO BY USING DATA ANALYSIS AT ISTANBUL STOCK EXCHANGE MARKET 38

HASAN AKYER, OZCAN MUTLU

ESTIMATING HOURLY SOLAR RADIATION IN TURKEY USING AIR TEMPERATURES 39

OZGUR SOLMAZ

SEGMENTATION AND APPLICATIONS OF VIDEOCAPILLAROSCOPY 40

KEMAL TUTUNCU, MUSTAFA BUBER

PREDICTING SEMINAL QUALITY USING ARTIFICIAL NEURAL NETWORK 41

I UNAL SERT, ILKER ALI OZKAN, MURAT KOKLU

A COMPARATIVE STUDY OF MODELING OF CHRONIC KIDNEY DISEASE USING DATA MINING 42

ARIF AYDIN, MURAT KOKLU, ILKER ALI OZKAN

THE DETECTION OF A TYPICAL RECTAL CANCER WITH SEMI AUTOMATIC IMAGE PROCESSING TECHNIQUES 43

ALI YASAR, ISMAIL SARITAS, HUSEYIN KORKMAZ INVESTIGATION OF OPTIUM DESIGN OF HETEROJUNCTION N ZNO P SI SOLAR CELL WITH PC1D 44

AYSEGUL TOPRAK, HAMDI SUKUR KILIC


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

DERMATOLOGICAL IMAGE LIBRARY AND DERMATOLOGICAL WOUND TRACING APPLICATION FOR DERMATOLOGISTS 45

SUMEYYA ILKIN, OKTAY DUMAN, SUHAP SAHIN

A COMPARATIVE PERFORMANCE EVALUATION OF CLASSIFICATION TECHNIQUES ON CAR EVALUATION DATASET 46

MURAT CINIVIZ, MURAT KOKLU

INTRODUCING LABORATORY EQUIPMENTS USING AUGMENTED REALITY 47

NIGMET KOKLU

MODELING OF PHOTOVOLTAIC MODULE AND PARAMETERS ANALYSIS OF A 335W PV MODULE 48

MELIH COBAN, MURAT FIDAN, OKTAY AYTAR

OUTRUNNER ROTOR SRM DESIGN FOR ELECTRIC MOTORCYCLE 49

CIHAN SAHIN, MEVLUT KARACOR

EXAMINATION OF THE ENERGY SAVING POTENTIAL OF A HYBRID CITY BUS WITH REGENERATIVE BRAKING 50

SEREF SOYLU, ZAFER DEMIR

MORPHOLOGICAL AND OPTICAL PROPERTIES OF TANTALUM TA TITANIUM TI AND MOLYBDENUM MO THIN FILMS PRODUCED BY PULSED LASER DEPOSITION

51

SERAP YIGIT GEZGIN, AHMET TOPRAK, ULFET ATAV, HAMDI SUKUR KILIC

DRIVE SYSTEM DESIGN FOR MUTUALLY SWITCHED RELUCTANCE MOTOR 52

MEVLUT KARACOR, CIHAN SAHIN, FERIHA ERFAN KUYUMCU

DESIGN ANALYSIS AND MANUFACTURING OF MULTIPURPOSE CYLINDRICAL SHAPE ROBOT 53

HILMI SAYGIN SUCUOGLU, ISMAIL BOGREKCI, PINAR DEMIRCIOGLU, OGULCAN TURHANLAR

AN INVESTIGATION ON A TYPE OF FUZZY MULTIFUNCTION 54

AHU ACIKGOZ, FERHAT ESENBEL A RESEARCH ON A SORT OF FUZZY MULTIFUNCTIONS 55

AHU ACIKGOZ


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

DETERMINATION OF TOOL GEOMETRY EFFECT ON TOOL VIBRATION USING ARTIFICIAL NEURAL NETWORK IN TURNING OPERATION 56

SULEYMAN NESELI, HACI SAGLAM, SULEYMAN YALDIZ, MUSTAFA ALTIN

THE THREE DIMENSIONAL ELECTROMAGNETICANALYSIS OF POWER TRANSFORMERS BASED ON ANSYS 57

TUGCE DEMIRDELEN, SELVA BAL, MEHMET TUMAY, MEHMET UGRAS CUMA, ABDULLAH CICIBAS, KEREM KOSEOGLU, MAHMUT AKSOY

DESIGN OF MASTER AND SLAVE MODULES ON BATTERY MANAGEMENT SYSTEM FOR ELECTRIC VEHICLES 58

ARDA KILIC, SELIM KOROGLU, AKIF DEMIRCALI, SELAMI KESLER, YUSUF ONER, ERKAN KARAKAS, PETER SERGEANT

RF TELEMETRY DESIGN FOR ELECTRIC VEHICLE 59

HANDE NEVIN KILICKIR, AKIF DEMIRCALI, SELIM KOROGLU, SELAMI KESLER, MUSTAFA TUMBEK, ARDA KILIC

AN INTERACTIVE TURKISH SIGN LANGUAGE LEARNING GAME USING LEAP MOTION CONTROLLER 60

FIDAN KAYA GULAGIZ, HIKMETCAN OZCAN, SUHAP SAHIN

STATE SPACE MODELLING AND SIMULATION FOR PEDELEC 61

OKAN UYAR, MEHMET CUNKAS

PERFORMANCE ANALYSIS OF SINGLE PHASE INDUCTION MOTOR FOR DIFFERENT ROTOR SLOT TYPES 62

AHMET FENERCIOGLU, MERVE SEN KURT, MEHMET AKAR, UNAL KURT

LQR REFERENCE TRACKING CONTROL OF BOEING 747 100 LONGITUDINAL DYNAMICS WITH CG SHIFTS 63

EZZEDDIN ELARBI, DINA LAILA, ABDULHAMID GHMMAM, NADJIM HORRI

SEAMLESS INTEGRATION OF CPS BASED FACTORIES ACROSS VARIOUS ENGINEERING AND SIMULATION DISCIPLINES 64

STEPHAN WEYER, CARSTEN HARMS, MARTIN RUSKOWSKI AN APPLICATION TO IMPROVE CIRCUIT PERFORMANCE OF A MULTIFUNCTION ANALOG FILTER USING INVERTING COMPOSITE CURRENT CONVEYOR

65

HASAN CICEKLI, AHMET GOKCEN


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

HIGH EFFICIENCY GASEOUS ACTIVE WATER COOLING SYSTEM 66

SELMAN TASAR, KEMAL TUTUNCU

ESTIMATION OF TURKEY S ELECTRICITY CONSUMPTION FOR 2023 BY USING MULTILAYER PERCEPTRON S MLPS 67

SEMIHA DURGUN, ALI OSMAN OZKAN

MINIMUM COST CONTROL OF PUMPING SYSTEMS WITH SCHEDULED ELECTRIC TARIFFS 68

ROBERTO SANCHIS LLOPIS, IGNACIO PENARROCHA ALOS

A MULTI CRITERIA DECISION MAKING APPROACH FOR GREEN SUPPLIER SELECTION PROBLEM 69

ZEHRA DURAK, ELIF OZGORMUS

IMPROVEMENT OF PROJECTION ERROR EVALUATION FOR MASSIVE DATA SETS 70

KOTRYNA PAULAUKSIENĖ, OLGA KURASOVA

DESIGN AND IMPLEMENTATION OF E LEARNING WEB PAGE FOR CHILDREN WITH INTELLECTUAL DISABILITIES 71

HANKYU LIM, YERIN JEONG, JAEJUN JANG, HYEMI DO, SYAOCHING JOU, EUNJU PARK

A MULTI OBJECTIVE MODEL FOR GREEN ORDER ALLOCATION PROBLEM IN MULTI ITEM MULTI SUPPLIER ENVIRONMENT 72

AHMET ALP SENOCAK, HACER GUNER GOREN

PDM IC MPPT CONTROLLED SERIES RESONANT CONVERTER FOR PV GENERATION SYSTEMS 73

AKIF KARAFIL, HARUN OZBAY, SELIM ONCU

SLIDING MODE CONTROL FOR A MIMO NONLINEAR COUPLED TANK SYSTEM 74

INAYET OZGE AKSU, RAMAZAN COBAN DESIGN AND FEM ANALYSIS OF LOW VOLTAGE OUTER ROTOR LINE START PERMANENT MAGNET SYNCHRONOUS MOTORS WITH DIFFERENT MAGNET ALIGNMENTS

75

MUSTAFA TUMBEK, SELAMI KESLER, YUSUF ONER


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

IOT (INTERNET OF THINGS) BASED HOME SECURITY AND AUTOMATION MODEL WITH A WEB AND MOBILE APP 76

FATIH SADAY, ABDULKADIR SADAY

THE USE OF ARTIFICIAL NEURAL NETWORKS FOR MODELING THE REMOVAL OF CR VI WITH POLYMERIC CALIX 6 ARENE 77

ABDULLAH ERDAL TUMER

SILICON RECOVERY IN WASTE CRYSTALLINE SILICON SOLAR CELLS BY A SINGLE CHEMICAL ETCHING 78

TERUAKI MATSUBARA, KEI TAKAMI, MD AZHAR UDDIN, YOSHIEI KATO, TAKANORI KAWANISHI, YOSHIAKI HAYASHI

A NOVEL FORMULATION USED FOR INDOOR AREA LOCATION ESTIMATION BASED ON WIFI RSSI LEVELS 79

MUHAMMED FAHRI UNLERSEN, KADIR SABANCI, MUCIZ OZCAN

AN APPLICATION FRAMEWORK FOR SHOP SCHEDULING OPTIMIZATION PROBLEMS 80

OGUZHAN KOCATEPE

A 3 WAY PLANAR MICROWAVE POWER DIVIDER COMBINER FOR HIGH POWER APPLICATIONS 81

MARCIN GORALCZYK


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


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

COMPARISON OF HARMONICS AT PHASE SHIFT CONTROLLERS OF ISOLATED BIDIRECTIONAL DC DC CONVERTER FOR DC GRID SYSTEMS

RAHMI ILKER KAYAALP1 , TUGCE DEMIRDELEN2 , MEHMET UGRAS CUMA3 , KAMIL CAGATAY BAYINDIR4 , MEHMET TUMAY5

1 Cukurova University, Turkey ; 2 Adana Science And Technology University, Turkey ; 3 Cukurova University, Turkey ; 4 Ankara Yildirim Beyazit University, Turkey ; 5Adana

Science And Technology University, Turkey

ABSTRACT

In this paper harmonic effects of three phase-shift controllers on full-bridge isolated bidirectional dc-dc converter are compared. Isolated Bidirectional DC-DC Converter (IBDC) is one of the power transferring systems in power quality issues and design of IBDC is expressed. Single Phase-Shift (SPS), Extended Phase-Shift (EPS) and Dual Phase-Shift (DPS) controllers are explained and discussed for harmonic elimination. Furthermore, a novel IBDC with CLLC topology is introduced and it gives the best harmonic elimination way with DPS control method. The topology is usable for medium power applications such as dc power transferring between renewable energy sources. The cases of study clearly show the capabilities of phase-shift control methods and IBDC with CLLC topology from the viewpoint of harmonic elimination in order to obtain clear dc-link voltage.

KEYWORDS - Isolated bidirectional dc-dc converter, H-bridge, phase-shift controllers, harmonics, comparison.


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

HEART SOUND RECORDING DEVICE DESIGN AND AUTOMATIC S1 AND S2 WAVES DETECTION

MEHMET FEYZI AKSAHIN1 , BURCU OLTU2 , BUSRA KUBRA KARACA3

1 Baskent University, Turkey ; 2 Baskent University, Turkey ; 3 Baskent University, Turkey

ABSTRACT

Cardiovascular diseases are the major cause of death in the world. Early diagnose of heart diseases provide an easy and economic treatment. Heart diseases can be diagnosed easily using data obtained from heart sounds. Today, heart sounds are listened by heart auscultation method and heart diseases are diagnosed according to this process. However since auscultation method depends on the experience and hearing ability of physicians, it is not exactly reliable. Therefore a device prototype that collects the heart sound from human body and also amplifies, filters, displays and records collected data on digital environment was designed. Subsequently recordings from 5 different patients group were collected within clinical application. By the use of these records, S1 and S2 waves were detected applying discrete wavelet transform and teager energy operator. As a result S1 and S2 waves were determined with high sensitivity, specificity and accuracy rates.

KEYWORDS - S1 and S2 detection ,discrete wavelet transform ,teager energy operator,heart sound recording device



IMPLEMENTATION OF CHEST DISEASE DIAGNOSIS OF ARTIFICIAL NEURAL NETWORKS

MEHMET BALCI1 , ERKAN ULKER2 , DILEK ERTAS3 , FIKRET KANAT4

1 Selcuk University Higher School Of Vocational And Technical Sciences, Turkey ; 2 Selcuk University Engineering Faculty, Turkey ; 3 Selcuk University Medicine Faculty,

Turkey ; 4 Selcuk University Medicine Faculty, Turkey

ABSTRACT

With the widespread use of computer systems all over the world, work in many disciplines has become easier. One of these disciplines is medicine. Nowadays, the use of electronic and computer systems have become almost indispensable elements of medicine. This situation also contributes significantly to the development of the studies carried out in the field of computer engineering. In particular, studies in advanced engineering fields such as artificial intelligence and machine learning are becoming widespread. Classification studies are at the forefront of the widely used techniques of machine learning techniques. Software that produces highly successful results in classification processes is being developed using artificial neural networks. Classification success is influenced by the neuron structure of the network and the activation functions used. Asthma and Chronic Obstructive Pulmonary Disease (COPD) are the most common pulmonary diseases worldwide. Both of these diseases are those diseases that narrow the airways in the lungs and occur with similar symptoms in patients. In this study, we have developed a system that allows pre-diagnosis of Asthma or Chronic Obstructive Pulmonary Disease (COPD) using data used in diagnosis of chest diseases. In the developed system, classification work was done using artificial neural networks, which are machine learning techniques. The designed artificial neural network was applied on the data set formed from the patients who applied to the chest diseases service and highly successful classification were obtained.

KEYWORDS - Artificial Neural Networks, Asthma, Chronic Obstructive Pulmonary Disease (COPD),Diagnosis



AFM CHALLENGES FOR CHARACTERIZATION OF SOFT BIOMOLECULAR NANOMATERIALS AND PRACTICAL INSIGHTS TO BE OVERCOME

ESMA ERYILMAZ1

1 Selcuk University Faculty Of Technology Department Of Biomedical Engineering, Turkey

ABSTRACT

Research in the fields of bioengineering and bionanotechnology require tools which are capable in reaching nanoscale level features of related biomaterials. Atomic force microscopy (AFM), a scanning probe microscope (SPM), is a powerful technique operating at multiple media and in different modes providing molecular and atomic level features of materials without destroying the sample surface. It is extensively used for not only surface characterization but also surface patterning with biomolecules for generating bio-functionalized nanomaterials. However, AFM procedure to investigate the nanoscale properties of biomaterials, unlike micron or macro scale material analysis, requires intensive effort and quite long time for achievement of consistent results. In this paper, we discussed possible problems which can arise during the characterization and the production of bionanomaterials and provided some practical insights to those problems. For that, a representative biomolecular system of a structural protein, collagen was used. Collagen is an extracellular matrix protein forming about 30\% of total protein mass in the body. We, in this study, investigated the technological and the experimental challenges of AFM technique to be overcome for achievement of consistent and significant results. For that matter, in addition to the AFM related artifacts, substrate and biomolecular system related issues were discussed with the representative AFM topography images of collagen fibrillar structures.

KEYWORDS - AFM artifact, Collagen, double tip, bionanomaterial, noise effect, mica, D-period, biomaterial characterization,biomaterial imaging,biomedical imaging



SUPERVISED SEGMENTATION OF RETINAL VESSEL STRUCTURES USING ANN

ESRA KAYA1 , ISMAIL SARITAS2 , ILKER ALI OZKAN3

1 Selcuk University, Turkey ; 2 Selcuk University, Turkey ; 3 Selcuk University, Turkey

ABSTRACT

In this study, a supervised retina blood vessel segmentation process was performed on the green channel of the RGB image using artificial neural network (ANN). The green channel is preferred because the retinal vessel structures can be distinguished most clearly from the green channel of the RGB image. The study was performed using 20 images in the DRIVE data set which is one of the most common retina data sets known. The images went through some preprocessing stages like contrast-limited adaptive histogram equalization (CLAHE), color intensity adjustment, morphological operations and median and Gaussian filtering to obtain a good segmentation. Retinal vessel structures were highlighted with top-hat and bot-hat morphological operations and converted to binary image by using global thresholding. Then, the network was trained by the binary version of the images specified as training images in the dataset and the targets are the images segmented manually by a specialist. The average segmentation accuracy for 20 images was found as 0.9492.

KEYWORDS - ANN, Blood Vessel, Retina, Segmentation, Supervised Learning



A NOVEL DATA PRE PROCESSING METHOD IN THE CLASSIFICATION OF DIABETIC RETINOPATHY DIFFERENCE BASED SUPERVISED WEIGHTING

BY DBSCAN

KEMAL POLAT1 , KADIR SABANCI2

1 Abant Izzet Baysal University, Turkey ; 2 Karamanoglu Mehmetbey University, Turkey

ABSTRACT

Diabetic retinopathy is a disease observed among diabetic people. Diabetic retinopathy dataset contains features extracted from the Messidor image in order to estimate whether an image is a DR sign. In this paper, a novel feature weighting scheme based on the differences between data points and cluster centers found by using density-based spatial clustering (DBSCAN), has been proposed. First of all, the cluster centers belonging to all features in the dataset have been found by DBSCAN. In the second phase, the differences between data points and cluster centers for each class in the dataset have been calculated. Finally, to find the weighting coefficients, the average values of all the differences for each feature have been computed and then these obtained average difference values have been assigned to each feature for each class separately. This weighting method has been proposed prior to classification or clustering algorithms. The promising results have been obtained in the classification of DR disease.

KEYWORDS - Diabetic Retinopathy, Supervised feature weighting method, classification, pattern recognition



INTERNET OF THINGS IN PRECISION AGRICULTURE FOR GREENHOUSE PARAMETER CONTROL

HIDAYET OGUZ1 , ABDULLAH ERDAL TUMER2

1 Necmettin Erbakan University, Turkey ; 2 Necmettin Erbakan University, Turkey

ABSTRACT

Today, human-human communication through internet is widely common. The developed technology made things to communicate each other possible. This technology, which is called the Internet of thins, is a technology that is expected to be used in many areas of our lives. In this technology, the sensors detect/sense events (heat, light, motion, humidity, any gas, etc.) where they are located and can send events to mobile phones, social media accounts or public web pages via data transmitters. Or they can determine the behavior of other things. Economical solutions can be produced in health, environment, education, agriculture, animal husbandry and industry. For instance, it is crucial to obtain optimum efficiency from greenhouse products. It is seen that farmers have huge financial losses due to incorrect weather forecasts and improper irrigation methods for crops. In this study, the monitoring and control of the greenhouse parameters were examined by using the internet of things technology. Precision farming with IOT technology is used to control greenhouse parameters very economically. In the proposed greenhouse design, farmers can easily provide the environmental conditions they desire for different products in the same greenhouse. To do this, the temperature sensor, ambient humidity sensor and soil moisture sensor are used in operation. The sensors are connected to a Wi-Fi module named ESP8266 with a microcontroller inside. Optimal parameters for each product are input into the microcontroller. For instance, ambient humidity is input as 65% and soil nematode is input as 16% for tomato. When the soil nematode in the tomato section falls below 16%, the ESP8266 module will run the irrigation system for 1 minute. If the new soil nematode values are still low after 1 min, the irrigation system will be switched on again; if it is normal or above, the irrigation will be terminated. The same system applies within the spinach section, but with a difference. The optimum soil nematode values for spinach are different. Data received from the sensors are transmitted to the farmer's social media account and to the mobile phone via ESP8266 before and after irrigation with certain periods (e.g. 3 hours). With the developed IOT system, both will be achieved; providing optimum irrigation and informing the farmer about greenhouse parameters. In this study, the results were examined by carrying out simulations.

KEYWORDS - Precision farming, IOT, sensor, microcontroller



WIND ENERGY SOURCE BASED NEW GENERATION ELECTRIC GENERATOR

FIRAT EKINCI1 , TUGCE DEMIRDELEN2 , ERHAN FIRAT3 , MEHMET BILGILI4

1 Adana Science And Technology University, Turkey ; 2 Adana Science And Technology University, Turkey ; 3 Munzur University, Turkey ; 4 Cukurova University, Turkey

ABSTRACT

This study is aimed at providing an innovative point for the wind tribunes’ technology to the researchers and the application engineers dealing with design issue. By this aim, New Generation Generators (NGG) will be investigated academically as well as implemented in our country. In this work, the main goal is increasing the utilization rate of wind energy which is low for our country and spreading the usage of this system. A new generation electric generator (NGG) based on vortex-induced vibration will be design and analyze in small-scale, compact, light (not heavy), portable and highly efficient for the usage in rural areas. This new research and development subject is also aimed to contribute to the improvement of technological developments and scientific knowledge in new generations of electricity generators based on vortex-induced vibrations from wind energy technologies.

KEYWORDS - Bladeless Tribune,Green Energy,Renewable Energy,Wind energy,Wind Tribune



COMPARISON OF SRF PLL AND E PLL TECHNIQUES FOR GRID CONNECTED INVERTER

FEHMI SEVILMIS1, HULUSI KARACA2

1 Selcuk University, Turkey; 2 Selcuk University, Turkey

ABSTRACT

In order to synchronize power electronics converters with the grid, the knowledge of the phase angle plays a very important role. In grid connected applications, the phase angle of grid voltage must be determined accurately and quickly. Therefore, Synchronous Reference Frame-Phase Locked Loop (SRF-PLL) technique is extensively used due to the advantages of its simple structure, easy to implement. Under ideal grid conditions, the SRF-PLL provides quick and accurate phase detection. However, under non-ideal grid conditions such as unbalanced and distorted grid, the dynamic response of the PLL is failed. In this study, the SRF-PLL and E-PLL are simulated and compared under different grid conditions such as unbalanced, distorted and variable frequency. The simulation results verify that E-PLL is a very good synchronization technique under non-ideal grid conditions for grid connected inverter.

KEYWORDS - E-PLL, grid synchronization, phase locked loop, PLL



PERFORMANCE ANALYSIS OF DUAL SECOND ORDER GENERALIZED INTEGRATOR PHASE LOCKED LOOP FOR GRID INTERACTIVE INVERTER

FEHMI SEVILMIS1, HULUSI KARACA2


ABSTRACT

Estimating the phase angle of grid plays a crucial role in grid interactive inverter in order to be synchronized the inverter and the grid. Phase locked loop (PLL) method is usually used in applications of grid synchronization. The angle information obtained by the PLL ensures the amplitude, phase and frequency of the inverter the same as grid. In case of unbalanced and distorted grid voltages, conventional PLL method which is called Synchronous Reference Frame (SRF)-PLL cannot accurately estimate the phase angle of grid voltages. So, errors occur in the grid synchronization. To overcome this, Dual Second Order Generalized Integrator (DSOGI)-PLL is a suitable synchronization method which determines the phase angle of grid voltages quickly and precisely. In this paper, the SRF-PLL and DSOGI-PLL are analysed and simulated under different adverse grid conditions such as unbalanced, harmonics and variable frequency in MATLAB/Simulink. Simulation results show that DSOGI-PLL has provided a very good solution for grid synchronization of three phase grid interactive inverter.

KEYWORDS - DSOGI-PLL, grid synchronization, phase locked loop, PLL, SRF-PLL



THE EFFETCS OF DIFFERENT PWM METHODS ON THE PERFORMANCE OF INDUCTION MOTOR DRIVE

M MUSTAFA ERTAY1

1 Duzce University, Turkey

ABSTRACT

PWM methods allow efficient use of power in UPS inverters, switched mode dc-dc power conversion, inverter based compensation devices and industrial induction motor drives. For this reason, it has a wide range of usage in the industry. Many PWM methods have been used in induction motor drives. These methods are divided into continuous and discontinuous PWM methods. Among the most common continuous PWM methods are sinusoidal PWM (SPWM) and space vector PWM (SVPWM). Discontinuous PWM methods (DPWM) are known as energy efficient PWM methods. In this study, the effect of different PWM methods on the voltage source inverter (VSI) fed induction motor performance is investigated. Simulation studies were performed in MATLAB / Simulink environment. In the simulations, when the CPWM and DPWM methods are applied; the current change, torque change, speed change and harmonic distortion values of the induction motor are investigated.

KEYWORDS - Cpwm,Dpwm, power quality, vsi fed induction motor drive



BLACKOUT DETECTION ALGORITHM IN SMART GRID

OKAN OZGONENEL1 , UNAL KURT2

1 Ondokuz Mayis University, Turkey ; 2 Amasya University, Turkey

ABSTRACT

Today, renewable energy sources are in service at the level of blackout that makes service restoration complicated. Presently, this problem cannot be handled (or solved) employing a single solution technique. Thus, intelligent techniques for recovering the whole system back to the service are specifically needed in smartgrid. K-means Clustering technique is used to identify and detect the nature of blackouts (also known as line outages) in this paper. The method is regarded as a signal processing method and many algorithms have proposed to get a local optimum so far. Data generation and load flow studies is the first stage of the detection algorithm and common IEEE buses are used to achieve it. For load flow studies a common approach known as Newton Raphson (NR) is employed to calculate bus voltages and their associated angles. Together with bus voltages and angles (δ), active/reactive powers are the data that the K-means Clustering needs. The results show that the suggested technique is able to detect line outages prior to the beginning.

KEYWORDS - Blackout, K-Means Algorithm, Newton-Raphson Load Flow Algorithm, Smartgrid



IOT AND ARM BASED DISABLED PERSON OR WHEELCHAIR FALL DETECTION

ABDULKADIR SADAY1 , FATIH SADAY2

1 Selcuk University, Turkey ; 2 Karatay University, Turkey

ABSTRACT

It becomes necessary to monitor and follow our old ones for their health and safety when it comes to old age. Because of the weakness of body system and weak joints, they have a big risk of falling down. It is very important to know if an old person has fallen or not, so that he or she can be helped on time without a risk. Also disabled people whom using wheelchair need to be checked and monitored for fall detection. For this purpose we are developing a smart fall detection system based on IOT and ARM controller. This system using a gyro sensor and an accelerometer to detect person movements in time. This items can be mounted on wheelchair for falling detection. When a fall event occurs, sensor that connected to an ARM microcontroller constantly transmit the acceleration data to monitoring system. The system analysis the data and keeps monitoring for fall detection and abrupt movement changes over wheelchair. A sudden abrupt change means that there is a risk in the system and also big changes treated as a fall. In this case, if the person did not fall and fall detection alarm was false, the monitoring system allows to stop the alert when accelerometer in its normal position. If accelerometer or gyro sensor data stays constant for a while except its normal position, the system detects that disabled person has fallen and needs help. When this occurs, the system automatically triggers alert through internet connection or GSM band to alert the related ones of the person about the crash and situation instantly. We designed an cheap fall detection system based on IOT and ARM with this paper.

KEYWORDS - iot,fall dedection



BARRIER LYAPUNOV FUNCTION BASED CONTROL OF AN EXPERIMENTAL BALL AND BEAM SYSTEM

FARUK DEMIRKIRAN1 , ESREF BOGAR2 , SELAMI BEYHAN3

1 Ege University, Turkey ; 2 Pamukkale University, Turkey ; 3 Pamukkale University, Turkey

ABSTRACT

In this paper, a barrier Lyapunov function (BLF) based backstepping constrained controller is adapted to control of cascaded nonlinear systems. Using the designed controller, the system output is naturally constrained to predefined bounds based on the tracking error convergence rate. The control law is determined based on the Barrier Lyapunov function approach. In application part, using BLF based controller, an experimental unstable nonlinear ball and beam system is controlled in simulation and real-time, and compared to the quadratic Lyapunov function (QLF) based controller. The tracking results in transient and steady-state periods are consistent with the developed theory and the designed constrained controller is applicable for the control of unstable cascaded nonlinear systems.

KEYWORDS - Backstepping control, real-time constrained control, barrier Lyapunov function, stability, ball and beam system



STRIP MAPS PRODUCTION WITH UNMANNED AERIAL VEHICLE

OMER MUTLUOGLU1 , ALPEREN ERDOGAN2

1 Selcuk Universty, Turkey ; 2 Selcuk Universty, Turkey

ABSTRACT

In engineering projects, the concepts of time, cost and accuracy have significant value. For this reason, engineering projects need to be done in a short time, with low cost and high sensitivity. Determining the usability of photogrammetric techniques in engineering projects with Unmanned Aerial Vehicle will give a new breath to the production of these projects and engineering projects with shorter and more precise and lower costs will be produced with this technique. Developing and changing technology has a strong influence on map engineering, especially photogrammetry branch. In recent years, many different options have begun to be used in the acquisition of digital terrain data. One of these is the collection of numerical data from unmanned aerial vehicles through the installation of non-metric cameras. In this study, a survey was made for the construction of the strip map from the data obtained with the Unmanned Aerial Vehicles. As a study area, a route of approximately 2 km was selected on the road project of 3rd Regional Directorate of Highways. Unmanned Aerial Vehicle (UAV) on this route, images were taken from 70m height with 70% overlay. Based on the evaluation of these images, the obtained Digital Elevation Model (DEM) was used and an actual map was created. In this study, a common route has been determined through the existing map, which is produced as 1/1000 photogrammetric with the current map obtained with high accuracy. In the determined route, the cross sections were produced and the excavation fill volume and differences were investigated.

KEYWORDS - UAV, Photogrammetry, Digital Elevation Model, Strip Map



CUTTING FORCES AND TOOL GEOMETRY BASED TOOL CONDITION MONITORING USING ARTIFICIAL NEURAL NETWORK IN TURNING

HACI SAGLAM1 , SULEYMAN NESELI2

1 Selcuk University, Turkey ; 2 Selcuk University, Turkey

ABSTRACT

It is important monitoring of on-line tool condition to determine optimized machining parameters and obtain reduced costs in industrial applications. Cutting forces affects surface quality and accuracy of workpiece which stands ultimate aim of an production line. Artificial intelligence methods provide prediction of machining performance with reduced experiments and contribute to enhancing of productivity. In this work, a combined statistical analysis (ANOVA) and artificial neural network (ANN) approach is implemented to turning operation to obtain minimum cutting forces with minimum experiment. Experiments were performed in dry conditions and inclined angle, approaching angle and feed rate were chosen as input parameters. An ANN based model was developed and compare with the experimental results. To check the validation of the model, additional experiments were performed and the results showed that the developed model is reliable and successfully implemented to similar applications. Finally, the most effective machining parameters were determined with statistical analysis using data obtained from experiments and ANN model. The aim of the study is obtaining of high quality materials with high efficiency and low costs.

KEYWORDS - Cutting force,Rake angle,Entering angle,Feed rate,ANN



REVIEW OF INDUSTRY 4 0

ABDURRAHMAN KARABULUT1 , HAKAN SAHMAN2 , MEHMET AKIF SAHMAN3 , ABDULLAH OKTAY DUNDAR4

1 Technology Faculty, Turkey ; 2 Institute Of Science, Turkey ; 3 Technology Faculty, Turkey ; 4 Vocational School, Turkey

ABSTRACT

This review aims to provide a summary of previous studies on Industry 4.0. The fourth industrial revolution, Industry 4.0, is the integration of mechanical and digital products into the real world; it first appeared at the (Germany Messe Fair) in 2011. Industry 4.0 includes simulation, system integration, the Internet of things, cyber security, cloud technology, additive manufacturing, augmented reality, big data, and autonomous robots. Since 2011, many innovative, developed countries have initiated adaptation studies for the Industry 4.0 Revolution. However, there are still many large companies around the world that not yet upgraded their manufacturing systems to be Industry 4.0 ready. This is mainly due to possible risks associated with this new industrial revolution. Initially, firms have doubts about return on investment. Industry 4.0 modernization entails higher costs; therefore, feasibility studies should be conducted prior to any modernization. In addition, companies require the knowhow and experience of specialists to understand the timelines for return on investment, increase in productivity, and quality improvement. There are previous studies in the literature on the processes and means of transition to Industry 4.0forbusinesses. These studies provide companies with approaches to identifying ways they can adapt to Industry 4.0, and the sections of their businesses that should be updated. This review presents some examples from previous studies of methods and processes that firms can use to adapt to Industry 4.0.

KEYWORDS - Industry 4.0



A COMPARISON STUDY ON INDUCTION MOTOR PARAMETER IDENTIFICATION

MEHMET CUNKAS1

1 Selcuk Univeristy, Turkey

ABSTRACT

This paper presents a comparison study on the parameter identification of induction motor from its manufacturer data by using Constrained Nonlinear Optimization, Genetic Algorithms and Simulated Annealing Algorithm. The estimation problem for the induction motor’s equivalent circuit parameters is based on minimizing the sum of the square of differences between the actual values and the estimated values. In order to validity of the proposed method, a three-phase squirrel-cage induction machine are used. The estimation methods are compared with each other. The results show that hybrid simulated annealing is capable of obtaining more successful solution than constrained nonlinear optimization, genetic algorithms and simulated annealing algorithm. However hybrid simulated annealing, constrained nonlinear optimization and genetic algorithms are competitive each other because of the accuracy values.

KEYWORDS - parameter identification, induction motor, genetic algorithms, simulated annealing, nonlinear optimization



GIS BASED PROCESSING OF IOT DATA KONYA EXAMPLE

ABDULLAH OKTAY DUNDAR1 , MEHMET AKIF SAHMAN2


ABSTRACT

The aim of this study is to develop software that will enable the processing of device data from the Internet of Things (IOT) in the Geographic Information System (GIS) environment. Thanks to the IOT, which is a sub-area of Industry 4.0, devices have become controllable over the internet. Therefore, the data from these related devices can be accessed and processed very cheaply in a very short time. It is of great importance to companies that serve a large number of customers in different geographical locations to be able to quickly detect failures. Detection of faulty devices has become easier with the IOT. If there is a failure in an area, it may not be easy to understand from the numerical data in the stack. This information, which is related to the geographical location, is important in terms of location planning and future failure statistics. GIS allows users to more easily understand the data and provides greater convenience in processing it. In this study, software was developed to process the data from IOT devices in Konya province on a map. The data from the IOT devices was produced virtually (as a test). Failure data were processed on a map of Konya and thematic maps were formed from it. Thus, it becomes easy to see regional failure density distribution. The software was developed on the .net platform using MapXtreme to process the GIS data.

KEYWORDS - IOT, GIS



PROGRAMMABLE AND MODULAR ELECTRONIC SYSTEM FOR FLUID NETWORKS

STEFANO NOLI1 , CARLA VACCHI2

1 University Of Pavia, Italy ; 2 University Of Pavia, Italy

ABSTRACT

This abstract describes a low power programmable electronic control system able to communicate with different sensors, actuators and wireless communication modules. Main field of application where this hardware platform has been specialized and tested is on a pipeline networks in which a fluid flows. The device analyzes the data coming from a pressure sensor, records on a flash memory only the significant events related to the pressure value and drop, evaluates the state of the system and sends an alarm and/or actuates a coil in order to lock a control valve in the pipeline. The device is portable and can be supplied even by batteries and renewable energy sources. The core of the hardware system is a low power Microcontroller Unit (MCU) that receives the pressure data from an analog or digital transducer and hosts the algorithm for recording, actuating and send the data to input-output interface and wireless communication modules. The embedded algorithm compares incoming pressure data with thresholds (minimum, maximum and drop) set up by the user and actuates the coil in order to lock/unlock the valve. The hardware system can be connected with a laptop through application-dependent connections: USB interface, short-range low power wireless (e.g. Bluetooth Low Energy, BLE) or, for installation in remote unreachable sites, GSM-GPRS . The designed prototype designed receive pressure from a digital transducer data compliant with RS485 communication protocol and actuate a coil with a maximum current absorption of 0.5A. The 128Mbit flash memory allows recording of process data (time, pressure, drop for three months every second). The power consumption depends on the connected peripherals; in the Nominal State it is less than 5mW @ 3.3V for MCU data acquisition and management. The device has been successfully tested in a pipeline with pressure until 80atm and drop rate up to 35atm/min.

KEYWORDS - MCU,USB,RS485,GPS,coil,GSM/GPRS,Memory,Control System,Actuators,Digital Electronics,Embedded System,Microcontroller,Fluid



DESIGN OF A PROTOTYPE FOR MONITORING OF THE FAULTY EQUIPMENT VIA GPRS

OMER FARUK BAY1 , OSMAN OZKARACA2

1 Gazi University, Turkey ; 2 Mugla Sitki Kocman University, Turkey

ABSTRACT

Because of the rapid changes in the technology, several new equipment has been joined in daily life. Being added new ones on existent equipment has exposed the difficulties of controlling and monitoring this equipment. The purpose of this study is to develop a prototype for transferring instantly faults of the electrical equipment to technical services via General Packet Radio Service (GPRS). By using the proposed system, equipment used at home or in industry can be monitored easily. In this way, faulty equipment can be determined instantly and they will be repaired in a short time by the technical services. Thus economic loses will be prevented. The proposed system was applied and tested for washing machines in a laundry. Faults of the washing machines were constituted artificially and these faults are transferred to the local main unit using RF receiver/transmitter. And then the fault codes along with the machine code are transmitted from the main unit to the web server located in the technical service using GPRS. Received faults and machine codes are recorded in the database in the web server using a Graphical User Interface Software. The results show that the proposed system is applicable and it can be used for technical services

KEYWORDS - GPRS, remote monitoring systems,fault transfer



SHORT TERM WIND POWER FORECASTING USING ANFIS AND REGRESSION METHODS

HASAN HUSEYIN CEVIK1 , MEHMET CUNKAS2


ABSTRACT

One of the most important tasks of electrical system operator is to keep the electricity generation / consumption in a balance. This balance is achieved by using forecast values which are taken from all electric generation/ consumption participants in electricity market. The power generated from the wind is a function of the wind speed and wind speed has uncertainty because of its nature. In this study, a day ahead hourly wind power forecast which is a mandatory for wind power plants (WPP) is carried out for seven WPP’s. While the forecasted hour, wind seed forecast and wind direction forecast are the system inputs, hourly generated wind power is the forecasted value. The proposed forecasting system consists of three steps and uses adaptive neuro fuzzy inference system (ANFIS) and regression methods sequentially. In first stage, five different models are formed by using the system inputs in different configurations. Secondly five different ANFIS structures are created for each models. In last stage, the outputs of ANFIS structures are considered as the inputs of regression. One-year data is selected as train data, six mount of data are forecasted. The averaged error values of the proposed forecast system are found as % 14.89, % 10.86 for NRMSE and NMAE, respectively.

KEYWORDS - ANFIS, energy management, regression, wind power forecasting.



ANALYSIS THE SINGLE PHASE DISTRIBUTION TRANSFORMER USING NUMERICAL METHOD

ATABAK NAJAFI1 , ABDOLREZA GHAFFARLOU2 , OKAN OZGONENEL3

1 Ondokuz Mayis University, Turkey ; 2 Middle East Technical University, Turkey ; 3 Ondokuz Mayis University, Turkey

ABSTRACT

The calculation of loss has been considered as an important step in the designing of distribution transformer. The 3D finite element analysis is a numerical method that solves differential and integral equations such as magneto static and electromagnetic. In this study finite element method that improved in MAXWELL software have been used to calculate the distribution of magnetic field, ‘no-load’ and ‘load’ losses of single phase 30 kVA and 100 kVA distribution transformer. The comparison between the simulation and experimental results shows that finite element method is a precise method for calculating the transformer losses.

KEYWORDS - Distribution Transformer, Losses, FEM



JET PULSE FILTER AUTOMATION

MUCIZ OZCAN1 , NIHAT CANKAYA2

1 Necmettin Erbakan Uni, Turkey ; 2 Necmettin Erbakan Uni, Turkey

ABSTRACT

Jet Pulse Filters (JPF) are the most commonly used filters in industrial installations. These filters generally do not have an automation system. In the case of automation, the efficiency of these filters can be increased, energy consumption can be reduced and the aspiration system can be made more stable.. In this work, the necessary features of JPF automation systems are explained. An automation system with these features has been developed and tested in real use by commissioning an industrial assembly. In a JPF system, there are commonly three main equipment. These equipments are JPF, fan and air lock. In a good automation system, every 3 elements must be controlled. In this study, all these main elements are controlled and suitable automation algorithm is developed. In order for the automation system to operate properly, all of the JPF operating parameters are measured. A large number of sensors are connected to the system for measurement. A PLC-SCADA-based automation infrastructure has been added to the system. An automation algorithm has been developed to provide the desired control conditions. Real-time trends of all operating parameters are taken. As a result of the work done, the system is operated at a constant air speed, energy consumption is reduced, filter efficiency is increased, and faults in JPFs are detected early.

KEYWORDS - PLC,Automation ,jet pulse filter,scada



THE USE OF ELECTRONIC CONTROL SYSTEMS IN ROAD TRAFFIC

AYHAN GOKTEPE1 , FATIH YIGIT2

1 Selcuk Universitesi, Turkey ; 2 Kgm, Turkey

ABSTRACT

In Turkey, road traffic has been increasing rapidly in province centers with the increase of the number of vehicles. The number of motor vehicles has shown an increase of 70 % within the last decade in Turkey. Inner-city traffic control is regulated by traffic lights and traffic control units. Electronic control systems have been developed to enable the control of the largest area by using the least human resource throughout all time periods. Electronic control systems and their areas of use are discussed in this study. The characteristics and advantages of electronic control systems are presented. The present speed warning systems on Ankara-Eskisehir highway and the speed lane system in Konya province are examined. The system was set up at 5 points as 10 speed lanes round in both ways. The ten lanes with lengths ranging from 1000 meters to 3500 meters were selected from areas where accidents intensified due to high speed and where the traffic did not stop considering the last three years. The communication of the lanes was maintained through fiber optic cable network and wireless network infrastructure. The entire road was monitored by using 2 license plate reading cameras for the three lanes. The detection of the number of vehicles that pass through the zones with respect to time, the speeds of all the vehicles, and the detection of the wanted vehicles (stolen, seized, restrained) is possible through the system. The numbers of speed violations before and after the installation of the system was examined. It was seen that speed violations decreased from around 90% to around 1% considering the number of the passing vehicles. The decrease in speed violations was found to be parallel to the decrease in accidents. As the result of the assessment of the traffic regulation information in addition to traffic control enabled the solution of a large number of security-related cases.

KEYWORDS - Electronic control systems; plate reading; traffic control.



EXAMINATION AND COMPARISON OF THE COMMUNICATION PROTOCOLS ON THE APPLICATION LAYER IN IOT

CEM GULTUNCA1

1 Computer Science, Turkey

ABSTRACT

Today the Internet has become ubiquitous, reached almost every corner of the globe, and is affecting human life in unimaginable ways. We are now entering an era of even more pervasive connectivity where various appliances will be connected to the web. We are entering an era of the “Internet of Things” (abbreviated as IoT). IoT is defined as a paradigm, in which objects equipped with sensors, actuators, and processors communicate with each other to serve a meaningful purpose. Several IoT protocols have been introduced in order to yield an efficient communication for resource-constrained applications. However, their performance has not been well understood yet. This paper evaluated and compared four communication protocols, namely, AMQP, MQTT, XMPP, and CoAP. In this article implemented a some IoT application using open source software for these protocols and measured their performance

KEYWORDS - AMQP, MQTT, XMPP,CoAP,IoT,Protocols



SLIMWEAR S SMART PATCH INTERFACE DESIGN FOR LIGHTWEIGHT MIDDLEWARE IN WEARABLE SYSTEM

TAE GYU LEE1 , YONG K KIM2

1 Wonkwang University, Korea Republic Of ; 2 Wonkwang University, Korea Republic Of

ABSTRACT

We propose the smart patch interface logics and wearable interface method for light-weighting of middleware in wearable mobile computing applications. Because the previous middleware did not properly support the lightweight and mobility of mobile user terminals. The previous mobile middleware was based on a smart device (or smart phones) with the limited hardware resources of memory, CPU, display, battery, etc. However, the mobile middleware cannot be applied to a thin client such as smart patch just with lightweight sensor or feedback controller. Therefore, we present the smart configuration methods of divided middleware components and the user’s mobility interface strategy. The proposed method for mobile middleware systems called of “SlimWear-S” provides the component dynamic logic and the user interface design.

KEYWORDS - Smart Patch, Interface design, Wearable system, Context awareness, Disconnected operation



EFFECTIVE HYPERLINK SETUP ALGORITHM IN WIRELESS SENSOR NETWORK

BYUN GON KIM1 , YONG K KIM2

1 Kunsan National University, Korea Republic Of ; 2 Wonkwang University, Korea Republic Of

ABSTRACT

It is important to select the most appropriate channel for efficient transmission of massive data in wireless sensor network. This thesis suggests the algorithm to search available channel for both 2 nodes effectively. Existing type shows markedly lower performance in case the number of channels and channel interference between the other nodes increase. For improving this, we suggested random exhaustive algorithm in this thesis. Generally, beacon signal is periodically waiting on the unused channel of receiving node, in that case the channel cannot be used because of other channel’s interferences during this period. Therefore, the existing algorithm just waits the fixed time period at the first time, our suggested algorithm reduces the beacon waiting time by sensing so that the effectiveness of searching channels could be improved.

KEYWORDS - Link Setup, cognitive radio network, random protocol, Interference Link.



INTERNET CONTROLLED ELECTRONIC SCOREBOARD DESING

FUNDA DEMIR1 , BATIKAN ERDEM DEMIR2

1 Karabuk University, Turkey ; 2 Karabuk University, Turkey

ABSTRACT

Recently, when athletes are selected for sports academies and vocational high schools, racecourse completion time is based on. This system is consist of motion sensors, are used at the starting and finishing points, an electronic control unit and an electronic scoreboard. Two different observers are needed to keep athlete durations and statistics in existing scoreboard systems. In this project that can be controlled over the internet, this two units are combined and informations processed in the user interface is displayed on the scoreboard and simultaneous comparisons with other athletes' data in the database, can be made at the same time. Statistics can be transferred to internet interface and online information can be shared without losing time. The system consisting of electronic control part and user interface is very easy to use and understandable. It is suitable for athlete selection in many areas such as handball, basketball, tennis, athletics.

KEYWORDS - Scoreboard, athlete selections, motion sensor, internet network.



A REALISTIC APPROACH TO SOLVE QUAY CRANE SCHEDULING PROBLEM

YUSUF YILMAZ1 , MUSTAFA EGEMEN TANER2

1 Pamukkale University, Turkey ; 2 Pamukkale University, Turkey

ABSTRACT

With the effect of globalization, it is important to deliver the goods to offshore markets in world trade. The volume of trade in container terminals is increasing day by day and marine lines are widely used for transportation. There are many papers in scientific literature about container terminal operations, especially scheduling of quay cranes in last decade. Most of papers about quay crane scheduling problem (QCSP) assumed that processing times of tasks are given. However in this paper, processing time of each bay is calculated from the sequence of containers to be handled in the bay by hybrid tabu search algorithm (HTSA). After the processing times of bays are computed, quay cranes are scheduled by genetic algorithm (GA). Reshuffle containers are positioned on the vessel as buffer in order to increase double-cycle rate because of minimization of empty movements with HTSA. For the first time in the literature, this paper has realistic assumptions that tasks are considered as container; processing times computed by position of each container and quay crane speed. Detailed numerical experiments are conducted to evaluate the performance of the proposed method.

KEYWORDS - Container Sequencing, Genetic Algorithm, Quay Crane Scheduling Problem



TECHNICAL ENVIRONMENTAL AND FINANCIAL FEASIBILITY OF WIND TURBINE FOR SHOEBURYNESS

RIZA BUYUKZEREN1 , ALI KAHRAMAN2

1 Necmettin Erbakan University, Turkey ; 2 Necmettin Erbakan University, Turkey

ABSTRACT

Wind energy is one of the most essential renewable energy sources which is cost effective and environmental friendly to generate electricity from renewable sources. Wind energy conversation systems convert kinetic energy to electricity or other kinds of energy. There is a huge growth in wind energy technologies in the last decades. From wind energy projects, which include off-grid applications, consist of wind energy produced at remote sites, such as detached from the electricity distribution network. These type of projects are using in the charging of batteries capable of storing the energy captured by the turbines of wind farms. The energy is also available on request directly to consumer. It is a competitive method for pumping water species. Other type of wind power solution is On-Grid. These applications are connected to the electricity network and directly spread to implement energy services. It is beneficial to distinguish between two types of on-grid applications, Isolated-Grid and Central-Grid. Isolated-Grid has used in this paper. A reference energy usage data, from a company in Shoeburyness in London, was evaluated. The annual energy consumption of the company is 311MW. Wind turbine which has 275 kW power capacity and 32m rotor diameter is selected for the company wind farm. The technical analysis shows that 497MW electricity is produced with selected turbine annually. Secondly, Environmental analysis illustrates that 370.5 tCO2 emissions is achieved which is equal to 159209 litter gasoline not consumed. Last, simple payback period is calculated as 5.6 years and it is an acceptable value for this type of huge investments.

KEYWORDS - Wind Energy, Retscreen, Shoeburyness



A MODIFIED CARNEGIE MELLON ESTIMATION METHOD FOR WHITE GOODS E WASTE QUANTITY OF TURKEY

ZEYNEP OZSUT BOGAR1 , ASKINER GUNGOR2


ABSTRACT

Technological developments have been gaining speed in the last twenty years or so and it seems that these rapid changes will continue increasingly. Electrical and electronics products with new technologies make life easier yet become obsolete as their life spans become shorter and shorter due to new technology driven innovations and changes in customer expectations. Thus, each new product either pushes the old one into the e-waste stream or become e-waste in the future at the end-of-life. Therefore, the estimation of e-waste potential is important and valuable to design efficient take-back systems and e-waste management processes. In this study, a Modified Carnegie Mellon (MCM) estimation method is proposed. The method is applied to estimate the e-waste potential of Turkey. The MCM method considers the variable product lifespans instead of fixed lifespan as in the original Carnegie Mellon Method. Various scenarios are designed based on parameters like sales, lifespan and disposal percentages to further understanding of the e-waste potential. Results are obtained for selected white goods in Turkey such as refrigerators, washing machines and dishwashers. Findings contribute to the better understanding of e-waste issue which may lead to develop effective collection, estimation and management systems necessary to sustain the environment for next generations.

KEYWORDS - WEEE, E-waste, Estimation, Carnegie Mellon



NUMERICAL SOLUTION OF MAGNETOCONVECTION IN A FERROFLUID FILLED CAVITY

BENGISEN PEKMEN GERIDONMEZ1

1 Ted University, Turkey

ABSTRACT

In this study, natural convection flow is numerically investigated in a ferrofluid filled cavity under the effect of an externally applied magnetic field. Induced magnetic field is also considered. The time independent, non-dimensional governing equations are solved by differential quadrature method. Numerical results are presented both in terms of streamlines, isotherms, vorticity contours and induced magnetic field lines, and average Nusselt number. Physically, the retarding effect of the increasing Lorentz force on convective heat transfer is pronounced. Convective heat transfer is enhanced with the increase in solid volume fraction of nanoparticles. Induced magnetic field lines are perturbed with the rise in magnetic Reynolds number.

KEYWORDS - Natural convection, Ferrofluid, Induced magnetic field, Differential Quadrature Method, Magnetoconvection



A COMPARATIVE STUDY BETWEEN COMPUTER AIDED STATISTICAL APPROACH AND ARTIFICIAL NEURAL NETWORK

MUSTAFA ALTIN1 , SAKIR TASDEMIR2 , SULEYMAN NESELI3


ABSTRACT

The widespread use of computer software in various disciplines is caused to obtain rapid and reliable results. Evaluation of results, which is obtained from the experimental study, can give clear clues about the solution way of the problem. Based on these clues, a model can be created for problem-solving by approaches such as artificial intelligence, statistics, fuzzy logic etc. A thumping majority of the stock building in our country constitutes stacking structures. Brick is one of the commonly used materials from stocked building material and it is necessary to learn of knowledge about pressure resistance of these materials. In order to determine the brick compressive strengths, experimental studies are carried out and whether they comply with the regulations. In this study, two comparative approaches were used to determine the compressive strength of stock brick elements. As input parameters in the modeled two methods; weight (gr), size (width-length-height, mm) applied a load (kgf) and as an output parameter, the pressure (kgf/mm2) was selected. Computer-aided statistical linear multi-regression (SLMR) model was developed as a first model. The data obtained by statistical regression models were compared with the experimental data and determined according to the model correlation value which gives the closest results to the experimental data. Correlation value was found as R = 0.977. The second model is consisting of Artificial Neural Network (ANN) which is an artificial intelligence approach. When the data obtained from the developed ANN model were compared with experimental data, success at a high level was obtained and correlation value was found as R = 0.983. The results were obtained from using these two models, compared and it was determined that the two groups of data were compatible with each other. When comparing these two different models, it can be said that the most successful results are obtained by ANN according to the correlation coefficients. As a result, it can be seen that the model developed in this study can be used safely with predicted structural strength compressive strengths.

KEYWORDS - Statistical Approach, Regression Model, Artificial Neural Network, Pressure Resistance.



ARTIFICIAL NEURAL NETWORK FOR DETERMINATION OF PRESSURE RESISTANCE OF CONSTRUCTION ELEMENTS

SAKIR TASDEMIR1 , MUSTAFA ALTIN2 , BURAK TEZCAN3


ABSTRACT

The widespread use of computer technology has been born the usage of computer systems as decision-support-solutions. In particular widely application of machine learning algorithms on different disciplines has been cause fast, accurate and logical results. Usages of like this or different computer-aided methodology and algorithms have become almost indispensable elements. One of the Artificial Intelligent Techniques, Artificial Neural Networks (ANN), is widely used for estimating study. According to the earthquake map, also, known that the ratio of 92% surface area of our country remains earthquake zone. It is necessary to know the strengths of the stacked bricks used as the basic elements in these structures when big part of Turkey is in the earthquake zone and about 51% of the buildings stock is thought to be formed by stacking structures. For this purpose, the conformity of the bricks to the standard which published in TS EN 771-1/April 2005 was investigated. In performed experimental study, brick samples were prepared in different weights and sizes which were placed in the center of between two steel plates on the computer controlled test press. After then load was applied with a constantly increase (0.3 - N/cm²/Sn) for pressing on the upper and lower steel plates until the brick was broken. The pressure values on brick were obtained by software. ANN’s model was created using data which gather from experimental samples. As an input parameters in the designed ANN model; weight (gr), size (width-length-height, mm) applied load (kgf) and the pressure (kgf/mm2) were selected as an output parameter. In the training-test phase 162 data set (3 fold) were used for estimation of pressure value in the ANN. In the training and testing phase Mean Square Error (MSE) were calculated as 0.00000529 and 0.0113, respectively. When the developed ANN model is compared with the experimental data, success at high level is obtained and it is determined that the two groups of data are compatible with each other. As a result, it can be seen that the model developed in this study can be used safely with predicted brick compressive strengths.

KEYWORDS - Artificial Neural Network, Bricks Pressure Resistance, Stock Building, Earthquake



NUMERICAL SIMULATION OF MIXED CONVECTION FLOW IN A CAVITY FILLED WITH FE3O4 WATER

BENGISEN PEKMEN GERIDONMEZ1 , SEVDI MERVE EROGLU2

1 Ted University Basic Sciences Unit, Turkey ; 2 Ted University Department Of Computer Engineering, Turkey

ABSTRACT

In this study, mixed convection flow in a cavity filled with Fe3O4-water under the effect of a magnetic field is investigated numerically. Induced magnetic field is also taken into account, and the governing dimensionless equations in the form of stream function, vorticity, temperature, and magnetic induction terms are solved utilizing radial basis function based pseudo spectral (RBF-PS) method. The parameters in the range of Rayleigh number Ra=10^3 to Ra=10^5, Hartmann number Ha=1 to Ha=200, magnetic Reynolds number Rm=1 to Rm=100, and solid volume fraction 0 to 0.20 are performed in the computations. In numerical results, streamlines, isotherms, vorticity contours, and induced magnetic field lines are presented as well as the average Nusselt number through the heated wall. Convective heat transfer and fluid flow decrease imposing the magnetic field. The increase in the concentration of nanoparticles causes convective heat transfer to increase inside the cavity.

KEYWORDS - Nanofluid, Magnetite, Mixed convection flow,MHD convection, Radial Basis Function



A FOURTH ORDER SWITCHED CAPACITOR LOW PASS FILTER WITH DYNAMIC SWITCHING BIAS OP AMPLIFIERS

HIROO WAKAUMI1

1 Tokyo Metropolitan College Of Industrial Technology, Japan

ABSTRACT

A two-stage biquadratic 4th-order switched capacitor low-pass filter (SC LPF) employing folded-cascode CMOS operational amplifiers with a dynamic switching bias circuit (DSBFC OP Amps) capable of processing video signals, which enables low power consumption, operation in wide bandwidths and low power supply voltages, is proposed. In the newly-proposed 4th-order SC LPF, the new DSBFC OP Amps performing a dynamic on-off switching operation, appropriate for providing low power consumption while maintaining high-speed switching operation, are used. This SC LPF consists of two-stage biquadratic circuits cascading two 2nd-order SC LPFs to achieve the sufficiently sharp roll-off. In this circuit, dynamic charging operations through two-phase clock pulses during the on-state period of the OP Amps and non-charging operation during their remaining off-state period are separated. Through SPICE simulations, it was shown that the 4th-order SC LPF with an OP Amp switching duty ratio of 50 % is able to operate at a 14.3 MHz high-speed dynamic switching rate, allowing processing video signals, and a dissipated power (65.7 mW) of 68 % of that observed in the static operation of the OP Amps. The stop-band gain below -31 dB in the frequency response, which is suitable, was obtained at over 4 MHz. This is much lower than below -21 dB in the 2nd-order LPF. The simulation results confirmed that the two-stage biquadratic circuit configuration for SC LPFs using DSB OP Amps is useful for achieving low stop-band gain, and reducing the power dissipation.

KEYWORDS - Switched Capacitor Circuit, Filter, CMOS,Operational Amplifier, Dynamic Switching



DETERMINING THE OPTIMAL PORTFOLIO BY USING DATA ANALYSIS AT ISTANBUL STOCK EXCHANGE MARKET

HASAN AKYER1 , OZCAN MUTLU2


ABSTRACT

Portfolio management is an area that has gained considerable importance in today's competitive world. Decisions of investment is an important stage to sustainable development for investors. Therefore, portfolio management has gained more importance day by day. Investors want to maximize their returns at minimum risk level. Portfolio of stocks were examined by using a method of Markowitz’s mean-variance model. The purpose of the study is to examine the relationship between historical price data and maturity time during a portfolio investment stage. In the study, portfolios are constructed of stocks listed in BIST 100 Index. As a result, in order to make short-term investments using short term data give better returns than long term data. However, using at most four years historical data is enough to make long term investmenst.

KEYWORDS - Portfolio Selection,Mean Variance Model,Risk Management



ESTIMATING HOURLY SOLAR RADIATION IN TURKEY USING AIR TEMPERATURES

OZGUR SOLMAZ1

1 Manisa Celal Bayar University, Turkey

ABSTRACT

Knowing solar radiation is important in meteorology, climatology, hydrology, and renewable energy applications. Even, the performance of these systems can be estimated more accurately with hourly solar radiation. Therefore, in this study, hourly solar radiation was estimated with ANN model by using data of 5 provinces (Antalya, Konya, Mersin, Mugla, Sanliurfa) selected in Turkey. These provinces is the leading cities of Turkey in terms of solar energy. In the ANN model, Levenberg-Marquard (LM) algorithm were used for training the artificial neural network. Input parameters are chosen as latitude, longitude, altitude, day of the year, hour, and temperature at that hour of that day and the previous day while the output parameter is the hourly solar radiation. The various networks designs were tested and then the most successful network was found as tree layer network with 6 neurons in hidden layer. The RMSE, MAE and R2 results for the most successful network are respectively calculated as 0.0663, 0.0570 and 0.9870. This neural network, therefore, can be used for estimating hourly solar radiation for locations where only ambient temperature data are available.

KEYWORDS - Hourly solar radiation, Artificial neural networks, Levenberg-Marquard algorithm, Meteorological data, Air temperature.



SEGMENTATION AND APPLICATIONS OF VIDEOCAPILLAROSCOPY

KEMAL TUTUNCU1 , MUSTAFA BUBER2


ABSTRACT

Nailfold and oral videocapillaroscopy have been used for long time as a non-invasive technique for viewing peripheral circulation and diagnostic and the monitoring phases of the connective tissue diseases. Videocapillaroscopy is the becoming more and more preferable today due to have non-invasive technique and be decision support system for experts. Videocapillaroscopy is the process of imaging small veins in nail bed or oral called capillary with the help of lens and camera. The possible changes that can be seen in small veins of related regions can help some diseases such as rheumatic, diabetes and etc be diagnosed at an early stage. This paper provides a state-of-the-art review and analysis of the different existing methods of videocapillaroscopy segmentation along with some common standards and guidelines drawn from the literature. This paper includes the important videocapillaroscopy segmentation methods and the main focus is on the review of application of these methods.

KEYWORDS - Videocapillaroscopy, Segmentation, Non-invasive, Capillary, Decision Support System



PREDICTING SEMINAL QUALITY USING ARTIFICIAL NEURAL NETWORK

I UNAL SERT1 , ILKER ALI OZKAN2 , MURAT KOKLU3

1 Necmettin Erbakan University, Turkey ; 2 Selcuk University, Turkey ; 3 Selcuk University, Turkey

ABSTRACT

Nowadays, potential fertility analysis is subject of interest. These researches show that life style and life habits are also effective on semen quality together with environmental factors. Decreasing fertility rates in the last decades is rising the importance of factors, which obstruct having high fertility rates. While field of medicine needs laboratory operations for diagnosis, with appropriate machine learning methods well-liked predictions can be acquired from patient’s medical background. Artificial neural networks are widely used as decision support systems in medical field and in this study; an artificial neural network is used to predict semen quality according to patient’s backstory of environmental factors and lifestyle. Performance of the artificial neural network is evaluated using precision, recall, accuracy and F1 score metrics based on a 10-fold cross-validation scheme.

KEYWORDS - Fertility, Artificial Neural Network, Diagnosis, Prediction, Semen quality



A COMPARATIVE STUDY OF MODELING OF CHRONIC KIDNEY DISEASE USING DATA MINING

ARIF AYDIN1 , MURAT KOKLU2 , ILKER ALI OZKAN3

1 Necmettin Erbakan University, Turkey ; 2 Selcuk University, Turkey ; 3 Selcuk University, Turkey

ABSTRACT

In traditional medicine, chronic renal disease is diagnosed after medical examinations. Nevertheless, computerized diagnostic systems that will assist to the doctors are the subject of many researchers. Various data mining techniques and methods are used to support doctors' diagnostic decisions and identify them. Data mining methods help to build the best model for supporting medical decision systems, extracting information, classifying, summarizing and evaluating according to given data set. In this study, different medical decision support systems were implemented using chronic renal disease dataset which is received from UCI Machine Learning Repository, including missing values and abnormal records. Different models for this disease have been established and also the effect of different preprocessing techniques on modeling of chronic renal disease (CKD) has been examined. It was performed with different preprocessing techniques. The first dataset was processed without any preprocessing, in the second dataset the missing values were completed and the abnormal data were discarded, in the third dataset the missing values and the abnormal data were completely discarded. Seven different classification studies including Naïve Bayes, Logistic, Decorate, Sequential Minimal Optimization (SMO), J48, Ridor and Grading have been performed for the three different datasets. The results obtained with 3 different models and 7 different algorithms are compared with each other.

KEYWORDS - Keywords: Data Mining, Classification, Chronic Kidney Disease, NaiveBayes, Logistic, Decorate, SMO, J48, Ridor, Grading



THE DETECTION OF A TYPICAL RECTAL CANCER WITH SEMI AUTOMATIC IMAGE PROCESSING TECHNIQUES

ALI YASAR1 , ISMAIL SARITAS2 , HUSEYIN KORKMAZ3


ABSTRACT

Rectal cancer is a disease in which malignant (cancer) cells form in the tissues of the rectum. In this study, our aims to detect the cancerous regions semi-automatically according to the colonoscopy images obtained. In this process, semi-automatic diagnosis of cancer is done according to the colonoscopy images received from the patients using image processing techniques and the results are compared with a typical rectal cancerous region determined from the colonoscopy images by specialized doctors. A typical rectal cancerous region obtained semi automatically using image processing techniques and determined by the specialized doctors are compared. 90.747 % accuracy rate is determined from the comparison results. Considering the results obtained, it can be seen that the suspected region can be determined via software by the specialized doctors.

KEYWORDS - A Typical Rectal Cancer, Image Processing , Colonoscopy Image, Adenocarcinoma



INVESTIGATION OF OPTIUM DESIGN OF HETEROJUNCTION N ZNO P SI SOLAR CELL WITH PC1D

AYSEGUL TOPRAK1 , HAMDI SUKUR KILIC2

1 Selcuk University Kadinhani Vocational School Department Of Electronic And Automation, Turkey ; 2 Selcuk University Faculty Of Science Department Of Physics,

Turkey

ABSTRACT

Silicon is one of the most suitable materials for solar cells because it is an abundant non toxic element in the earth's crust. In this study, design and optimization of heterojunction n-ZnO/p-Si solar cell was performed by using one-dimensional solar cell simulation software PC1D. The output characteristics of the solar cell vary depending on the cell configurations, the type of materials used, the doping ratios and the region thicknesses. The influence of the thickness and doping ratios of the region on the silicon solar cell characteristics were investigated and the performance of the solar cell was determined. From this study, the ideal characteristics were determined for the best efficiency of the solar cell. The efficiency was observed under AM1.5 solar radiation, and constant intensity of 0.1 W/cm2. Simulations were estimated to have a conversion efficiency of 17.9% and a fill factor (FF) of 80%. It was determined that the thickness and doping ratios are critical parameters for heterojunction silicon solar cell performance.

KEYWORDS - PC1D, Heterojunction, Thickness, Doping



DERMATOLOGICAL IMAGE LIBRARY AND DERMATOLOGICAL WOUND TRACING APPLICATION FOR DERMATOLOGISTS

SUMEYYA ILKIN1 , OKTAY DUMAN2 , SUHAP SAHIN3

1 Kocaeli University, Turkey ; 2 Kocaeli University, Turkey ; 3 Kocaeli University, Turkey

ABSTRACT

Applications developed for mobile devices which are using the Android operating system are constantly on increasing. Along with technological advances, devices using the Android operating system have become more complicated and have more hardware capabilities that similar to the desktop computer. Therefore, applications for these devices are becoming more diverse. In addition, various automation programs can be created by providing databases and other application links to these devices. In recent years, applications developed for mobile devices have also started to be used in the medical field in order to provide convenience for doctors/patients. In this study, a mobile dermatological image library application was developed that allows dermatologists to store patients’ wound images and track dermatologic wounds. Dermatologists take images of their patients with the help of camera applications in the application. The developed application allows a dermatologist to track a patient’s wounds by making a comparison with previous images which belong to the same patient. This procedure is done by using patient images that previously recorded in the dermatologist’s account that it created on a cloud server by using this application. Thus, the dermatologist is able to examine the recovery process of his patient. Dermatologist can create a new patient record for the first time in the application as well as create a new image record for an existing patient. In addition, the application allows to dermatologist can display the patient's information or clear the patient's information. This application was written in java language and has been developed for using in the android operating system. The information was saved to a database by using the existing SQL database in Android, and an online dermatologic image library was created by using Dropbox cloud server.

KEYWORDS - Mobile Application, Android, Dermatologic Images, Wound Area Tracing, Dermatologic Image Library



A COMPARATIVE PERFORMANCE EVALUATION OF CLASSIFICATION TECHNIQUES ON CAR EVALUATION DATASET

MURAT CINIVIZ1, MURAT KOKLU2


ABSTRACT

Classification is a two step process consisting of pre-designating data classes and developing a new model on top of them to use for classification. Performance of classification algorithms are mostly evaluated by their classification accuracy. Accuracy of the classification is determined by the correctness of classified objects in relatively true class. Success of classification applications depends on the accuracy of the used algorithm so that algorithm in use has a huge effect on application’s performance. In this study, WEKA is used with car evaluation dataset from UCI machine learning datasets to compare various classification techniques. In total of 8 classification methods; multilayer perceptron, ridor, decorate, j48, kstar, bayesnet, grading and classification via clustering are used. Performances of all algorithms are compared with each other. Hence the results are, the most accurate algorithm was multilayer perceptron (99,54%) and lest accurate algorithm was classification via clustering (54,84%).

KEYWORDS - Classification Techniques, Car Evaluation Dataset, Multilayer Perceptron



INTRODUCING LABORATORY EQUIPMENTS USING AUGMENTED REALITY

NIGMET KOKLU1

1 Selcuk University, Turkey

ABSTRACT

Developments in computer, internet and mobile technologies enhanced communication and interaction in field of education thus introduced many new practices. One of those prominent technologies, augmented reality is becoming widespread in area of utilization and getting easier to use every day. Every process of augmented reality is constituted on computers. Sound effects and images created by computers attached to existing real objects or even GPS data, thereby helps presenting an enhanced content. In this study, promotional materials are prepared for physics laboratory equipment’s demonstrations using augmented reality. Aurasma and Color Mix programs are used to create these materials. In this way, promotional material became more appealing.

KEYWORDS - Keywords, Augmented Reality, Laboratory Instruments, Aurasma, Color Mix



MODELING OF PHOTOVOLTAIC MODULE AND PARAMETERS ANALYSIS OF A 335W PV MODULE

MELIH COBAN1 , MURAT FIDAN2 , OKTAY AYTAR3

1 Abant Izzet Baysal University, Turkey ; 2 Abant Izzet Baysal University, Turkey ; 3 Abant Izzet Baysal University, Turkey

ABSTRACT

The model parameters of photovoltaic (PV) cell are directly related to ambient conditions. Especially, temperature and irradiance level show big importance. The main goal of this study is to examine the equivalent circuit of PV cell and to analyze parameters with using Matlab® software program. This study also investigates effects of shunt, series resistors and diode ideality factor. Obtained I-V and P-V curves in this paper are compared to provided by X21-335-BLK model datasheet. Finally, this work can be used for numerical examples and real case on PV modules.

KEYWORDS - PV CELL MODEL, PV MODULE



OUTRUNNER ROTOR SRM DESIGN FOR ELECTRIC MOTORCYCLE

CIHAN SAHIN1 , MEVLUT KARACOR2

1 Bilecik Seyh Edebali Universitesi, Turkey ; 2 Manisa Celal Bayar Universitesi, Turkey

ABSTRACT

This work is a combination of electrical and mechanical disciplines. During the study, the minimum motor power required for the electric vehicle to travel was detected. For an electric vehicle it is desirable that the motor be of optimum strength. For this reason, it is necessary to determine the power that the vehicle needs to move by calculating the resistance that the vehicle is subjected to during its movement. Electric motorcycles have been preferred as electric vehicles in this study. The reason for the selection of motorcycles is that they are heavily preferred as a means of transportation in countries with crowded populations. The optimum motor power is calculated as 750 watts after the electric motosikletin has been subjected to resistances. If the motor efficiency is 85%, the motor power is calculated as 875 watts. Switched Reluctance Motor (SRM) was selected as the electric motor and the design and analysis were performed according to the determined power. Made design is verified using the Finite element method. The moment, current and inductance curves obtained after the simulation study are presented.

KEYWORDS - SRM, Outrunner,Electric Motorcycle,Torque Ripple



EXAMINATION OF THE ENERGY SAVING POTENTIAL OF A HYBRID CITY BUS WITH REGENERATIVE BRAKING

SEREF SOYLU1 , ZAFER DEMIR2

1 Bilecik S Edebali University, Turkey ; 2 Anadolu University, Turkey

ABSTRACT

The effects of the speed and altitude profiles of real world urban driving routes on the basic operating characteristics of a hybrid city bus were examined. The tests were conducted on the Go and Return bus routes in Karaman, Sakarya, Turkey. The following characteristics were examined: the traction and braking energies of the bus, the traction and braking energies of the motor/generator (M/G), the charging and discharging energies of the ultra-capacitor (Ucap) and the brake engine energy (BEE). The results of the examinations indicated that the braking energy, which would normally be wasted with conventional buses, corresponds to 46% and 60% of the BEE for the Go and Return routes, respectively. Under these conditions, charging the Ucap with the braking energy corresponds to 79% and 78% of the M/G braking energy for the Go and Return routes, respectively. The Return route, where the altitude decreases from 86 m to 36 m, has 12% higher braking energy and 14% less traction energy than the Go route. As a result, regenerative braking is one of the most important parameters to be considered to improve the energy efficiency of a city bus that operates with frequent and heavy braking under urban driving conditions. As the braking energy losses are recovered, the energy efficiency of the city bus can be improved significantly, because the rolling and aerodynamic (R&A) losses of the bus correspond to 27% and 34% of the BEEs only for the Go and Return routes, respectively.

KEYWORDS - energy efficiency, regenerative braking, hybrid bus, city bus, urban transport



MORPHOLOGICAL AND OPTICAL PROPERTIES OF TANTALUM TA TITANIUM TI AND MOLYBDENUM MO THIN FILMS PRODUCED BY

PULSED LASER DEPOSITION

SERAP YIGIT GEZGIN1 , AHMET TOPRAK2 , ULFET ATAV3 , HAMDI SUKUR KILIC4

1 Department Of Physics Faculty Of Science Selcuk University Konya, Turkey ; 2 Department Of Electric And Energy Bozkir Vocational School Selcuk University

Konya, Turkey ; 3 Department Of Physics Faculty Of Science Selcuk University Konya, Turkey ; 4 Department Of Physics Faculty Of Science Selcuk University Konya, Turkey

ABSTRACT

Ta, Ti and Mo thin films have received much attention in microelectronics, X-ray optics, mechanical systems or biomedical implants because of their many attractive properties, such as stability, high melting point, good resistance for corrosion and high electrical conductivity. In this study, we present some optical and structural properties of Ta, Ti and Mo thin films, which we have deposited on glass substrates by Pulsed Laser Deposition (PLD) technique by using a Nd:YAG laser in the high vacuum condition at room temperature. The PLD system has a great flexibility in adjusting the parameters for thin film production. By setting these parameters at proper values, thin films can be formed with desired morphological and optical properties. We have fabricated all thin films under the same conditions except the laser power, four different laser powers were used, 20 mJ, 25 mJ, 30 mJ and 40 mJ. After producing thin films, post-annealing process is performed by using rapid thermal annealing at 150°C, 300°C and 450°C. The produced thin films were investigated and characterized by performing Atomic Force Microscopy (AFM), X-Ray Diffraction (XRD) and Ultraviolet-Visible (UV-VIS) absorption spectroscopy. As the laser power and annealing temperature were changed, the thickness, optical properties, morphological properties and crystal structure of the thin films changed and these properties were investigated and interpreted in detail. As a result, these parameters will be referenced in our future studies like a solar cell and thin film transistor (TFT).

KEYWORDS - PLD, Thin film



DRIVE SYSTEM DESIGN FOR MUTUALLY SWITCHED RELUCTANCE MOTOR

MEVLUT KARACOR1 , CIHAN SAHIN2 , FERIHA ERFAN KUYUMCU3

1 Manisa Celal Bayar Universitesi Hasan Ferdi Turgutlu Teknoloji Fakultesi Mekatronik Muhendisligi Bolumu, Turkey ; 2 Bilecik Seyh Edebali Universitesi Bilecik Meslek

Yuksek Okulu Elektronik Ve Otomasyon Bolumu, Turkey ; 3 Kocaeli Universitesi Elektrik Muhendisligi Bolumu, Turkey

ABSTRACT

In this study, Mutually Coupled Reluctance Motor (MCSRM) to reduce torque ripple, a new drive system and control algorithm is proposed. According to Clasical Switched Reluctance Motors, which are capable of producing higher torque MCSRMs, due to the high rate of ripple of torque, have not received due attention in industrial applications. Torque produced by the ARM, high rate ripple in order to reduce the problem,various proposals presented in the literature. Studies have generally concentrated within the framework of the geometrical structure of amendment to reduce the torque ripple. Made some changes to the structure of the machine to resolve the issue, there has been limited in terms of the solution for the machines. In this study, analyzing the results of analysis of electric motor static torque, produced the "m-curve" is controlled with multi-level drive system is introduced. In the classical control, whereas the torque ripple 8.97%,with the proposed new controller system 4.82%, the value has been reduced.Between classical control and the proposed control method with a value close to 50% of the torque ripple is reduced.Efficiency is 64.1% of classical control ', the m-curve in the control increased to 67.7%.

KEYWORDS - Short pitched, Full pitched, SRM, Torque ripple



DESIGN ANALYSIS AND MANUFACTURING OF MULTIPURPOSE CYLINDRICAL SHAPE ROBOT

HILMI SAYGIN SUCUOGLU1 , ISMAIL BOGREKCI2 , PINAR DEMIRCIOGLU3 , OGULCAN TURHANLAR4

1 Adnan Menderes University, Turkey ; 2 Adnan Menderes University, Turkey ; 3 Adnan Menderes University, Turkey ; 4 Adnan Menderes University, Turkey

ABSTRACT

In this study, cylindrical shape mobile robot was designed, analysed and produced with rapid prototyping technique. Design and Analyses stages were carried out using Computer Aided Design (CAD) and Computer Aided Engineering (CAE) tools. The shape of the robot was selected as cylindrical for multipurpose tasks. It can serve as a hobby and education robot for children and also used for security and military applications with small upgrades. Analytical calculations, structural and dynamic analyses were conducted to find torque and power used for both actuator selection and determination of structural strength of the robot. Fused Deposition Modelling (FDM) and Polylactic Acid (PLA) material were selected for manufacturing light weight, cheap and strong robot. The weight of manufactured parts and commercially available components were about 1.5 and 1 kg, respectively. In the first part of the calculations and analyses; the density of PLA, torque and power requirements, and loads on cylindrical robot were calculated analytically using Engineering Equation Solver (EES). In Ansys, a road was designed with 12 meters straight, 18 meters with 10° inclination and 12 meters straight, respectively for Rigid Body Dynamics analysis tool to find required torque values. The maximum velocity of robot was defined as 4 m/s. Then, reaction forces and torques obtained from calculations were applied to the cylindrical robot body to calculate strength of the robot in dynamic conditions with Finite Element Analysis (FEA) method. Torque and power requirements were found as about 0.7 Nm and 21 W per wheel, respectively. The results showed that safety factors were almost 15 and 3.4 for whole body and critical component (motor-wheel connector), respectively. The results imply that robot can execute its tasks safely under dynamic conditions.

KEYWORDS - Cylindrical Robot, Finite Element Analysis, Fused Deposition Modelling, Rapid Prototyping, Rigid Body Dynamics



AN INVESTIGATION ON A TYPE OF FUZZY MULTIFUNCTION

AHU ACIKGOZ1 , FERHAT ESENBEL2

1 Balikesir University, Turkey ; 2 Balikesir University, Turkey

ABSTRACT

In this document, we extend the concept of strongly na precontinuous multifunction to fuzzy topology and discuss its characterizations. Also, the definitions of fuzzy strongly continuous multifunction, fuzzy semi strong na continuous multifunction, fuzzy na continuous multifunction, fuzzy super continuous multifunction[11], fuzzy strongly semi-irresolute multifunction[4], fuzzy strongly M-precontinuous multifunction, fuzzy strongly α-irresolute multifunction[4] are given and the relationships of strong na precontinuous multifunction with these types of fuzzy continuous multiunctions are also investigated and we obtain a diagram. Moreover, properties of strong na precontinuity about restriction mutlifunction, graph multifunction, composite multifunction are discussed on some fuzzy spaces and product spaces.

KEYWORDS - Fuzzy topology, fuzzy preopen set, fuzzy δ-open set, fuzzy upper (lower)strongly na-precontinuous, fuzzy upper (lower) strongly M-precontinuous.



A RESEARCH ON A SORT OF FUZZY MULTIFUNCTIONS

AHU ACIKGOZ1

1 Balikesir University, Turkey

ABSTRACT

In this study, we expand the concept of na-continuous multifunction to fuzzy topology and discuss its characterizations. Also, the definitions of fuzzy strongly continuous multifunction, fuzzy α-irresolute multifunction [7], fuzzy super continuous multifunction [10], fuzzy continuous multifunction [15] are given and the relationships of fuzzy na-continuous multifunction with these types of fuzzy multivalued maps are also investigated and we obtain a diagram. Moreover, properties of this function about restriction mutlifunction, graph multifunction, composite multifunction are discussed on some fuzzy spaces and product spaces.

KEYWORDS - fuzzy regular-open sets, fuzzy α-open sets, fuzzy δ-open sets, fuzzy upper (lower) strongly continuous, fuzzy upper (lower) na-continuous.



DETERMINATION OF TOOL GEOMETRY EFFECT ON TOOL VIBRATION USING ARTIFICIAL NEURAL NETWORK IN TURNING OPERATION

SULEYMAN NESELI1 , HACI SAGLAM2 , SULEYMAN YALDIZ3 , MUSTAFA ALTIN4

1 Selcuk University, Turkey ; 2 Selcuk University, Turkey ; 3 Selcuk University, Turkey ; 4 Selcuk University, Turkey

ABSTRACT

Cutting tool vibration is a complex phenomenon that often appears in the process of metal cutting and become a common limitation to productivity and product quality. Since the vibration is the main factor for resulting in manufacturing and machine faults such as machine spindle, cutting tool, or workpiece and leave behind a poor quality machined surface. So the analysis of cutting tool vibration in the cutting process has become necessary for improving the operational efficiency. The focus of the present study is to find a correlation between tool geometry and cutting tool vibrations in turning. The cutting tool vibration signals were collected through a data acquisition system supported by Matlab software. To increase the reliability of the experiments, L27(313) Taguchi Orthogonal Design was used. On the completion of the experimental test artificial neural network (ANN) has been modeled using the data obtained to predict the tool vibration. Multilayer perceptron neural network model has been constructed with feed forward back-propagation algorithm using the acquired data. In the training and test phase Mean Square Error (MSE) were calculated as 5.0871x10-33 and 0.0149, respectively. Experimental data collected were tested with analysis of variance (ANOVA) to understand the influence of the cutting parameters. Comparison of the experimental data and ANN results shows that ANN has been used confidently for this type prediction process. This work may help new workers to avoid operating machine tools incorrectly, and hence can decrease manufacturing costs. It is expected that this study can enhance the performance of machine tools in metalworking sectors.

KEYWORDS - Tool geometry, Tool vibration, ANN, ANOVA, Turning.



THE THREE DIMENSIONAL ELECTROMAGNETICANALYSIS OF POWER TRANSFORMERS BASED ON ANSYS

TUGCE DEMIRDELEN1 , SELVA BAL2 , MEHMET TUMAY3 , MEHMET UGRAS CUMA4 , ABDULLAH CICIBAS5 , KEREM KOSEOGLU6 , MAHMUT AKSOY7

1 Adana Science And Technology University, Turkey ; 2 Iskenderun Technical University, Turkey ; 3 Adana Science And Technology University, Turkey ; 4 Cukurova University, Turkey ; 5 Best Transformer, Turkey ; 6 Best Transformer, Turkey ; 7 Best Transformer,

Turkey

ABSTRACT

The transformer is an electrical device that transfers energy from one circuit to another purely by magnetic coupling. The transformer is one of the well-known electrical devices due to its simplicity, reliability, and economy of conversion of voltages. Its basic design, materials, and principles have changed little over the last one hundred years, yet transformer designs and materials continue to be improved. In power transformer, insulation performance of the main insulation is an important factor for transformer reliable operation, while the end insulation is the most important part of the main insulation. The end insulation of power transformer will affect the reliability of power transformer, and also affect the economy of power transformer, that is manufacture cost. As a general software on finite element, ANSYS is widely used in real world application. Many researchers have been done on power transformers, voltage transformers and current transformers using ANSYS, among which however seldom were about the electromagnetic interface on the above mentioned on transformer. In this paper, the three-dimensional electromagnetic analysis of power transformers based on ANSYS researches will be investigated in detail.

KEYWORDS - Power transformer,ANSYS,energy transfer,electromagnetic simulation,transformers



DESIGN OF MASTER AND SLAVE MODULES ON BATTERY MANAGEMENT SYSTEM FOR ELECTRIC VEHICLES

ARDA KILIC1 , SELIM KOROGLU2 , AKIF DEMIRCALI3 , SELAMI KESLER4 , YUSUF ONER5 , ERKAN KARAKAS6 , PETER SERGEANT7

1 Pamukkale University, Turkey ; 2 Pamukkale University, Turkey ; 3 Pamukkale University, Turkey ; 4 Pamukkale University, Turkey ; 5 Pamukkale University, Turkey

; 6Pamukkale University, Turkey ; 7 Ghent University, Belgium

ABSTRACT

Nowadays, electric vehicle usage and the use of LiFePO4 batteries in electric vehicles gradually increase. However, there are important features to be considered to use these batteries safely and efficiently. Incorrect use of these batteries can lead to burning, explosion or shortening of the life of batteries. In this paper, a Battery Management System (BMS) for lithium based batteries is designed that operates more efficiently and communicates with UART between master and slave modules and can communicate via CAN protocol with external devices. Micro controller based control and protection equipment is designed that help to measure and monitor the voltage, temperature and current values of the batteries. They protect the battery cells from the conditions such as over charge, over discharge, high current, high temperature. BMS balances battery cell voltages during charging process with passive cell voltage balancing. In addition to the main controller module in the BMS, slave controller modules have been added to provide high resolution voltage and temperature tracking. A modular BMS has been devised which can be used in groups of batteries of different voltage values thanks to electrically isolated slave control modules.

KEYWORDS - Battery management system, battery monitoring, cell balancing, LiFePO4 batteries,electric vehicle



RF TELEMETRY DESIGN FOR ELECTRIC VEHICLE

HANDE NEVIN KILICKIR1 , AKIF DEMIRCALI2 , SELIM KOROGLU3 , SELAMI KESLER4 , MUSTAFA TUMBEK5 , ARDA KILIC6

1 Pamukkale University, Turkey ; 2 Pamukkale University, Turkey ; 3 Pamukkale University, Turkey ; 4 Pamukkale University, Turkey ; 5 Pamukkale University, Turkey

; 6Pamukkale University, Turkey

ABSTRACT

Electric vehicles are promising solutions to today’s transportation and environmental problems. To improve the electric vehicles and use them safely could be possible with the online monitoring and recording the important parameter of the vehicle at the whole operating period. However, it is a complicated problem and has some restrictions about the range, reliability and stability. In this work, a telemetry system is presented to overcome these struggles. The proposed system uses the RF signals to transfer the data and reliable and stable ARM microprocessors to proceed communication assignment. Functional operation of the proposed telemetry system is tested in the competition of International Tubitak Efficiency Challenge Electric Vehicle and the results are presented in this paper.

KEYWORDS - Electric vehicles, RF signals, Telemetry, CAN communication, System design



AN INTERACTIVE TURKISH SIGN LANGUAGE LEARNING GAME USING LEAP MOTION CONTROLLER

FIDAN KAYA GULAGIZ1 , HIKMETCAN OZCAN2 , SUHAP SAHIN3

1 Kocaeli University, Turkey ; 2 Kocaeli University, Turkey ; 3 Kocaeli University, Turkey

ABSTRACT

With the development of technology, the control of devices has become possible without touching anything. Leap Motion controller which is developed for this purpose is spent the control process beyond touching and it has been widely used in many areas. In this study, a desktop application aimed at teaching Turkish sign language is presented and leap motion controller is used to detect hand movements of users. Communication and sign language is extremely important for people with hearing impairments. Learning the sign language with the developed game based on Leap Motion controller has become very easy. At the end of the study, the usability analysis of the application is made in terms of users. For this purpose, SUMI (Software Usability Measurement Inventory) method is used and the results were analyzed.

KEYWORDS - Sign Language,Leap Motion Controller,SUMI



STATE SPACE MODELLING AND SIMULATION FOR PEDELEC

OKAN UYAR1 , MEHMET CUNKAS2

1 Dept Of Electric Electronic Engineering Technology Faculty Selcuk University, Turkey ; 2 Dept Of Electric Electronic Engineering Technology Faculty Selcuk University, Turkey

ABSTRACT

This paper presents the mathematical model of the electric assisted bicycle also called pedelec. This model includes the vehicle and the electric motor dynamics and all equations in the model. Model equations are constructed in the Simulink in the time domain instead of s-domain. The main benefit of this method is the ability to simulate different e-bikes characterized by different parameters such as rolling resistance, chain gear ratio and road gradient. PID control is used for testing the state-space based model with constant but updatable parameters and using different road slopes obtained from Selçuk University Campus map. The cruise control for pedelec is achieved with 2 per mille error as mean value.

KEYWORDS - electric assisted bicycle, pedelec, time domain, modelling,state space,power-assist strategy



PERFORMANCE ANALYSIS OF SINGLE PHASE INDUCTION MOTOR FOR DIFFERENT ROTOR SLOT TYPES

AHMET FENERCIOGLU1 , MERVE SEN KURT2 , MEHMET AKAR3 , UNAL KURT4

1 Fen Bilimleri Enstitusu, Turkey ; 2 Fen Bilimleri Enstitusu, Turkey ; 3 Fen Bilimleri Enstitusu, Turkey ; 4 Fen Bilimleri Enstitusu, Turkey

ABSTRACT

The Single Phase Induction Motor (SPIM) is widely used in house hold appliances because of simple structure, low cost, without driver, line started and fault-free. But its efficiency and starting performance are lower than three phase induction motors and synchronous motors. It needs starting equipment and capacitor to start. SPIM’s rotor slot geometry can be optimized to improve efficiency and starting performance. Rotor slot types are significant indicator for output performance. In this study; a SPIM is modeled with two poles, rated speed is 2950 rpm, output power is 100 W, it is connected to 220 V 50 Hz AC network under the full load. Rotor slots which are designed as pear, tapered and trapezoidal shaped analyzed by 2D Finite Elements Method (FEM). Transient and steady state operations are taken into considiration in analyses. Simulation software is Ansys Maxwell EM. SPIM’s electrical and electromagnetic parameters which are efficiency, output power, rated speed, electromagnetic torque, windings currents, power factor are solved and compared with FEM analysis for each slot type. Magnetic flux density (B) and flux lines in cores and current density (J) in rotor bar are demostrated visually.

KEYWORDS - Single phase induction motor, Rotor slot types, Rotor slot geometry , FEM analysis



LQR REFERENCE TRACKING CONTROL OF BOEING 747 100 LONGITUDINAL DYNAMICS WITH CG SHIFTS

EZZEDDIN ELARBI1 , DINA LAILA2 , ABDULHAMID GHMMAM3 , NADJIM HORRI4

1 Tripoli University, Libyan Arab Jamahiriya ; 2 Coventry University, United Kingdom ; 3 Tripoli University, Libyan Arab Jamahiriya ; 4 Coventry University, United Kingdom

ABSTRACT

The stick-fixed longitudinal flight of Boeing® 747-100 is compared at three Mach-altitude conditions: 0.2 at sea-level; 0.5 at 20,000ft; and 0.9 at 40,000ft, as centre of gravity (CG) shifts. Velocity, pitch rate, attitude and altitude with elevator and throttle commands have been simulated for each flight condition. These simulations indicate stable complex eigenvalues with low stability margins. Stick-fixed flight is dominated by insufficiently damped long-period responses with an initial short-period effect. Dynamic stability and performance of the stick-free autopilot is investigated using a linear quadratic regulator (LQR) algorithm in the presence of CG shifts. Steady state convergence is achieved within a few seconds with acceptable flying qualities. The controller has been revised to cope with Mach-altitude constraints and demonstrates autopilot ability to mainain stable cruise flight. The results emphasise the plane is dynamically stable and satisfactorily trimmed at those CG cruise conditions.

KEYWORDS - Boeing® 747-100 longitudinal stability; coupling states with elevator and throttle; stick-fixed responses; LQR stick-free autopilot; CG-shifts flight conditions



SEAMLESS INTEGRATION OF CPS BASED FACTORIES ACROSS VARIOUS ENGINEERING AND SIMULATION DISCIPLINES

STEPHAN WEYER1 , CARSTEN HARMS2 , MARTIN RUSKOWSKI3

1 German Research Institute For Artificial Intelligence Gmbh Dfki, Germany ; 2 German Research Institute For Artificial Intelligence Gmbh Dfki, Germany ; 3 German Research

Institute For Artificial Intelligence Gmbh Dfki, Germany

ABSTRACT

Smart Factories are characterized by their scalability and modularity whereby field devices and production units will turn into so-called Cyber-Physical-Systems (CPS) and enable dynamic adjustments during running production. This paper outlines an approach for a seamless integration of future CPS-based factories across various engineering and simulation disciplines. They are intended to support and evaluate such (re-) engineering processes but require relevant model data about modifications made. This includes the identification, recognition of available CPS-based units as well as their model-based integration. For this purpose, a generic data model and vendor-independent system architecture has been developed. Across a varied set of digital tools it allows data to be enriched and used as needed for each CPS. The objects’ characteristic models and data can be im- and exported across various departments. In a final step, the approach is being demonstrated within an industry-relevant use case in the SmartFactoryKL. Besides selected results of the prototypical implementation, an insight into the ongoing work is presented.

KEYWORDS - Automation ,Distributed Systems,Future Manufacturing,Design Tools and Modelling,Actuators and Sensors,Electronic Components ,Simulation in Production,Semantic Descriptions,Artificial Intelligence



AN APPLICATION TO IMPROVE CIRCUIT PERFORMANCE OF A MULTIFUNCTION ANALOG FILTER USING INVERTING COMPOSITE

CURRENT CONVEYOR

HASAN CICEKLI1 , AHMET GOKCEN2

1 Mustafa Kemal University, Turkey ; 2 Iskenderun Technical University, Turkey

ABSTRACT

This paper presents an application that improves the circuit performance of a multifunction analog filter by employing second generation inverting composite current conveyor (ICCCII) as active element in the same circuit topology which is employing second generation inverting current conveyor (ICCII). ICCCII used in this work is implemented by using two current conveyors as sub-circuit. The validity of the presented application is verified and a comparative performance analysis is done by PSPICE simulation program using MOSIS 0.35 μm CMOS process parameters . Simulation results are included to demonstrate that employing ICCCII improves the performance and working accuracy of the circuit and reduce the distortion when compared to the same topology employing ICCII.

KEYWORDS - Second generation inverting current conveyor, Composite current conveyor, Multifunction filter,CMOS ,Circuit Simulation



HIGH EFFICIENCY GASEOUS ACTIVE WATER COOLING SYSTEM

SELMAN TASAR1 , KEMAL TUTUNCU2

1 Adil Karaagac Mesleki Ve Teknik Anadolu Lisesi, Turkey ; 2 Selcuk University, Turkey

ABSTRACT

Although there are lots of sizes and types of hermetic compressors used in gaseous cooling systems according to the needs, high consumption of energy of the smallest sized ones even poses a big obstacle in production of portable refrigerator. A hermetic compressor consuming less energy than existing hermetic compressors having the smallest structure, can work with battery for a long while with portable dimensions is designed with the aim of contributing to solution of the relevant problem in this study. Prototype production of the compressor designed is done after being simulated. Results of the test on the prototype and necessary data to reach sufficient pressure are compared on a graph

KEYWORDS - Gaseous cooling, portable refrigerator, hermetic compressor, energy, prototype



ESTIMATION OF TURKEY S ELECTRICITY CONSUMPTION FOR 2023 BY USING MULTILAYER PERCEPTRON S MLPS

SEMIHA DURGUN1 , ALI OSMAN OZKAN2

1Necmettin Erbakan University, Turkey; 2Necmettin Erbakan University, Turkey

ABSTRACT

Accurate estimation of electricity for consumption is one of the important issues in energy planning of developing countries In this study, the demand for 2023 Turkey's electrical energy consumption has been estimated by using the. Multilayer Perceptron’s (MLPs). The supervised learning problem of the MLP can be solved with the back-propagation algorithm, MLP, consist of three layers; these are the input layer, the hidden layers and the output layer. For this aim, various variable such as Gross Domestic Product (GDP) between the years of 1980-2015, population and meteorological data were used in the prediction model. By using the model, Turkey's electrical energy consumption between the years of 2016-2023 was estimated.

KEYWORDS - Electricity consumption forecasting, Multilayer Perceptron’s



MINIMUM COST CONTROL OF PUMPING SYSTEMS WITH SCHEDULED ELECTRIC TARIFFS

ROBERTO SANCHIS LLOPIS1 , IGNACIO PENARROCHA ALOS2

1 University Jaume I, Spain ; 2 University Jaume I, Spain

ABSTRACT

This paper studies the optimization of the operation of a water supply pumping system by means of standard solvers. The system consists of several tanks that supply the water to several districts in a town. Each tank can be filled from several wells through a hydraulic system that can be reconfigured by means of several valves. The automatic operation of the system tries to determine which valves and pumps must be active at each instant in order to minimise the operation cost, taking into account the tariff periods. A mathematical model of the problem is proposed in order to formulate, in matrix form, the cost index and the constraints, to be able to use standard solvers as Mosek or CBC. It is a mixed integer optimization problem with a high computational cost. To reduce the overall computational cost, it is proposed to reduce the number of integer variables, replacing them with real variables from a given moment in the optimization horizon. A simple output flow prediction strategy is proposed. A simulation example based on a a real water supply system model is analysed to demonstrate the validity of the approach, using Yalmip as parser and CBC as solver.

KEYWORDS - pumping optimization, electric tariff, pumping scheduling



A MULTI CRITERIA DECISION MAKING APPROACH FOR GREEN SUPPLIER SELECTION PROBLEM

ZEHRA DURAK1 , ELIF OZGORMUS2


ABSTRACT

Today’s environmental conscious world forces companies to be more sensitive about energy and water consumption. Customers are highly conscious and sensitive about protecting the environment. Consequently, companies cannot ignore environmental issues if they want to maintain their competitive advantage in this globalization trend. As a critical decision of companies, supplier selection problem needs to consider not only economic factors, but also the environmental factors. This paper presents a well-known multi criteria decision making approach: fuzzy The Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) for green supplier selection problem. The proposed model is illustrated with a case study from Turkey and the results are compared with conventional fuzzy axiomatic design.

KEYWORDS - Green Supplier Selection, Fuzzy TOPSIS,Fuzzy Axiomatic Design



IMPROVEMENT OF PROJECTION ERROR EVALUATION FOR MASSIVE DATA SETS

KOTRYNA PAULAUKSIENĖ1 , OLGA KURASOVA2

1 Institute Of Mathematics And Informatics Vilnius University, Lithuania ; 2 Institute Of Mathematics And Informatics Vilnius University, Lithuania

ABSTRACT

When the dimensionality of initial data set is reduced by different methods they need to be compared. The most common projection evaluation measure – projection error has been investigated. A shortcoming of main projection error calculation ways is such that they require a large amount of computer memory or computation time is not acceptable when massive data sets are analysed. The problem arises because huge distance matrices (or vectors) are used in projection error calculation and they require large memory resources. The goal of this research is to present and explore an improvement of the projection error calculation for massive data sets using only a common personal computer without any particular technologies. In this research, we propose to calculate the projection error by dividing the data set into the smaller data sets. The proposition can be summarized as follows: 1) the initial data set and the data set of the reduced dimensionality are divided into the smaller data sets; 2) Euclidean distances between instances are found; 3) intermediate calculations are performed; 4) the projection error is obtained. The experimental investigation with real and artificial data sets has shown that the projection error by dividing the data set into the smaller data sets allows us to calculate the projection error for massive data sets. For example, dividing the data set into the smaller data sets allows us to process the data set of 450 000 instances in an appropriate time (1 hour and 14 minutes) comparing to the basic way when the projection error is calculated using the loop for each data point (7 hours and 18 minutes). The proposed way of projection error calculation decreases computation time and avoids running out of a computer memory. The proposed way doesn’t influence the projection error value i.e., it remains the same as it would be calculated for not divided data set.

KEYWORDS - projection error, massive data set, dimensionality reduction



DESIGN AND IMPLEMENTATION OF E LEARNING WEB PAGE FOR CHILDREN WITH INTELLECTUAL DISABILITIES

HANKYU LIM1 , YERIN JEONG2 , JAEJUN JANG3 , HYEMI DO4 , SYAOCHING JOU5 , EUNJU PARK6

1 Andong National University, Korea Republic Of ; 2 Anu, Korea Republic Of ; 3 Anu, Korea Republic Of ; 4 Anu, Korea Republic Of ; 5 Anu, China ; 6 Anu, Korea Republic Of

ABSTRACT

The e-Learning market is growing rapidly along with the development of electronic media based on the Internet. E-learning not only enables learning using multimedia but also may rapidly reflect demand on or content of new education. Accordingly, this paper designed and embodied an e-Learning web page for intellectually retarded children in order to complement the current status of insufficient education on the disabled in Korea. HTML (Hyper Text Markup Language) 5 Canvas was employed so that its embodiment was made possible in the web without installation of a platform or plug in. In addition, the web was produced as a responsive web, enabling content delivery while maintaining the layout in diverse platforms, thereby improving convenience of the user and making it possible for the user to utilize it without constraints. The e-learning webpage produced in this paper is expected to enhance mentally retarded children's educational effects and quality by utilizing their leisure time after school.

KEYWORDS - e-Learning, HTML5, Responsive Web, Design of e-Learning Contents, HTML5 API



A MULTI OBJECTIVE MODEL FOR GREEN ORDER ALLOCATION PROBLEM IN MULTI ITEM MULTI SUPPLIER ENVIRONMENT

AHMET ALP SENOCAK1 , HACER GUNER GOREN2


ABSTRACT

The role of purchasing departments has become more important in success of organizational functions due to increasing competition, globalization and specialization in recent years. Supplier selection and order allocation process, which deploys a vast amount of company’s financial resources, involves much more than ranking the suppliers based on traditional economic approaches. With the growing pressure by governments and customers on companies to be green, firms have started to consider environmental criteria along with economic criteria in decision-making process. This study presents a fuzzy multi-objective linear programming model for green order allocation problem. The problem consists of four objective functions that aims to minimize the net cost, the number of rejected products, the number of late deliveries and greenhouse gas emission level. To gain more realistic outcomes for the solution and add vagueness into mathematical model, fuzzy logic has been utilized. Moreover, the changes in the weights of the objective functions have been analysed in order to present alternative solutions to decision-maker for various scenarios.

KEYWORDS - Multi-objective linear programming, fuzzy set theory, order allocation, greenhouse gas emission, food industry



PDM IC MPPT CONTROLLED SERIES RESONANT CONVERTER FOR PV GENERATION SYSTEMS

AKIF KARAFIL1 , HARUN OZBAY2 , SELIM ONCU3

1 Bilecik Seyh Edebali University, Turkey ; 2 Bilecik Seyh Edebali University, Turkey ; 3 Karabuk University, Turkey

ABSTRACT

In this study, pulse density modulation (PDM) controlled full bridge series resonant converter is simulated at PSIM for 600 W PV system. Zero current switching (ZCS) is provided and therefore switching and electromagnetic interference (EMI) losses are decreased and the circuit is operated at high frequency. The maximum power transferred from PV panels is obtained by the proposed phase locked loop (PLL) based 8 PDM control technique at 100 kHz switching frequency. Series resonant converter is operated at constant frequency and maximum power is transferred from PV panels at ZCS condition by the proposed control technique. Incremental conductance (IC) technique is used as maximum power point tracking (MPPT) technique. The control of PDM-IC MPPT algorithm and resonant converter is provided by DSP.

KEYWORDS - PDM, resonant converter, ZCS, MPPT, PLL,IC



SLIDING MODE CONTROL FOR A MIMO NONLINEAR COUPLED TANK SYSTEM

INAYET OZGE AKSU1 , RAMAZAN COBAN2

1 Adana Science And Technology University, Turkey ; 2 Cukurova University, Turkey

ABSTRACT

The work is aimed to control a Multi Input Multi Output (MIMO) system, which is a nonlinear and uncertain system, using the conventional Sliding Mode Control (SMC) technique. This paper defines the level tracking control of two input two output system. For this, the SMC method, which is a robust method against model parameter variations and noises, is used. The SMC method has been applied to the tank system in the MIMO and interconnected structure. The method is used to the coupled tank system to control the liquid level. At the end of the study, the experimental results obtained are shown graphically.

KEYWORDS - conventional Sliding Mode Control (SMC), multi-input multi-output systems, coupled tank system, nonlinear system, experiment



DESIGN AND FEM ANALYSIS OF LOW VOLTAGE OUTER ROTOR LINE START PERMANENT MAGNET SYNCHRONOUS MOTORS WITH

DIFFERENT MAGNET ALIGNMENTS

MUSTAFA TUMBEK1 , SELAMI KESLER2 , YUSUF ONER3

1 Pamukkale University, Turkey ; 2 Pamukkale University, Turkey ; 3 Pamukkale University, Turkey

ABSTRACT

— In this study, an outer rotor low voltage Line Start Synchronous Motor is designed and simulated by using Finite Element Method for electric vehicles to be used instead of asynchronous and synchronous motors. First, asynchronous motor design, one of the methods for determining initial parameters of LSSM in literature, is carried out and the magnets are placed in various size and alignments. The most suitable motor geometry and magnet shape is determined by evaluating the simulation results. The results of simulations show that the desired output dynamics are achieved by taking into consideration the motor performance presented in the literature.

KEYWORDS - Line Start Synchronous Motor, Motor Design, Synchronization Problem, Electric Vehicle



IOT (INTERNET OF THINGS) BASED HOME SECURITY AND AUTOMATION MODEL WITH A WEB AND MOBILE APP

FATIH SADAY1 , ABDULKADIR SADAY2

1 Karatay University, Turkey ; 2 Selcuk University, Turkey

ABSTRACT

Internet of Things (IoT) based control systems allows us to implement many kind of home automation system that can be controlled remotely through internet. The proposed system can monitor, control and follow different parameters like gas, light, motion detection, temperature and more using the sensor datas and also trigger a process according to the requirement of user. Collected datas from the sensors are uploaded to a cloud server by an interface and these datas can be used to analyse any type of the parameters. This home security and automation system continuously monitors the activities and environment in the home and if there is any harmful activity it is intimated to the owner of the system. The sensor data is stored in the cloud by using internet. Smart homes or automation is one of the main applications of IOT and a low cost home monitor and automation system model with a web and mobile app is presented in this paper. The whole system has an embedded web server with fully internet connectivity. An Android application can be used to remotely control and monitor the home appliances over the internet. For continuous connectivity, either mobile networks like 3G/4G or Wi-Fi can be used for this system. The IOT based home automation system uses GSM and internet for controlling the home appliances remotely. It can be monitored and controlled from anywhere in the world. Datas from the web server are converted into GSM commands by an interface and sent to the receiver. This controls the appliances.

KEYWORDS - IoT,home automation,home security,mobile-app,web-app



THE USE OF ARTIFICIAL NEURAL NETWORKS FOR MODELING THE REMOVAL OF CR VI WITH POLYMERIC CALIX 6 ARENE

ABDULLAH ERDAL TUMER1

1 Necmettin Erbakan Universty, Turkey

ABSTRACT

A sorption process requires costly equipment, expertise and time. The most important jobs in the sorption process are modeling and optimization to the process performance at the lowest cost. One way of low cost modeling is the use of artificial neural networks. In addition, both ANNs have been used for modeling of sorption process in many studies and ANN models have provided reliable and powerful results. In this study, an artificial neural network-based model was developed to modeling the removal efficiency of calixarene for Cr(VI) in aqueous solutions. Nineteen experimental datasets were used to develop the ANN model. Optimum parameters (initial concentration of Cr(VI), adsorbent dosage, contact time and pH) have been determined before, so they have been used for the training and simulation of the network in the current work. The training results were tested using the input data (simulated data) that were not shown to the network. The training was conducted using feed-forward backpropagation artificial neural network architecture. Models were designed to generate the minimum mean absolute error (MAE) and root mean square error (RMSE), and the maximum coefficient of determination (R2) (measures the model performance) values, with 4-7 neurons in hidden layers, and tansig, logsig, and purelin transfer functions. Network training data and simulated data were compared, in order to confirm the designed models. In this architecture, 2 hidden layers were kept constant. The network was trained with different neuron numbers (4,5,6 ve 7) and different transfer functions. According to the indicator (the highest R2 and the lowest MSE), the optimum and most reliable model was found based on the test results. In the forecast model developed for removal efficiency, the transfer function was found as logsig and the number of neurons was found as 5.

KEYWORDS - Artificial Neural Network; Modelling; sorption; removal efficiency; sorption capacity



SILICON RECOVERY IN WASTE CRYSTALLINE SILICON SOLAR CELLS BY A SINGLE CHEMICAL ETCHING

TERUAKI MATSUBARA1 , KEI TAKAMI2 , MD AZHAR UDDIN3 , YOSHIEI KATO4 , TAKANORI KAWANISHI5 , YOSHIAKI HAYASHI6

1 Graduate School Of Environmental And Life Science Okayama University, Japan ; 2 Graduate School Of Environmental And Life Science Okayama University, Japan ; 3Graduate School Of Environmental And Life Science Okayama University, Japan ; 4 Graduate School Of Environmental And Life Science Okayama University, Japan

; 5Toho Kasei, Japan ; 6 Toho Kasei, Japan

ABSTRACT

Reusing and recycling of useful elements in waste silicon solar cells are important for environmental issues of photovoltaic industry. In this study, we studied two kinds of chemical etching process with and without dissolving the elements. Method-I were carried out with mixed HF/HNO3 solution to leave the maximum silicon wafer from the piece. The etching rates of the electrodes and anti-reflection film (AR) in the solar cells were examined as a function of HF and HNO3 concentrations. The maximum silicon yield was attained to 96.2 % after etching the other elements. On the other hand, in Method-II, the crushed wire and electrode containing silicon, copper, aluminum, silver, tin etc. were the etching target. The mixed solution of HCl/HNO3 was used. As copper etching rate were slowest in the above etched materials, the effect of HNO3 and HCl concentration on copper etching was studied and the mathematical model was developed. The maximum silicon yield became 98.0% after etching.

KEYWORDS - solar cell; silicon; recycling; etching



A NOVEL FORMULATION USED FOR INDOOR AREA LOCATION ESTIMATION BASED ON WIFI RSSI LEVELS

MUHAMMED FAHRI UNLERSEN1 , KADIR SABANCI2 , MUCIZ OZCAN3

1 Necmettin Erbakan University, Turkey ; 2 Karamanoglu Mehmetbey University, Turkey ; 3 Necmettin Erbakan University, Turkey

ABSTRACT

In our daily life, navigation devices have become indispensable. Especially outdoors, thanks to the GPS sensors, a stranger of a city could beseen as an experienced driver. Outdoor localization problem can be solved very accurately thanks to the inclusion of GPS sensors into the mobile devices. But indoor localization is still a problem because the satellite signals cannot penetrate into the buildings. To overcome this problem a WLAN fingerprint based position estimation system has been proposed. A database which was created by collecting various wireless access point (WAP) received signal strength (RSSI) in one of building of Universitat Jaume, has been investigated. There are 1356 records that collected from 85 WAP in the database. Each record has been collected various points of the building but same floor. A novel formulation has been proposed to estimate latitude and longitude. The estimation errors of new formulation has been presented.

KEYWORDS - Indoor location estimation,WiFi, RSII



AN APPLICATION FRAMEWORK FOR SHOP SCHEDULING OPTIMIZATION PROBLEMS

OGUZHAN KOCATEPE1

1 Beykoz University, Turkey

ABSTRACT

Sequencing is the process of defining the order in which the set of jobs to be completed are done. Scheduling is the process of adding start and finish information to the job order dictated by the sequence. There are 5 basic types of scheduling: Single Machine Scheduling Parallel Machine Scheduling Open Shop Scheduling Flow Shop Scheduling Job Shop Scheduling This paper is about the implementation of the algorithms for these basic scheduling and types and developing heuristic approaches to more complex scheduling problems. A heuristic is an algorithm which tries but not guarantees to find the optimum solution but gives good results for NP-hard or NP-complete problems. NP means non-deterministically polynomial time

KEYWORDS - Machine Scheduling, Shop Scheduling Problems, Optimization Algorithms, Heuristics,Np-Complete Problems



A 3 WAY PLANAR MICROWAVE POWER DIVIDER COMBINER FOR HIGH POWER APPLICATIONS

MARCIN GORALCZYK1

1 Institute Of Radioelectronics And Multimedia Technology At Warsaw University Of Technology, Poland

ABSTRACT

The biggest advantage of the branch-line power divider is its ability to maintain high input return losses despite the matching of the succeeding sections (amplifiers) as long as those sections are both the same. The problem arises when there is a need for splitting the power into three such devices. A lot of work has been done in this field. Generally, the solutions assures high isolation between the paths, but they are either hard to realize (especially in planar manner) or do not improve input return losses. An arrangement of three-way planar microstrip power divider and combiner, suitable for high power applications, is presented in this paper. The structure is based on branch directional coupler and is calculated to deliver equal power to all three paths. It assures high isolation and, unlike e.g. Wilkinson power divider, improves input return losses. The structure is also easy to implement and do not require additional phase compensating lines or meanders, which are common in other solutions.

KEYWORDS - microwave,power divider,power combiner,three-way


International Conference on AdvancedTechnology & Sciences

ICATScience@

LATVIA SECTION

Technology & Sciences ABSTRACT BOOKKemal Tutuncu, Selcuk Universitesi – Turkey . Alla Anohina Naumeca, Riga Technical University – Latvia ... International Conference on Advanced

Documents