292 jagannath

Investigations on the effect of Waste fried oil methyl ester blends and load on performance and smoke opacity of diesel

engine using response surface methodology

*Authors* Jagannath HirkudeDeepa VedarthamAtul S. Padalkar

Mechanical Engineering Department Padre Conceicao of College of Engineering

Verna, Goa,India. ICAER 2013 @ IIT Bombay on 11/12/13 1

Overview of the presentation

Objectives

Introduction

Waste Fried Oil Methyl Ester preparation

Properties of WFOME

Engine Set Up

Response Surface Methodology

Results

Conclusions

ICAER 2013 @ IIT Bombay on 11/12/13 Page 1of 20

Objectives

To investigate interactive effect of Load and blend on

performance of CI engine using WFOME

Optimization of input parameters for best performance

and least smoke emissions.

ICAER 2013 @ IIT Bombay on 11/12/13 Page 2 of 20

4

Introduction

Enormous amount of waste fired oil generated from restaurants,

food processing industries and fast food shops everyday.

The high viscosity and poor volatility of WFO is major limitation.

The most convenient method to use waste fried oil as fuel in CI

engines is to convert it into biodiesel.

RSM - Collection of mathematical and statistical techniques for

empirical model building.

Objective of RSM- to optimize a responses (output variables),

influenced by several independent input variables.

ICAER 2013 @ IIT Bombay on 11/12/13Page 3 of 20

Preparation of WFOME

ICAER 2013 @ IIT Bombay on 11/12/13 Page 4 of 20

Collectionfrom different hotel of Goa, India

FiltrationRemoval of food residues and solid particles

Heating and dryingMoisture removal at 100oC for 10 min

MixingPotassium hydroxide with methanol oil + WCO at 60oC

Stirring for 15 min at 600 rpm

SeparationGlycerin is allowed to settle for 24 hours and separated from biodiesel

Preparation of WFOME

ICAER 2013 @ IIT Bombay on 11/12/13 Page 4 of 20

Properties Comparisons

Page 5 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Engine Specifications

Page 6 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Experimental Setup

12

3

78

4

1011

9

5

6

Page 7 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Response Surface Methodology (RSM)

1. Identifying the important input variables (Load & blend).

2. Development of design matrix.

3. Recording the responses (BTE, BSFC, EGT, & OP)

4. Development of second-order quadratic model.

5. Checking the adequacy of the models developed.

6. Testing the significance of the regression coefficients.

7. Presenting the interactive effects of the input parameters on the responses.

8. Determination of optimized input parameters for the multiple responses.

Page 8 of 20ICAER 2013 @ IIT Bombay on 11/12/13

RSM Analysis

Blends five levels – B0, B50, B70, B90 and B100.

Load eight levels - from 0.5 kW to 4 kW in steps of 0.5 kW

each.

Design matrix contains 40 experimental runs.

Development of quadratic models.

Significance and validation of models was analyzed Based on

the ANOVA.

Optimization and Validation of model.

Page 9 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Quadratic Model

BTE = 5.21 – 0.046 B + 14.6 L – 6.19× 10-

3×B×L+3.17×10-4×B2 – 2.2×L2 (1)

BSFC = 0.876 + 1.834×10-3×B – 0.418 L – 1.923×10-

4×B×L–4.782×10-6×B2+0.070×L2 (2)

EGT =103.89 + 0.09 B + 30.01 L – 7.63×10-

3×B×L+1.71×10-3×B2 + 3.91× L2 (3)

OP =13.9 + 0.049×B +22.2×L+5.01×10-3×B×L + 6.86× 10-4× B2 – 1.4×B2×L2 (4) Page 11 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Interactive Effect on BTE

Max. BTE = 29.97% @ B0 & 3.5 kW

Page 12 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Interactive Effect on BSFC

Min. BSFC = 0.275 kg/kWh @ B0 & 3.5 kW

Page 13 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Interactive Effect on EGT

Min. EGT = 189oC @ B 20 & 2.20 kW

Page 14 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Interactive Effect on OP

Min. OP = 58 HSU @ B30 & 2.5 kW

Page 15 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Optimization

High desirability of 0.962 was obtained at the optimum input

parameters viz. 20% of WFOME blend (B20) and 2.38 kW of

brake load, where the values of the BTE, BSEC, EGT and

smoke opacity were found to be 27.40%, 0.28, 1960C and 58

HSU respectively.

Page 16 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Validation of Results

Exp. No.

Blend(%)

Load(kW)

  BTE(%)

BSFC(kg/kWh)

EGT(0C)

OP(HSU)

 1

 20

 2.5

Predicted 26.84 0.295 205 62

Actual 26.55 0.298 207 63

% Error -1.09 1.00 0.96 1.58

Page 17 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Conclusions

The prospectuses of WFO as fuel are very attractive for

developing country like India.

More than 6.5 million diesel engines being used in the Indian

agricultural sector for various activities.

With increase in load increase in BTE, EGT and Smoke

opacity with decrease in BSFC was observed.

Page 18 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Conclusions

Reduction in WFOME blend from B100 to B20 could help to

increase BTE and reduce BSFC, EGT and smoke opacity.

Desirability approach of the response surface methodology was

found to be simple and efficient optimization technique.

Page 19 of 20ICAER 2013 @ IIT Bombay on 11/12/13

Conclusions

Optimal Input parameters – Blend of B20 and Brake Load of

2.38 kW for Maximum BTE and Minimum BSFC, EGT and

OP.

Validation of result showed that developed models were quite

accurate and in good agreement.

WFOME can be suitable alternative fuel for diesel at higher

load and lower blend.

Page 20 of 20ICAER 2013 @ IIT Bombay on 11/12/13

ICAER 2012 @ IIT Bombay on 11/12/13 22

“THANK YOU”