Fuzzy Logic Modeling and Multiple Performance Optimization in Turning GFRP Composites Using Desirability Function Analysis

Abstract

Glass fibre-reinforced polymer (GFRP) composites are extensively used now-a-days in manufacturing various components in aerospace, oil, gas and process industries. It replaces conventional materials due to their excellent properties such as light weight, corrosive resistance and superior properties. Development of predictive modeling and optimization of machining process in producing components is important for machining industries. In this work, fuzzy logic based multi response predictive model development and multi objective optimization of processes parameters using Desirability Function Analysis (DFA) in turning GFRP composite has been attempted. The input variables are cutting speed (v), feed rate (f) and depth of cut (d). The responses are surface roughness (Ra), metal removal rate (MRR) and tool wear (VB). The average percentage error in fuzzy logic prediction obtained as 2.74%, 12.67% and 3.06% for Ra, MMR and VB respectively. The optimum level of input parameters for composite desirability was found v2 f1 d3. The corresponding optimum parameter is v=100 m/min, f=0.10mm/rev and d= 1.5mm for obtaining combine optimization of Ra, MRR and VB having equal weightage. The analysis of variance of composite desirability at 95% confidence level showed that depth of cut is the most significant parameter with 39.38% contribution followed by feed and cutting speed.