Application of Taguchi, ANFIS and grey relational analysis for studying, modeling and optimization of wire EDM process while using gaseous media

Experimental studies on ultrasonically assisted friction stir spot welding of AA6061

“…The step by step implementation of grey relational analysis along with details and equations are found in Ref. [18]. Various stages for implementation are:…”

Surface Integrity Analysis and Simultaneous Optimization Using Taguchi Grey Relational Analysis

“…The process parameters and their levels are presented in Table 2. Specimens of size 13x12x6 mm 3 were machined by varying the cutting parameters. Brass wire is used as wire electrode with demineralised water as dielectric fluid.…”

“…In WEDM, the moving wire is a negative electrode whereas the work piece is a positive electrode. The sparks will generate between two closely spaced electrodes under the influence of dielectric fluid [3]. These sparks generate craters, micro cracks and recast layer on the machined surface [4].…”

Simultaneous optimization of WEDM responses using grey relational analysis coupled with genetic algorithm while machining EN 353

“…Huang and Liao [11] carried out MO optimization with conflicting objectives to yield a set of solutions known as trade-off, nondominated, noninferior, or pareto-optimal solutions. Shajan and Shanmugam [12], Chiang and Chang [13], Ramakrishnan and Karunamoorthy [14], Mahapatra and Amar [15], Prasad and Gopala [16], Saurav and Siba [17], Muthu et al [18], Kamal et al [19], Susanta and Shankar [20], Kamal et al [21], Balasubramanian and Ganapathy [22], Somashekar et al [23], Kapil and Sanjay [24], Nixon and Ravindra [25], Kamal et al [26], Neeraj et al [27], Bagherian et al [28], and Rao and Krishna [29] also reported on MO optimization. An easy way of solving MO optimization problem is converting MO optimization problem into SO optimization by multiplying weights to individual objectives or using grey entropy method.…”

“…Hajela and Lin [32] proposed the WBGA for multi-objective optimization. Each solution x i in the population uses a particular weight Konda et al [5] Taguchi (L 9 )→S/N ratio→regression→nondominated point Single MS, SR Puri and Bhattacharyya [6] Taguchi (L 27 )→ANOVA→F test Single MRR, SR, DD Huang and Liao [11] Taguchi (L 18 )→grey→ANOVA→F test Multi MRR, SR, GW Shajan and Shanmugam [12], Taguchi (L 18 )→regression→NSGA Multi CV, SR Chiang and Chang, [13] Taguchi (L 18 )→grey Multi MRR, SR Manna and Bhattacharyya [7] Taguchi (L 18 )→ANOVA→Gauss elimination model Single MRR, SR, SG, GC Ramakrishnan and karunamoorthy [14] Taguchi (L 16 )→S/N ratio→Weighting (MRSN)→ANOVA Multi MRR, SR, WWR Mahapatra and Amar [15] Taguchi (L 27 )→S/N ratio→regression→GA→weighting factor Multi MRR, SR Prasad and Gopala [16] Factorial CCD (32)→ANOVA→regression→NSGA Multi MRR, SR Saurav and Siba [17] Taguchi (L 27 )→grey→ANOVA Multi MRR, SR, kerf Muthu et al [18] Taguchi (L 9 )→grey→ANOVA Multi MRR, SR, kerf Kamal et al [19] Taguchi (L 18 )→grey→ANOVA Multi MS, SR, DD Susanta and Shankar [20], Taguchi (L 18 )→S/N ratio→Grey→MRSN→WSN→ANOVA Multi MRR, SR, kerf Kamal et al [21], Taguchi (L 18 )→grey→ANOVA Multi MRR, SR Balasubramanian and Ganapathy [22] Taguchi (L 8 )→grey→ANOVA Multi MRR, SR Somashekhar et al [23] Taguchi (L 9 )→grey→ANOVA Multi MRR, SR, overcut Kapil and Sanjay [24] Taguchi (L 27 )→S/N ratio→regression→NSGA-II Multi MRR, SR Nixon and Ravindra [25] Taguchi (L 16 )→ANOVA→regression→GA Multi MRR, SR, DE Kamal et al [26] Taguchi (L 18 )→grey+entropy→ANOVA Multi MRR, SR, AE, ROC Neeraj et al [27] RSM (32)→ANOVA→regression→desirability Multi CS, DD Zhang et al [8] RSM (32)→ANOVA→regression→BPNN-GA Single MRR, SR Bagherian et al [28] Taguchi (L 27 )→ANFIS→GRA Multi CV, SR Rao and Krishna [29] Taguchi (L 27 )→ANOVA→regression→NSGA-II Multi MRR, WWR MS machining speed, DD dimensional deviation, GW gap width, CV cutting velocity, SG spark gap, GC gap current, AE angular error, ROC radial overcut, CS cutting speed, BPNN-GA backpropagation neural network combining with genetic algorithm, ANFIS adaptive neuro-fuzzy inference system vector w i ={w 1 , w 2 ,….w k } in the calculation of summing objective function. The weight vector w i is embedded within the chromosome of each solution.…”