Surface Finish Analysis of D2 Steel In WEDM Using ANN & Regression Modelling with Influence of Fractional Factorial Design of Experiment
|International Journal of Engineering Trends and Technology (IJETT)||
|© 2015 by IJETT Journal|
|Year of Publication : 2015|
|Authors : U.K.Vates, N.K. Singh, Mr. B.N.Tripathi
U.K.Vates, N.K. Singh, Mr. B.N.Tripathi "Surface Finish Analysis of D2 Steel In WEDM Using ANN & Regression Modelling with Influence of Fractional Factorial Design of Experiment", International Journal of Engineering Trends and Technology (IJETT), V19(3), 159-167 Jan 2015. ISSN:2231-5381. www.ijettjournal.org. published by seventh sense research group
Wire electrical discharge machining (WEDM) is an important metal removal process in precision manufacturing of mould and dies, which comes under non traditional machining processes. It is also quite difficult to find the correct input parametric combinations to give lowest possible values of surface roughness of D2 steel under WEDM.. Non-conventional WEDM process under low temperature dielectric (DI water) is more robust and powerful approach than conventional machining process to obtaining better surface finish in low temperature treated tool steels. Low temperature dielectric cooling medium implementation generally used as secondary treatment to enhance the surface smoothness. Present work aimed to effect of WEDM parameters on surface finish of low temperature treated AISI D2 tool steel is investigated. Montgomery fractional factorial design of experiment, L16 orthogonal array was selected for conducting the experiments. The surface roughness and its corresponding material removal rate (MRR) were considered as responses for improving surface finish. The Analysis of variance (ANOVA) was done to find the optimum machining parametric combination for better surface finish. The experimental result shows that the model suggested by the Montgomery’s method is suitable for improving the surface finish. Regression (RA) analysis method and Artificial Neural Network (ANN) were used to formulate the mathematical models. Based on optimal parametric combination, experiments were conducted to confirm the effectiveness of the proposed ANN model
(1] Boothroyd, G.; Winston, A.K., (1989). Non-conventional machining processes, in Fundamentals of Machining and Machine Tools, Marcel Dekker, Inc, New York, 491.
 Erzurumlu, O.H., 2007. Comparison of response surface model with neural network in determining the surface quality of moulded parts. Materials and Design, vol. 28, no. 2, pp. 459–465.
. Gatto A., L. Luliano., 1997. Cutting mechanisms and surface features of WEDM metal matrix composite, Journal of Material Processing Technology 65. (209-214).
. Gencay Ramjan, Qi Min. Pricing and bedging., 2001. Derivative securities with neural networks, Bayesian regularization, early stopping and bagging. IEEE Trans Neural Networks 12(4): 726-34.
 Ho K.H, Newman, S.T., Rahimifard, S., Allen, R.D., (2004). State of the art wire electrical discharge machining (EDM), Int J Mach Tools Manuf. 44:1247 – 1259.
(6) Das D, Dutta A.K., Ray K.K., Influence of varied cryo treatment on wear behavior of AISI D2 Tool steel,Wear, vol. 266, pp. 297- 309, 2009.
(7) Das D, Ray K.K. Dutta A.K., Influence of sub- zero treatment on the wear behavior of die steel, Wear,vol. 267, pp. 1361- 1370, 2009.
(8) Esme U., Sabas A.and Kahraman F., Prediction of surface roughness in wire electrical discharge machiningusing design of experiments and neural network, Iranian Journal of science and technology, Transaction B.Engineering, vol. 33, pp. 231-240, 2009.
(9) RossP.J.,Taguchi techniques for quality engineering, McGraw-Hill Book Company, New York, 1996.
Montgomery method, WEDM, ANOVA, ANN, RA and surface finish.