Title: Modelling and analysis for the machinability evaluation of Udimet-L605 in wire-cut electric discharge machining
Authors: Somvir Singh Nain; Dixit Garg; Sanjeev Kumar
Addresses: Department of Mechanical Engineering, CMR College of Engineering & Technology, Kandlakoya, Hyderabad, Telangana, India ' Mechanical Engineering Department, National Institute of Technology, Kurukshetra, Haryana 136119, India ' Mechanical Engineering Department, PEC, University of Technology, Chandigarh, 160012, India
Abstract: The intention of this research is to examine and analyse the machinability of Udimet-L605 in wire-cut electric discharge machining. The pulse-on time, pulse-off time, peak current, wire tension, spark gap voltage, wire feed have been considered as input variables and two response variables as material removal rate and surface roughness. The four models, such as support vector machine using two different kernels, nonlinear regression and multi-linear regression have been proposed to check the variation among the experimental and predicted consequences. In addition to this, the sensitivity analyses test has been performed on prominent model to examine the influence of input variables on the material removal rate and surface roughness in wire electric discharge machining and epitomised that pulse-on time is the momentous parameter for both material removal rate and surface roughness. The support vector regression based on the polynomial kernel model has conferred enhanced resulting in a contest with other models.
Keywords: support vector machine; SVM; regression; sensitivity analysis; material removal rate; surface roughness; Udimet-L605.
International Journal of Process Management and Benchmarking, 2019 Vol.9 No.1, pp.47 - 72
Received: 05 Jul 2016
Accepted: 12 Nov 2016
Published online: 12 Feb 2019 *