Constrained optimization of plasma enhanced turning by Electromagnetism-like Algorithm

The process of plasma-assisted machining for turning applications utilizes a high-temperature plasma arc to provide a controlled source of localized heat, which softens only that small portion of the work material removed by the cutting tool. The goal of this study is to present a methodology for determination of the optimal cutting parameters (cutting speed, feed and undeformed chip temperature) during plasma enhanced turning of hardened steel AISI 4140 to maximize the material removal rate by considering surface roughness as the constraint through coupling neural network (NN) and Electromagnetism-like Algorithm (EM). For this purpose, a model was developed to relate current of plasma to the undeformed chip temperature. To this end, experiments on hardened AISI 4140 are conducted to obtain surface roughness values based upon full factorial design of experiments, and then, analysis of variance (ANOVA) is performed. Material removal rate constitutes the main function for the Electromagnetism algorithm, and surface roughness is applied as the constraint of the EMA function. The function is optimized by the EMA code, and finally, the optimum variables are obtained, and the results of EMA are tested experimentally

Keywords:

Electromagnetism-like Algorithm , Constrained optimization , Neural network modeling , Plasma aided turning , Design of experiments