Experimental investigation and optimisation of WEDM process for machining maraging steel using neural network based Jaya algorithm. (2018)

• Record Type: Journal Article
• Authors: Sen, Ruma; Choudhuri, Bikash; Barma, John Deb; Chakraborti, Prasun
• Is Part Of: International journal of machining and machinability of materials. Volume 20:Number 4(2018)
• Page Start: 387
• Page End: 399
• Publication Date: 2018
• Subjects: wire electric discharge machining -- WEDM -- maraging steel -- BPNN -- Jaya; Machining -- Periodicals; Metals -- Machinability -- Periodicals; 671.35
• Languages: English
• ISSNs: 1748-5711

Abstract

Wire electric discharge machining has become one of the popular machining processes used for generating complex geometry on electrically conductive materials. Due to its complex behaviour, a correlation between parameters and machining characteristics has been established by a neural network model. This study also recommends an optimal setting of process parameters with an aim to improve machining performance, which is achieved by using Jaya algorithm. Experiments were conducted on maraging steel 300 using silver coated brass wire to study the effects of process parameters (pulse on time, pulse off time, peak current, and servo voltage and wire tension) on the performance characteristics such as root mean square roughness, cutting speed, and kerf width. From the study, it is revealed that pulse on time is the predominant factor that mostly influences the machining characteristics. According to the analysis of results, the most suitable parametric combinations which obtained from Jaya algorithm offer the best performance characteristics.