Affiliation:
1. The Kavery Engineering College, Department of Mechanical Engineering, India
2. Sona College of Technology (Autonomous), Department of Mechatronics Engineering, India
3. Sona College of Technology (Autonomous), Department of Mechanical Engineering, India
Abstract
Electrical discharge machines (EDM) are widely employed in machining components containing complex profiles of hard-to-cut and machining materials. However, the fabrication-of-tool time for the EDM process is excessively high in the traditional machining method, which significantly affects the machining rate. Therefore, in this paper, a powder metallurgy (PM) technique is employed to fabricate the tool electrode using copper (Cu), titanium carbide (TiC), and zirconium silicate (ZrSiO4) for different combinations. An L18 orthogonal array (OA) is planned using the following input parameters: three types of tools (Cu, Cu90, Cu80), peak current (PC) [A], pulse on time (PT) [µs], and gap voltage (GV) [V]. The performance of EDM is evaluated through the material removal rate (MRR), tool wear rate (TWR), and surface roughness (SR). The process parameters are optimized using two different techniques: the technique for order of preference by similarity to the ideal solution (TOPSIS) and grey relational analysis (GRA). TOPSIS and GRA optimization techniques produce the same optimal parametric solution for less TWR, SR, and higher MRR with the combination of the Cu90 tool, E8 APC, 15 µs pulse PT, and 75 V GV. Based on the ANOVA table of TOPSIS, pulse on time plays a major role, contributing 46.8 % of the machining performance; peak current shows the most significant contribution of 39.3 % of the machining performance using GRA values. Furthermore, the scanning electron microscope (SEM) image analyses are carried out on the machined workpiece surface to understand the effect of tools on machining quality.
Publisher
Faculty of Mechanical Engineering
Subject
Mechanical Engineering,Mechanics of Materials
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献