Multi-Objective Optimization of PMEDM Process Parameter by Topsis Method

Copyright © 2019 by author(s) and International Journal of Trend in Scientific Research and Development Journal. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0) (http://creativecommons.org/licenses /by/4.0) ABSTRACT In this study, MRR, SR, and HV in powder mixed electrical discharge machining (PMEDM) were multi − criteria decision making (MCDM) by TOPSIS method. The process parameters used included work piece materials, electrode materials, electrode polarity, pulse−on time, pulse−off time, current, and titanium powder concentration. Some interaction pairs among the process parameters were also used to evaluate. The results showed that optimal process parameters, including ton = 20 μs, I= 6 A, tof = 57 μs, and 10 g/l. The optimum characteristics were MRR = 38.79 mm3/min, SR = 2.71 μm, and HV = 771.0 HV.


INTRODUCTION
Conductive powder is often mixed into dielectric fluid in electrical discharge machining (EDM) because it is an effective solution for improving productivity and the machined surface quality after machining [1]. Many types of powder materials have been used, such as Al, Si, SiC, etc. [2]. They are mixed into dielectric fluid to improve the material removal rate (MRR), surface roughness (SR), and electrode wear ratio (EWR) in EDM. Recently, the Taguchi method has been combined with several other methods, such as grey relational analysis (GRA), TOPSIS, particle swarm optimization (PSO), and fuzzy logic [3]. This has contributed to improving the efficiency of the optimization problem in PMEDM.
Recent research has shown that Taguchi combined with several other methods, such as GRA, TOPSIS, and PSO, can MCDM in EDM, and results have been good.
Taguchi-GRA has been used to simultaneously optimize MRR, EWR, and OC expenditures in micro-EDM of CP Ti [4]. SR and kerf width have been optimized simultaneously in WEDM using Taguchi-GRA [5]. Taguchi-GRA was used to simultaneously optimize MRR, SR, recast layer thickness (RLT) and micro hardness (HV) in PMEDM of H11 die steel [6]. Many quality indicators have been optimized by Taguchi-GRA in dry EDM using a Cu electrode of AISI D2 steel [7]. In addition, the surface topography of H11 steel was significantly improved. The TOPSIS method has been used to MCDM in both traditional machining (milling, turning, drilling, grinding), non-traditional machining (EDM, abrasive jet machining, micromachining) and many other areas [8]. TOPSIS algorithms can simultaneously optimize a large number of quality characteristics, and its optimal results are better than other methods, such as Taguchi and GRA.
This study presents the results of simultaneous optimization of the MRR, SR, and HV in PMEDM. The Taguchi-TOPSIS method, seven process parameters, and three kinds of interactions between them were studied.

Experimental setup and methods
In this study, an electrical discharge machine, the AG40L (Sodick, Inc. USA), was used to perform the experiment. Ti powder was mixed into the dielectric fluid (oil HD-1) during the experimental process. Work piece dimensions were 45×27×10 mm. The electrode materials were Cu and Gr. Seven factors were considered as shown in Table 1. Experimental results are shown in Table 2.
The weight different of work pieces before and after the performance trial were measured by an electronic scale, AJ 203 (Shinko Denshi Co. LTD − Japan). Its accuracy was ±0.001 g. The SRwas measured by a strain gauge transducer contact, SJ − 301 (Mitutoyo − Japan). The surface hardness (HV) was measured by 1106 Met Indenta (Buehler Motor, USA). The surface morphology was verified by scanning electron microscope (SEM) JEOL 6490 (Jeol -Japan).

Results and discussion
Step1-The decision matrix: The indicators selected for optimization in PMEDM, the assigned quality characteristics, are as follows: xMRR with MRR, xSR with SR, and xHV with HV.
Step 2-The normalized decision matrix: The normalized values are showed in Table 3   Table 3. Normalized data Step 3-The weighted normalized decision matrix: WMRR = 0.2 for MRR, WSR = 0.4 for SR, WHV = 0.4 for HV. The weighted decision-making matrix is shown in Table 7.
Step 4-The positive ideal solutions (PIS) and negative ideal solutions (NIS): S hown in Table 4. Step 5-The separation measures: Shown in Table 7.
Step 6-The relative closeness to the ideal solution: The relative closeness index is calculated using Eq. 11, and shown in Table 7.
Step 7-Ranking: The results clearly show that the 17 th run is getting the first rank and good performance of the alternative Ai (Table 5).  From the above tables, it is clear that the 17 th running receives the 1 st rank. Hence, the corresponding input parameters such as ton = 20 µs, I= 6 A, tof = 57 µs, and 10 g/l were found to be the optimum combination.

Conclusion
In this study, MRR, SR, and HV in PMEDM using Ti powder have been MCDM by the TOPSIS method. The results of multicriteria optimization in PMEDM using powder Ti show that: Optimal results using the TOPSIS method show that the 17 th experiment was the best. However, values of the S/N ratio show that the optimal combination is ton =5 µs, I = 4A, tof = 57 µs, and 10 g/l. Optimal values are MRR = 38.79 mm 3 /min, SR = 2.71 µm, and HV = 771 HV.