A novel approach of GEF and GA for the optimization of multi-objective wire EDM process during the machining of DC53 super alloy

Abstract

In components manufacturing, the Wire EDM is more popular due to its outstanding features of high dimensional accuracy, lower cost of production and good surface finish as there is no physical contact between the wire and work piece. However, for the steel with higher hardness, it is difficult to obtain these features up to the required extent. Moreover, with the help of optimum process parameters selection the WEDM performance characteristics should be improved. The most commonly studied responses for this process are material removal rate (MRR), Surface finish, Kerf width, wire consumption, roundness error. Among these, the responses like Kerf width, MRR and surface roughness are the primary responses. It is observed that pulse on time, pulse off time, servo voltage, peak current and wire feed are the influencing input parameters to these primary responses. Therefore, in the present work, the optimal level of input parameters is estimated for these responses using the Grey-Entropy-Fuzzy (GEF) and Genetic Algorithm (GA) during wire EDM of steel grade DC53 with high hardness of HRC58normally used in stamping dies, injection moulding and compression Moulding etc. To convert the multi objective problem into a single objective, the grey relational coefficient (GRC) has been calculated using Grey Relational Analysis and the weight-age of each response is approximated during the entropy method. To estimate the relation among the input parameters and single objective (GRC) fuzzy mathematical modelling technique termed as GEF has been applied. The optimal performance has been calculated using GA and GEF model is considered as fitness function. Five conformational tests on optimum parameter combination suggested by GA has been performed. The predicted values and experimental values have been found to be in good agreement with a standard error of 3.31% hence the prediction performance of the GREG-Fuzzy is quite satisfactory.

Energy Consumption Optimization Strategy of Hybrid Vehicle Based on NSGA-II Genetic Algorithm

“…Xue B. applied the ant colony algorithm and the GA to wireless power transmission. For the multi-objective optimization of energy transmission and information transmission, this method could optimize both objectives at the same time and had a high efficiency and information transmission capacity [11]. The vehicle battery system has a strong correlation with EMS.…”

A novel hybrid approach GREG-fuzzy-GA for minimizing work piece temperature during 2.5D milling of Inconel625 super alloy

“…The researchers used multiple ways to give weights to the replies, including the weighted sum method (Marler and Arora, 2010), expert scoring method (Zhang et al , 2014), entropy method (Zhi-hong, 2002) and scalarization method (Gunantara, 2018). Before the desirability approach, a layer of grey relationship grade was included by a few studies (Gopal and Soorya Prakash, 2018; Kumar et al , 2021; Gok, 2015).…”

“…As a result, for such AI mathematical models, the maximum number of experiments is preferred. Similarly, Kumar et al (2021) used an innovative and hybrid AI technique based on learning algorithm to optimize WEDM process parameters. Researchers use these techniques all the time, and they are even used in real-world production scenarios.…”

Micro-EDMing of SKD-5 die steel stimulated with magnetic field and ultrasonic vibration

“…Several sparks take place between the electrodes, which are submerged in the dielectric medium. 2 The erosion of the material takes place by the vaporization and melting action produced from the electric arc. [3][4][5][6][7] The powder form of chips is swilled out by the stream of the dielectric.…”

“…μ-EDM can be used to drill holes in materials having higher thermal resistance and strength. 2 However, the geometrical accuracy of the job is significantly affected by the wear of the electrode in the μ-EDM process. 12 Further, the deterioration of the hole is another challenge in the case of hard-to-machine materials.…”