Performance analysis of WEDM during the machining of Inconel 690 miniature gear using RSM and ANN modeling approaches

• Published in: Reviews on advanced materials science Vol. 62; no. 1; pp. id. 106722 - 600
• Main Authors: Raj, Atul, Misra, Joy Prakash, Singh, Ravinder Pal, Singh, Gurminder, Sharma, Shubham, Eldin, Sayed M.
• Format: Journal Article
• Language: English
• Published: De Gruyter 29-03-2023
• Subjects: artificial neural network; FE-SEM; gear noise characteristics; Inconel 690; optimization; response surface methodology; surface roughness
• ISSN: 1605-8127; 1605-8127
• DOI: 10.1515/rams-2022-0288

The present work aims to carry out a feasible study of wire electro-discharge machining (WEDM) during the machining of Inconel 690 superalloy gears. Processing conditions of power-on time, power-off time, current, and spark-gap voltage are varied to evaluate the process performance in terms of material removal rate (MRR), surface roughness (SR), and wire consumption. Parametric optimization has been carried out using combined approach of response surface methodology (RSM) and artificial neural network (ANN). Results revealed that ANN predicted values are 99% in agreement with the experimental results which validates its effectiveness as compared to RSM predicted values. A viability study of noise characteristics of the processed gear is also done using a noise testing setup. Additionally, FE-SEM has been used to analyze the machined surface’s topography. Greater discharge energy brought by a longer pulse length raises the values of MRR, SR, and recast layer thickness (RLT). This study explores the capability of WEDM to produce a more precise and smooth gear profile as compared to other conventional machining methods. Additionally, RLT and microhardness of the machined surface have been critically studied to comprehend the better understanding of the process mechanism.