Authors: Pankaj Kumar Shrivastava; Avanish Kumar Dubey
Addresses: Mechanical Engineering Department, AKS University, Satna – 485001, Madhya Pradesh, India ' Mechanical Engineering Department, Motilal Nehru National Institute of Technology, Allahabad – 211004, Uttar Pradesh, India
Abstract: It has been found that wheel wear rate (WWR) and surface finish is adversely affected in order to improve the material removal rate (MRR) in electrical discharge diamond grinding (EDDG) process. Therefore, simultaneous optimisation of above three responses is always desired. This research paper presents the modelling and multi-objective optimisation of EDDG using AI-based hybrid ANN-GA approach. The effect of wheel grit size has also been considered along with electrical parameters such as peak current, pulse-on time and pulse-off time. The significant control parameters for different responses have been found and effect of their variation has been discussed. The developed ANN models for different responses have been found reliable with negligible prediction errors. The optimisation results show considerable improvement of 97% in MRR with marginal increase in WWR and surface roughness.
Keywords: artificial neural networks; ANNs; electrical discharge diamond grinding; EDDG; genetic algorithms; grey relational analysis; GRA; high speed steel; electro-discharge diamond grinding; electrical discharge machining; electro-discharge machining; EDM; wheel wear rate; WWR; surface finish; material removal rate; MRR; modelling; multi-objective optimisation; wheel grit size; peak current; pulse-on time; pulse-off time; surface roughness; surface quality.
International Journal of Abrasive Technology, 2016 Vol.7 No.3, pp.226 - 245
Received: 07 Oct 2015
Accepted: 05 Feb 2016
Published online: 29 Jul 2016 *