Title: Artificial neural network modelling and multi objective optimisation of hole drilling electro discharge micro machining of invar
Authors: Rajesh Kumar Porwal; Vinod Yadava; J. Ramkumar
Addresses: Department of Mechanical Engineering, Motilal Nehru National Institute of Technology, Allahabad-211004, India. ' Department of Mechanical Engineering, Motilal Nehru National Institute of Technology, Allahabad-211004, India. ' Department of Mechanical Engineering, Indian Institute of Technology, Kanpur-208016, India
Abstract: Hole drilling electro discharge micro machining (HD-EDMM) is one of the potential method for creation of micro-holes in difficult to machine electrically conductive workpiece materials. Maintaining quality and accuracy of the drilled micro-holes along with better performance characteristics have always been a challenge for the researchers and manufacturers. Keeping cost and time of manufacturing into consideration, modelling and optimisation of EDMM is required. In this paper, attempts have been made to model the HD-EDMM process using feed forward back propagation neural network (BPNN) and further combined with GRA-based PCA for its optimisation. The developed ANN model and finally optimised results are validated with our own experimentally obtained results. The approach used in the present paper would be extendable to other configuration of EDMM such as milling-EDMM, wire-EDMM and grinding-EDMM.
Keywords: hole drilling; electro-discharge machining; micro EDM; HD-EDMM; modelling; multi-objective optimisation; Taguchi methods; ANNs; GRA; PCA; electrical discharge machining; invar machining; artificial neural networks; grey relational analysis; principal component analysis; micro-holes; hole drilling.
International Journal of Mechatronics and Manufacturing Systems, 2012 Vol.5 No.5/6, pp.470 - 494
Received: 21 Jul 2011
Accepted: 10 Apr 2012
Published online: 21 Aug 2014 *