INVESTIGATION AND MODELLING OF EDM PROCESS PARAMETERS IN MACHINING OF INCOLOY-800

Harmanpreet, Harvinder Lal, Gaurav Dhuria

Abstract: The experimentation has been performed on EDM for the modelling and optimization of the process parameters. In this work modelling of the processes are done by artificial neural networking and optimization by genetic algorithm. For fitness function regression analysis has been performed. Design of experiment with mixed level L18 orthogonal array has been considered. All 18 experiments were performed on INCOLOY-800. Copper as an electrode considered. In this work TWR considered as the response variable. Dielectric fluid with two level; where current, pulse on time and gap voltage considered with three levels as the process parameters. 4-17-1-1 network considered in ANN by feed forward back propagation for process modelling. Two types of dielectric fluid considered as kerosene and EDM oil. For multi objective-optimization Genetic Algorithm used (GA).

Keywords: Artificial neural network, genetic algorithm, machining time, material removal rate

DOI: https://doi.org/10.15623/ijret.2015.0408021