Determining alloyed layer thickness in electric discharge alloying by inverse estimation of energy distribution

Abstract

Electric discharge alloying (EDA) is an important surface modification process, however, accurate determination of alloyed layer thickness is essential for effective utilization of this process in the industry. The spark energy distribution among the tool and workpiece controls alloyed layer thickness. In this work, a novel method is developed for accurate and quick determination of alloyed layer thickness through inverse estimation of energy distribution factor. A nonlinear transient finite element method based direct numerical model is developed for the analysis of EDA. The inverse estimation of the energy distribution factor is carried out by reducing the difference between actual and direct model computed data. For this purpose, an integrated finite element method–artificial neural network based method is developed. Physical experiments at the shop floor were also performed, which suggested that the inversely obtained energy distribution factor is sufficiently accurate in estimating the alloyed layer thickness at changed operating conditions.