Modeling of Electrical Discharge Coating Parameters Using Artificial Neural Network

Electrical discharge coating (EDC) is modified form of electrical discharge machining (EDM) to modify the surface of ZE41A magnesium alloy with WC/Cu coating to enhance the tribological properties. Variables such as compaction load, current and pulse on time were varied during EDC resulted varying layer thickness. A central composite design (CCD) of response surface methodology (RSM) has been employed to perform the experiment and collect data for analysis. In this study, an artificial neural network (ANN) model was developed for the analysis and prediction of EDC parameters while coating ZE41A magnesium alloy. The process variables and the responses metal transfer rate (MTR) and surface roughness (Ra) were used as input data set to train the three-layered feed-forward, back-propagation artificial neural networks. The networks were trained to predict MTR and surface roughness separately. The results show that the correlation coefficients between the neural network predictions and experimental values of MTR and surface roughness was 0.9991, recommending the reliability of the neural network model for analysis and optimization of EDC process. Microstructure of the machined surface has been characterized using scanning electron microscopic (SEM).