Optimization of process parameters in wire electrical discharge machining of TiB₂ nanocomposite ceramic

Abstract

Wire electrical discharge machining (WEDM) is a widely accepted non-traditional material removal process used to manufacture components with intricate shapes and profiles. The selection of optimum machining conditions for obtaining higher machining efficiency and increasing the accuracy of products are the most important task when the WEDM process is used for machining new advanced material such as nanocomposite ceramics. In this paper, a series of experiments has been carried out over a wide range of machining conditions. An L32 orthogonal array based on the Taguchi method for design of experiments is used to conduct the experiments. Then, by using a multilayer perceptron neural network, process modelling is performed and the most effective parameters on the process variables (i.e. material removal rate and surface roughness) are determined. Results demonstrate a very good modelling capacity of the proposed neural model. Finally, a genetic algorithm is used to optimize the process performance of WEDM. Additional experiments are performed to verify the adequacy of the optimization method. The optimization results are shown to be in good agreement with the experimental process outputs.