Modelling and prediction of machining parameters in composite manufacturing using artificial neural network

Abstract

This paper presents a consistent way to deal with optimize Wire EDM response parameters for the aluminum metal matrix composites utilizing Artificial Neural Network strategies. Wire cut discharge machining is a progressed machining strategy controlled by an outsized scope of assortment of interrelated complex parameters like discharge current, pulse on time, pulse off time and servo speed rate. Any slight variations in one will have an exertion on the machining execution measures like surface roughness and material removal rate. In the present work aluminum 7075 is utilized as matrix and activated carbon as reinforcement metal matrix composites. 27 trails of investigations considering response surface procedure are done and the perceptions are made. ANN has been produced based on back propagation for expectation of the numerous reactions. The ANN was consequently prepared with test information. Testing of the ANN is done utilizing exploratory information not utilized amid preparing. The outcomes demonstrate that the results of the ANN are in great concurrence with the trial information; this shows the created neural system can be utilized as an option path for figuring response parameters for given process parameters.