Performance evolution in machining parameter of Al-Si (LM6) alloy using neural network

Abstract

The global manufacturing age has increased competition in the manufacturing sector as markets become more agile and customer-focused. Because of the fierce rivalry, manufacturers are paying more attention to automation. In the last several decades, computer numerically controlled (CNC) machine tools have been utilized to achieve complete automation in machining. The artificial neural network theories are used in this research to train a solution for the problem of selecting machine setup settings for a turning operation. A collection of input and output values can be mapped out by an artificial neural network. A network may be used to anticipate output values for a certain set of input values once it has been trained. Process inputs like cutting speed, feed, depth of cut, and coolant flow rate can be generated by a trained program, as can their related process outputs like surface finish. Forward mapping of process inputs and outputs is accomplished using back propagation neural networks in this approach. The optimum machine setup parameter may then be selected interactively using these networks. To solve the model, a MATLAB program has been created. The experiment's findings are verified using a neural network.