PREDICTION MODEL AND OPTIMIZATION OF MACHINING PARAMETERS USING INTEGRATED ANN-GA METHOD ON CNC MILLING MACHINE

Abstract

The surface roughness is one of the important indicators widely used to evaluate surface quality in mechanical processing. This paper introduces a predictive model and optimizes machining parameters by integrating the artificial neural network (ANN) model and genetic algorithm (GA) when machining on CNC milling machines. To evaluate the capability of the ANN-GA method for prediction and optimization of surface roughness, a real experiment on machining C45 steel with a high-speed steel tool was performed on an AGMA - A8 CNC milling machine. The results show that the 3-8-1 network structure of ANN proposed model has the correlation coefficient (R) values ​​greater than 0.9, indicating that it can predict the surface roughness accurately and reliably. In addition, GA is integrated with neural network model to determine the optimal machining parameters leading to minimum surface roughness. The results of this study demonstrate that the ANN-GA method is capable of predicting and optimizing the optimum machining parameters.