Comparative Analysis of Simulation of Different ANN Algorithms for Predicting Drill Flank Wear in the Machining of GFRP Composites

摘要

Optimum selection of machining conditions significantly results in the increase of productivity and the reduction of costs. So, the present research paper focusses on an Artificial Neural Network (ANN) based approach to optimize the HSS drill flank wear by simulating the machining parameters in the drilling of GFRP composite laminates. The present research paper is also focused on comparison of different ANN algorithms to predict the drill flank wear while machining. ANN is trained with the data collected from the experimentation. The experimental data is generated by performing drilling operation on CNC machine using different machining factors and levels. Further optimization of the ANN structure is done through performance evaluation of the selected algorithms by changing its structural parameters. This optimized ANN can measure drill flank wear under the specified work material, tool material and machining conditions efficiently.