Optimization of Surface Roughness When Turning Polyamide using ANN-IHSA Approach

This study presents an approach by coupling artificial neural network (ANN) and improved harmony search algorithm (IHSA) to determine the optimum cutting parameter settings for minimizing surface roughness when turning of polyamide material. An ANN model surface roughness was developed in terms of cutting speed, feed rate, depth of cut, and tool nose radius using the data from the turning experiment conducted according to Taguchis L27 orthogonal array. The optimal cutting parameter settings were determined by applying the IHSA to the developed ANN surface roughness model. The results show that the proposed optimization approach can be efficiently used for optimization of cutting parameter settings when turning polyamides. Although determining ANN and IHSA parameters is quite complex and problem dependent, it can be simplified by using Taguchis experimental design as in this study.