Artificial Neural Network Assisted Sensor Fusion Model for Predicting Surface Roughness During Hard Turning of H13 Steel with Minimal Cutting Fluid Application

Abstract

Abstract Surface roughness plays an important role in manufacturing process and is a factor of great importance in the evaluation of cutting performance. Performance parameters such as cutting force, cutting temperature, acoustic emission signals etc. can be used to predict surface roughness. It is expected that more accurate prediction would be possible if these factors are considered collectively with cutting parameters since each of these factors predict surface roughness in their own characteristic fashion. In this present work, an attempt was made to fuse cutting temperature along with cutting parameters to predict surface roughness during turning of H13 tool steel having a hardness of 43 HRC. An artificial neural network model was developed exclusively with cutting parameters then it was compared with another ANN model developed by fusing cutting temperature with cutting parameters and their ability to predict surface roughness (Ra) was analyzed. The fusion model developed based on the artificial neural network was found to be superior to ANN model without sensor fusion.