Application of NN technique for predicting the in-depth residual stresses during hard machining of AISI 52100 steel

In machining of parts, surface quality is one of the most impellent customer requirements. The most relevant issues are surface roughness and residual stresses. In particular, the latter are affected by tool geometry, material characteristics and process conditions. Residual stresses can have a significant effect on the service quality and the component life. Residual stresses can be determined by both empirical and numerical investigations for selected configurations, however, these are expensive procedures. This paper presents a hybrid model based on the artificial neural networks (ANNs) and finite element method (FEM) that can be used to predict the residual stress profile. A three layer neural network has been trained and tested on the data obtained by numerical investigations of hard machining of 52100 bearing steel. The numerical results are consistent with experimental data.