Optimization of parameters for the shot peening effect On Welded specimen using Artificial Neural Network

Abstract

The welding process induces Residual tensile stress that is detrimental to Fatigue life. Tensile stress act to stretch or pull apart the surface of the material. With enough loads cycle at a high enough tensile stress, a metal surface initiate a crack. Significant improvement in Fatigue life can be obtained by modifying the Residual stress level in the material. The intent of this paper is for Butt welded similar and dissimilar joint, evaluate Residual stress level by using most accurate and best developed X-Ray diffraction method, before and after surface modification. In this paper, welding compressive residual stress can be modified by Shot peening, which required simple equipment and treatment is extensively employed as a method to improve Fatigue strength. This study uses MATLAB software to analyze two experimental data sets using a neural network fitting tool. According to the results, artificial neural networks (ANNs) are accurate and results show that ANN models can be used successfully for predicting surface roughness changes during shot Peening Process.