Process Parameters Selection for Obtaining of Certain Residual Stresses Distribution on Cylindrical Aluminum

Surface compressive residual stresses have significant effects on the fatigue behavior of components. In this paper a procedure for process parameters selection for obtaining of certain residual stresses distribution on cylindrical aluminum were proposed. The surface cold rolling process was simulated by to study effects of cold rolling process parameters on surface compressive residual stresses. By developing a parametric FEM method the effects of process parameters like roller and specimen diameter and interface magnitude between them were considered. By using a neural network for achieving every distribution and magnitude of residual stress on the surface, process parameter will be selected. Some experimental residual stresses by HDSG method were done and results of FEM analysis and neural network prediction were compared to experimental data and good agreement were found.