Optimization of stress distribution and surface deformation of Ti-6Al-4V alloy in laser surface shock peening by Finite element method and statistical analysis

Laser shock peening process is a complex phenomenon of interactions among various parameters in a nanosecond time frame with the goal of generating residual stress in the surface. In this research, the distribution of residual stress and surface deformations after the laser shock peening process has been investigated via finite element method (FEM) and statistical analysis. To simulate the process, ABAQUS explicit dynamic finite element method coupled with Johnson-Cook model to analyze the nonlinear behavior of Ti-6Al-4V alloy and the results were compared with experimental data from other articles. The design of experiment was used to investigate various process sates with the aid of simulation and analyzing the effect of process parameters on the distribution of residual stress and surface deformation. The aforementioned process parameters are laser spot size, laser spot overlap, laser power density, number of laser passes and laser pulse width. To achieve the correlation between the parameters and the process outcomes namely, the distribution of residual stress and surface deformation, linear regression was incorporated. Additionally, genetic algorithm was used to optimize the process and achieve more accurate results. Defining the compound variable for effective laser parameters and implementing the genetic algorithm resulted in linear regression with 93% accuracy for predicting the surface deformation and 94% accuracy for the prediction of the distribution of residual stress. A good correlation between the simulated model and other articles regarding shock peening in specific conditions was observed. The simulation results indicated that the value of compressive residual stress and surface deformation increase when power, pulse width, overlapping and the number of repetitions are increased. On the other hand, an increment in laser spot size has an opposite effect on the aforementioned parameters.