Determining heat source parameters based on particle swarm optimization for temperature field simulation of EH40/316L high power laser welding

The welding transient temperature field has an important influence on the simulation results of welding residual stress and deformation. To improve the accuracy of the temperature field simulation in EH40/316L high power laser welding, an intelligent calculation method is developed to optimize the empirical parameters of heat source model. This method uses MATLAB software to randomly generate the empirical parameters, calls ANSYS software to execute the APDL command of the welding temperature field simulation, and returns the result data when the calculation is completed. Then, based on the swarm intelligence and evolutionary intelligence of particle swarm optimization algorithm, new empirical parameters of the heat source are generated according to the prediction results. The iterative calculation is carried out until the prediction results are the optimal results. 10 cases of empirical parameter optimization are performed, and the parameter sensitivity of heat source model is analyzed according to the optimization results. The results show that the average calculation time of the 10 cases is 30.3 hours, and the maximum, minimum and average prediction errors are 2.84%, 2.06% and 2.16% respectively. This indicates a great improvement in the simulation accuracy of welding temperature field. Meanwhile, it can be seen from the response surface of the prediction error and the empirical parameters, that the mathematical function between them is a multi-valley function and has a complex nonlinear relationship.