Optimization of the Process Parameters for Controlling Residual Stress and Distortion in Friction Stir Welding

Cem C. Tutum, Henrik B. Schmidt, Jesper H. Hattel (Technical University of Denmark)

In the present paper, numerical optimization of the process parameters, i.e. tool rotation speed and traverse speed, aiming minimization of the two conflicting objectives, i.e. the residual stresses and welding time, subjected to process-specific thermal constraints in friction stir welding, is investigated. The welding process is simulated in 2-dimensions with a sequentially coupled transient thermo-mechanical model using ANSYS. The numerical optimization problem is implemented in modeFRONTIER and solved using the Multi-Objective Genetic Algorithm (MOGA-II).

In conclusion, a multi-objective optimization application in the friction stir welding process has been presented. Minimization of the residual stresses and maximization of the welding speed have been considered simultaneously using the tool rotation speed and the traverse welding speed as the design variables.

In addition to the description of the process goals, process-specific thermal limitations, i.e. lower and upper bounds on the peak temperature, have been added to the optimization problem in order to take the tool loads and tool life issues into account. At the end of the optimization study, feasible and unfeasible solutions are discussed and the Pareto solutions are presented. The results show that a tool rotation speed of 200 to 400 rpm can be considered as robust working conditions for almost all possible welding speeds. Depending on the Pareto designs, ranking of the trade-off solution alternatives has been discussed in addition to looking to the optimization problem from a manufacturer point of view.