Design, evaluation, and implementation of a model-predictive control approach for a force control in friction stir welding processes

Friction press joining is an innovative joining process for bonding plastics and metals without additives in an overlap configuration. A model-based approach for the design of an axial force controller for friction press joining is presented in this paper. A closed-loop control was set up on the machining center, in which the plunge depth was used as the controlling variable. In order to support the controller development, a nonparametric dynamic process model was developed via a data-based system identification. Subsequently, various control concepts were designed off-line and verified on the actual system. The most promising ones, a proportional controller, a controller created with the pole placement method, and a model predictive controller, were selected for further investigations. The three controllers were re-evaluated and compared by means of a defined input of disturbance variables and reference variables. The model predictive control (MPC) approach as well as the proportional controller were also tested for model uncertainties. For this purpose, different material combinations were joined using the different controllers. Thereby, it was shown that the MPC controller resulted in smaller standard deviations when encountering large model uncertainties. The investigations demonstrated the high potential of friction press joining of plastic components with metals. The results form the basis for future research, whereby the force can be specified as an additional input parameter instead of the plunge depth.