Model predictive controller for machine tool feed drives

One of the most important requirements of machine tools is an accurate and fast positioning behavior of the installed feed drives. The closed-loop position controllers have a central role in fulfilling these requirements. At present as closed-loop position controllers, mainly P-controllers are used, whereby model predictive controllers (MPC) are largely still unnoticed. Therefor a model predictive controller as closed-loop position controller for feed drives in machine tools will be presented in this article. A focus is an increase of the positioning speed and the robustness concerning unmodeled machine behaviors. In addition, a method based on a genetic algorithm is presented to design the controller parameters according to requirements to the positioning behavior.