Model predictive force control for multi-axis rough milling

• Applicants: Professor Dr.-Ing. Thomas Bergs; Dr.-Ing. Sebastian Stemmler
• Subject Area: Metal-Cutting and Abrasive Manufacturing Engineering ; Automation, Mechatronics, Control Systems, Intelligent Technical Systems, Robotics
• Term: since 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 450794086

Milling is a flexible and highly dynamic machining process for production of simple geometries up to complex free-form surfaces. The tool performs a rotary cutting movement on a clamped workpiece. Superimposed with a translatory feed movement, a cyclically interrupted cutting process occurs, which is characterized by a continuously varying chip thickness. The result is a pulsating load on the workpiece, tool and machine. In the first phase of the project, a model predictive controller (MPC) for controlling the active force in a multi-axis milling process without simultaneous table movement was investigated. An MPC for two-axis milling was used as preliminary work. The force is measured using a table dynamometer. When the machine table and, thus, the dynamometer are deflected in the multi-axis milling process, the measurement of the process forces is disturbed, and the measured forces rotate in relation to the simulated engagement conditions. A measurement model was formulated to correct these effects. In addition, a method for online estimation of the parameters of a force model was developed. The existing two-layer MPC, consisting of a trajectory generator of the optimum feed rate and an MPC for setting the speed, was further developed for the more complex engagement. In addition, further control approaches were developed which take the active force directly into account in the MPC itself. The control system was validated by simulation and experiment. In the project phase that has now been applied for, the developments from the first project phase are to be continued in order to apply the researched force control to a multi-axis simultaneous milling process. For this purpose, the complex intervention conditions in the control system must be taken into account and dynamic measurement disturbances must be compensated for with the aid of the measurement model. The research results to date indicate that the considerable computational effort required for force control is the decisive obstacle to using the control system in an industrial production environment. Therefore, methods to reduce the computing time of the various force control components will be investigated. Finally, the force control system is to be validated under production-related conditions.

• DFG Programme: Research Grants
• Co-Investigator: Professorin Dr.-Ing. Heike Vallery