Appropriate tuning and robust design of a generalized predictive speed controller for drive systems with resonant loads
Nils Hoffmann, Sönke Thomsen, Friedrich W. Fuchs
Abstract
Nils Hoffmann, Sönke Thomsen, Friedrich W. Fuchs
Abstract
This paper presents an analysis of the appropriate tuning and robust control design for a model based predictive speed controller - the Generalized Predictive Controller (GPC) - used in a drive system with a resonant load. The resonance results from a finite stiffness of a long drive shaft. The achievable control quality of a model based predictive controller dependents on the appropriate choice of the control design parameters and the accuracy of the applied system model. If the predictive controller is not tuned adequately or unmodeled dynamics occur in the control system, the speed control performance must be reduced dramatically to avoid instability. To analyze these issues an overview about the influence of the GPC design parameters to the control performance is carried out. Moreover a possibility to tune the predictive control to be robust against model mismatches and control plant parameter uncertainties is presented in this paper. The theoretical approaches are verified in experiments with a 5.5 kW drive system. Finally, a study of the impact of backlash to the proposed speed control is presented.
OpenAlex reports 5 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
This paper presents an analysis of the appropriate tuning and robust control design for a model based predictive speed controller - the Generalized Predictive Controller (GPC) - used in a drive system with a resonant load. The resonance results from a finite stiffness of a long drive shaft. The achievable control quality of a model based predictive controller dependents on the appropriate choice of the control design parameters and the accuracy of the applied system model. If the predictive controller is not tuned adequately or unmodeled dynamics occur in the control system, the speed control performance must be reduced dramatically to avoid instability. To analyze these issues an overview about the influence of the GPC design parameters to the control performance is carried out. Moreover a possibility to tune the predictive control to be robust against model mismatches and control plant parameter uncertainties is presented in this paper. The theoretical approaches are verified in experiments with a 5.5 kW drive system. Finally, a study of the impact of backlash to the proposed speed control is presented.
Key concepts: Control theory (sociology), Model predictive control, Backlash, Controller (irrigation), Robust control, Control engineering, Control system, Stiffness