Repetitive control mechanism of disturbance cancellation using a hybrid regression and genetic algorithm

• Published in: Mechanical systems and signal processing Vol. 62-63; pp. 356 - 365
• Main Authors: Lin, Jeng-Wen, Shen, Pu Fun, Wen, Hao-Ping
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01-10-2015
• Subjects: Basis functions; Disturbances; Errors; Genetic algorithm; Genetic algorithms; Mechanical control systems; Periodic disturbance; Regression; Repetitive control; Tracking error; Tracking errors; Regression; Tracking error; Periodic disturbance; Genetic algorithm; Repetitive control
• ISSN: 0888-3270; 1096-1216
• DOI: 10.1016/j.ymssp.2014.12.014

The application of a repetitive control mechanism for use in a mechanical control system has been a topic of investigation. The fundamental purpose of repetitive control is to eliminate disturbances in a mechanical control system. This paper presents two different repetitive control laws using individual types of basis function feedback and their combinations. These laws adjust the command given to a feedback control system to eliminate tracking errors, generally resulting from periodic disturbance. Periodic errors can be reduced through linear basis functions using regression and a genetic algorithm. The results illustrate that repetitive control is most effective method for eliminating disturbances. When the data are stabilized, the tracking error of the obtained convergence value, 10−14, is the optimal solution, verifying that the proposed regression and genetic algorithm can satisfactorily reduce periodic errors.
•The analytic method of repetitive control proposed in this paper eliminates the interference from a feedback control system.•The start point influences the convergence rate directly.•When the data are stabilized, the tracking error of the obtained convergence value, 10−14, is the optimal solution.