Controller performance plays an important role in the servo control system of the photoelectronic detection system. However, it is difficult to redesign the controller parameters in real time when the system performance degrades. This paper proposed an adaptive PI controller using Model Reference Adaptive Control (MRAC) techniques with the MIT rule and applied it to the current-loop of the servo control system. In the MRAC, the desired behavior of the process is specified by a model, and the parameters of the controller are adjusted based on the difference between outputs of the closed-loop process and the model. The MIT rule is an adjustment mechanism that makes the error becomes zero. The presented algorithm is carried out in two steps. Firstly, given the time and domain specification, a reference model based on the open-loop transfer function of the plant and a PI controller is designed to parallel with the real process. The adaptation gains of the MIT rule and the external signal are determined to guarantee the convergence of the parameters of the PI controller and minimize the squared model error. Secondly, the adaptation is started and the process is excited adequately in order to enable the parameter adaptation to take place. When the parameters have adapted after a period of time, then the adaptation was stopped and the PI controller operates again with fixed parameters. The proposed algorithm was simulated on the servo control system of the photoelectronic detection system and the results show the effectiveness of the algorithm.

Date Published: 22 February 2008
PDF: 8 pages
Proc. SPIE 6622, International Symposium on Photoelectronic Detection and Imaging 2007: Laser, Ultraviolet, and Terahertz Technology, 66220U (22 February 2008); doi: 10.1117/12.790809

Published in SPIE Proceedings Vol. 6622:
International Symposium on Photoelectronic Detection and Imaging 2007: Laser, Ultraviolet, and Terahertz Technology
Liwei Zhou, Editor(s)