Periodic Disturbance Control And Compensation

Sine the 20th century,70th years, a variety of excellent control algorithms have been brought forth, greatly enriched the academic branch of control theory and control engineering. These control algorithms include Smith predictive control algorithm, the inter model algorithm and predict PI algorithm. However, widespread in the industry periodic disturbance has not specifically a treatment program, and have no a good solution to such problems. And the research direction has few been studied in the domestic. Therefore, based on the main loop control, this article carry out control strategy research for periodic disturbance.This paper will design and modify Smith predict algorithm and predictive PI algorithm for the main circuit. And presents a predictive control algorithm based on Smith. In this control algorithm, charged subjects will be divided into delay part and the non-delay part based on the principle of Smith, then we can design instable non-delay part. This algorithm has the advantage of time delay part can be very large, and the controller design is simple and mismatch is effective.Periodic disturbance widely exists in process control, only using the main loop control methods are difficult to control the object. Therefore, a periodic disturbance of the time-delay control is a difficult point of delay control area. This paper will raise a effective control method for the difficulty. For the periodic disturbance time delay control, this paper firstly design repetitive controller and part of frequency identification, then adding the identified frequency to repetitive controller. According to this method, this paper presents a new algorithm of online control for frequency's identification. According to this algorithm, we can identify the frequency of periodic disturbance signal on-line by identification device. And the output frequency will be sent into the controller, then we can get control signal on-line. It cancels the periodic disturbance signal on-line, and the system will be stable ultimately. This algorithm has the advantage of strong frequency tracking, high tracking accuracy, fast system response speed, and large delay-time coefficient.