Internal Model Control Strategy Of Complex Object Based On Active Disturbance Rejection Structure

The most basic requirement of the control system design is that the controlled variables can track the given reference signal quickly and accurately. However, the system often can not be controlled to maintain the set value because of some elements in the actual system, these elements collectively referred to as disturbance. In order to control the variables do not occur offset or as soon as possible back to the set value,in the design of the control system, it must be considered of the disturbance rejection effect of the system. When good tracking of system was pursued by conventional control method, the result in disturbance rejection effect is often not ideal, so more effective controller need be designed for the disturbance rejection performance of the system. In this paple, on the basis of earlier research, studies the following several aspects to use equivalent conversion, the disturbance observer and effective open loop transfer function:, this paper studies the following aspects using equivalent transform, disturbance observer and effective open-loop transfer function:1. A novel intelligent cluster cuckoo algorithm is proposed to solve the closed-loop identification problem under different test signals.Currently, most of the identification methods can only identify a particular input signal, but can not applied to the square wave signal,white noise input signal, etc. In order to solve the problem of parameter estimation of different test signal systems, this paper presents a cluster cuckoo algorithm. Different from the basic cuckoo search algorithm, this algorithm increases the probability of discovery, which makes the algorithm have a strong global search capability in the search range. The simulation consequence show that the new analyzing algorithm can improve the accuracy of the identification results compared with the particle swarm optimization algorithm.2. An internal model control method based on active disturbance rejection structure is proposed. In the design of multi-variable system controller, an effective method is to use an effective open-loop transfer function to decompose the system into multiple single-loop systems. In the decomposition process, errors will be inevitably produced. For reducing the influence of errors on system performance, an internal model control method based on active disturbance rejection structure is proposed and applied to a multi-variable system with time delay. The simulation results show that this method not only can effectively reduce the influence of errors on system performance, but also has good robustness and disturbance rejection.3. The active disturbance rejection internal model structure is applied to the two - input - two - output system with time- delay, which solves the problem of model reduction error when designing the controller. In the design of two-input-two-output system controller, a multi-loop control system is decomposed into independent single-loop by decoupling matrix and get a complex equivalent model. In order to facilitate the design of the controller, the Mclaughlin series method is used to reduce the order of equivalent model. Based on the reduced model, IMC-PID method is used to adjust the controller parameters.Simulation results show that the proposed method can reduce the influence of the order-reduced error and improve the disturbance rejection ability and robustness of the control system.4. The active disturbance rejection structure is extended to the singular system and the active disturbance rejection internal model control of the singular system is studied. A robust IMC method based on the Kronecker canonical multivariable singular system is proposed, which provides a simpler and convenient method for the analysis of the singular linear system and control. The method first decomposes the controlled object into a fast subsystem and a slow subsystem by using restricted equivalence. Then, the feedforward controller is designed to eliminate the impulse of the fast subsystem, and the internal model controller is designed for the slow sub-system. In order to facilitate the realization of the feedforward controller, the time scale conversion is introduced. The method is also proved to be suitable for no-pulse singular systems. In order to improve the control precision, using the integral square error as the index, the cluster cuckoo algorithm is used to search the filter parameters of the system. Finally, the simulation consequence show the availability of new analyzing method.