Research And Application On Disturbance And Time-delayed Process Control Techniques

Disturbances and time-delay are common phenomenon in the complex process industry, such as Chemical Engineering, Oil refining, Metallurgy, and Paper-making industry. Disturbance has important effect for system performance. Removing and controlling disturbance is key for design robust controller of high performance. Since the controlled variable in time-delayed system can not reflect the manipulate variable in time, it causes high overshoot and long settling time. The stability of closed-loop system will be degraded. It is difficult to control time-delayed system. Hence, there are much important theory and practical value on the study of control on disturbances and time-delay system. More and more researchers devote to research on control method for controlling disturbances and time-delayed processes.It is very difficult to get exact mathematical models of industrial processes, models used in design procedure of control systems are commonly mathematical models through approximation. Uncertainties of mathematical models must be considered in control systems design procedure. Hence, how to deal with effects of uncertainties is becoming urgent requirement, and it is also a big challenge facing control theory and application recently. Some problems in disturbance and time-delayed system control techniques and applications are researched in this thesis. Some improvements of tradition control methods have been made at different degree, and some new control algorithms are also proposed in this thesis.The main working results and contribution of this dissertation are stated as follows:Researched the character of coloured disturbance. A particle filter is proposed for the nonlinear non-Guassian filter problem in the state tracking process of coloured disturbance. A statistical autocorrelation function is proposed for power spectral character of random orders. So, using biased estimator, which produces a stable system with a matrix that numerically is positive definite.Based on the feedforward-feedback control structure, a set of design method and tuning rule of feedforward-feedback control system is proposed based on particle filter. The particle filter is utilized for rejecting the non-Guassian disturbance. Firstly the setpoint tracking controller is designed in terms of the robust optimal performance objective. Then according to the load disturbance rejection requirement during the system steady operation, based on robust optimal performance index, the design of controller is proposed based on the expected sensitivity function. The setpoint response and the load disturbance response are decoupled from each other so that both of them can be separately regulated and optimized on line. The performance is improved.For robust control of time-varying uncertain time-delay disturbance non-linear continuous systems, the characters are researched in time domain. It need not matched uncertainties. The robust control problems and related problems are presented for uncertain time-delay disturbance non-linear continuous systems. The design algorithms are given about robust state feedback controller and robust dynamic output feedback controller. The asymptotic stability of robust controller is guaranteed.The problem of robust control for time-varying disturbance discrete linear system with uncertain time-delay is investigated. Based on LMI method, design algorithms are presented about memoryless robust state feedback controller for system stability and performance index. On the other hand, based on new Lyapunov function, the design method of state feedback controller is presented for time-varying disturbance discrete non-linear systems with uncertain time-delay. The asymptotic stability of robust controller is guaranteed.To demonstrate the validity and superiority of the proposed control structure, a two-degree-of-freedom control structure with particle filter feedforward controller and disturbance observer is designed for PMSM speed control. It has strong rejecting the disturbance. The speed precision of motor can be effectively improved by the scheme. On the other hand, a new repeat control system is proposed for PMSM position control. The tracking precision and system stability of motor can be effectively improved by the scheme. The results of simulation examples show that the proposed control method is satisfied.