Control System With Large Time Delay And Controller's Performance Assessment

There are some reasons such as technology,system design,object properties make the phenomenon of most control loops are lagged in the chemical,thermal and other process industries.The existence of the lag makes the controller design,controller parameters tuning difficulties.In the thesis,the PID controller is analyzed by amplitude margin and phase margin.It is concluded that the greater the lag time,the smaller the controller gain is required to meet the stability of the closed-loop control system.So the use of PID controller in the hysteresis system cannot meet the stability of the system at the same time the speed of the system output tracking set value.In order to comprehensively understand the lagged object,this topic mainly studies the system identification,control and performance assessment of the controller.In order to meet the problem of response time and stability of the critical stability system,the identification excitation signal chosen in the thesis is the bias relay signal with hysteresis.The relay signal forms a negative feedback closed loop with the lagged object.The traditional relay feedback identification method transfers the system output cycle oscillation signal into the time domain,and then read the oscillation period and the oscillation amplitude to resolve the calculation of the object parameters.However,the amplitude is too easily affected by the interference to the identification results are not accurate even completely untrustworthy.In the thesis,by reading the oscillation output of half a cycle,using the least squares method to identify the object model,greatly improve the accuracy of the object identification.At the same time,compared with the least squares identification with the step signal,the system response time is shortened.In the thesis,two improved methods of controlling lagged objects are proposed.First of all,the improved standard auto-disturbance-rejection control(ADRC)loop structure is used to compensate the hysteresis system through the expansion state observer,and the whole ADRC structure is simplified as a compensation module and controller module.The parameters of the observer and the controller parameters in the ADRC are obtained by establishing the minimization performance index H? of the sensitivity function and the complement sensitivity function,and obtained by PSO optimization.Secondly,the double closed-loop control structure of IMC-Dahlin is proposed for critical stable object in the thesis.The inner ring uses the Dahlin controller to stabilize the object and optimize the target by selecting the time constant of the inner loop closed-loop transfer function.The outer ring uses the IMC controller to comprehensively adjust the system's ability to track the set point,controller consumption,system immunity and system robustness.The parameters of the two controllers are given by linear quadratic regulator LQR optimization.Finally,the controller performance assessment method is given for the deterministic interference loop.The thesis takes willing closed-loop transfer function of the IMC controller as the basis to design performance assessment.The time constant values of the closed-loop transfer function need to be given by the user.The output variance of the theory is solved by the linear matrix inequality(LMI),and is used as the performance benchmark value of the controller.The controller performance index is obtained by comparing theory output variance and actual output variance.It is worth noting that any method that can improve the performance of the controller can be used as a tuning method for the controller.