Optimal Design, Based On The Pid Parameters Of The Correlation Method Of System Identification

Ordinary PID control as a traditional control method is used in process control widely because of its small computation, good real-time and the easiness to realize.However, in real practice, many of the mechanisms of controlled process are complex, highly nonlinear, time-varying and uncertain. In the influence of noise and load disturbance, process parameter and even model structure will change with the variation of time and work condition. In addition, with the rapid development of modern industrialize, when designing process control system, both the higher stability and dynamic static performance, as well as the good robustness are demanded .In the case, the traditional PID control, which is based on linear, is often hard to obtain the optimal control effect.As for the design and usage of the PID controller, the core problem is PID parameter tuning. When the controlled subject changes, how to deal with the conflict dynamic and static performance, follow-setting and disturbance-rejecting, robustness and control performance are the topic of many researchers and producers. To solve this problem, people put forward many theories and improve techniques. A lot of PID parameter tuning methods come up all the time.The PID parameter tuning of non-linear control system is set as the engineering background. The aim is the optimal design of PID parameter. On the basis of the design and research of the advanced PID parameter tuning, the mathematical model-transfer functionis developed through correlation analysis on statistical method.And with the optimization algorithm under constraint condition, the non-linear, Isqnonlin is set as object function. Then with connection of the non-linear constrain optimization algorithm and simulink technique, a set of adjust parameters which fit the performance index of the control system are found. And optimization design process and steps are provided. A simulation experiment of the robustness for second-order model and third-order model is made by combining example. Using the optimization design method enables non-linear system to not only meet the needs for performance index, possess a good control specific property but also have strong robustness.In the end, the following direction of research and improvement is put forward on the base of the realization for simulation system.