Relay Feedback Auto-tuning Of PID Controllers And Its Application

PID is a control method which is most widely used in industrial control. After dozens of years, the PID control method is still play an important role in each fields. It has the simple structure, good robustness, high reliability, easy algorithm etc. With the development of artificial intelligence, PID combine with some advanced control algorithms and break the limitation of the traditional PID. Therefore, PID has a greater breakthrough in the control field. PID controller has three important parameters: the proportion, integral and differential. Each parameter values mutual combine to realize the powerful function of PID controller. However, in the process of industrial the PID parameters setting has been a difficult problem. Therefore, the research of the PID controller parameter setting method has important a significance.The Relay auto-tuning method which was put forward by Astrom and Hagglund has successfully used in industry for more than 25 years. Combine with the control characteristics of relay feedback closed-loop and the method of auto-tuning PID controller,the Relay auto-tuning method is a simple and good robustness auto-tuning method. In this paper, a model-based relay feedback auto-tuning method for cascade control systems with time delay is proposed. The dynamic model has two types: FOPDT( first-order-plus-dead-time) and SOPDT(second-order-plus-dead-time), therefore, a higher-order system must be reduced order. According to the damping behavior, process is classified into two groups, it’s more intended than ATV test combines with Z-N.rules The controller of each group is derived in terms of ultimate periods and ultimate gains in order to get the intermediate variable of PID parameters. The model-based cascade controller tuning by relay feedback test is adopted by two step auto-tuning method. The model structure is derived by reducing the order of entirety of outer loop process which is using BP neural network and the relay feedback method, and then choose the proper control method by model-based rules. Finally, the simulation results indicate that the model-based relay feedback auto-tuning method is superior to the ATV-ZN in cascade control system.With the thermal power boiler unit upgrade, the parameters of superheated steam temperature control is becoming more and more high quality requirements and it directly affects the safe and economic operation of power plant. The reaction of superheated steam temperature under the disturbances of spraying water is slow, so it is often adopts the cascade control method. However, the equivalent of the controlled object has the large time constant and highly nonlinear, etc. So there are more strict requirements for the cascade control method. In this paper,the relay self-tuning method based on the model was put forward. This method has the better simulation effect, better robust performance, also can adapt to change at any time in the model parameters, and it can adapt the model parameters change at any time.