In industrial practice, PID control

In industrial practice, PID control (proportional plus integral plus derivative control) is the most widely used control method. More than 95% of industrial process control loops are PID structure. PID controller has been used in industrial process control for 35 years, and has become the most important and reliable technical tool. Nowadays, the most common PID model first appeared in the pneumatic regulator in 1930s, and then gradually applied in various fields: machinery, chemical industry, metallurgy, electric power, computer technology, etc.; in 1980s, PID controller developed rapidly with the application of microprocessor. Up to now, PID control is still in the leading position, and with the development of advanced control technology, there are some improvements - the combination of traditional PID controller and advanced control algorithm, such as: fuzzy PID controller, neural network PID controller, predictive PID controller, etc. [1]. PID controller can solve a variety of control problems, it can provide feedback, or eliminate system steady-state error through integral link, or predict the future through differential link. Its structure and algorithm are relatively simple, and there are many kinds of controllers for different model control requirements: proportional (P) controller, proportional integral (PI) controller, proportional differential (PD) controller and proportional integral differential (PID) controller; it has good robust performance, high reliability performance, and does not depend on the model. The following factors need to be considered when designing PID controller: load disturbance, noise measurement, model demand and accuracy [2]. In the actual field, 30% of the control loop adopts manual PID parameter adjustment [3], but the theoretical parameter setting value is often not ideal. This is because the controlled object of industrial process has nonlinear or time-varying uncertainty, it is difficult to establish accurate mathematical model, and the parameters of the system often change, so the process of model identification using modern control theory is not only tedious and difficult, but also difficult to achieve the desired control effect. Therefore, the research on PID parameter tuning method is of great significance and needs further development.