Controller is the core component in the automation control system, responsible for receiving sensor signals, processing data, issuing control instructions to achieve precise control of the controlled object. Controller control methods are varied, and different control methods are suitable for different control scenarios and needs. In this paper, we will introduce in detail several control methods often used by controllers, including PID control, fuzzy control, adaptive control, predictive control, neural network control and intelligent control.
1. PID control
PID control (Proportional-Integral-Derivative Control) is a classical control method, widely used in industrial production, aerospace, transportation, etc. PID controller controls the controlled object through the three links of Proportional (P), Integral (I) and Derivative (D).
1.1 Proportional control
Proportional control is the basis of PID control, the control law is: u (t) = Kp * e (t), where u (t) for the control quantity, Kp for the proportionality coefficient, e (t) for the deviation. The main function of proportional control is to reduce the deviation and improve the response speed of the system.
1.2 Integral Control
The function of integral control is to eliminate the static difference of the system and improve the stability of the system. The control law is: u(t) = u(t-1) + Ki * ∫e(t)dt, where Ki is the integral coefficient.
1.3 Differential control
The main function of differential control is to suppress the oscillation of the system and improve the anti-interference ability of the system. Its control law is: u(t) = u(t-1) - Kd * de(t)/dt, where Kd is the differential coefficient.
1.4 Characteristics of PID control
PID control has the advantages of simple structure, easy adjustment of parameters, adaptability and so on, but at the same time there are some limitations, such as poor control of nonlinear and time-varying systems, and higher requirements for the adjustment of parameters.
2. Fuzzy control
Fuzzy control is a type of control based on fuzzy logic, which is suitable for dealing with uncertainty and ambiguity. The fuzzy controller realizes the control of the controlled object through three parts: fuzzy rule base, fuzzy inference machine and defuzzifier.
2.1 Fuzzy Rule Base
The fuzzy rule base is the core of the fuzzy controller, which contains a series of fuzzy rules for describing the relationship between input variables and output variables. A fuzzy rule is of the form IF input variable IS fuzzy set, then output variable IS fuzzy set.
2.2 Fuzzy Inference Machine
The fuzzy inference machine reasons about the input variables according to the rules in the fuzzy rule base to get the fuzzy values of the output variables. The process of fuzzy inference includes four steps: fuzzification, rule matching, rule fusion and defuzzification.
2.3 Defuzzifier
The role of defuzzifier is to convert the fuzzy values obtained from fuzzy reasoning into actual control quantities. Commonly used defuzzification methods include maximum affiliation method, weighted average method, etc.
2.4 Characteristics of fuzzy control
Fuzzy control has the ability to deal with uncertainty and fuzzy problems, with low requirements for parameter adjustment and high adaptability. However, fuzzy control also has some limitations, such as the construction of the rule base requires a lot of experience and knowledge, and the control accuracy is affected by the division of the fuzzy set and the inference method.
3. Adaptive control
Adaptive control is a kind of control method that can automatically adjust the control parameters according to the characteristics of the controlled object and environmental changes. Adaptive controller usually includes three parts: model identification, parameter estimation and control law design.
3.1 Model Recognition
Model identification is the basis of adaptive control, through the input and output data to establish the mathematical model of the controlled object, to provide a basis for parameter estimation and control law design.
3.2 Parameter Estimation
Parameter estimation is to estimate the parameters of the controlled object online according to the information obtained from model identification, which provides real-time parameter information for control law design.
3.3 Control law design
The control law design is to design the control law adapted to the characteristics of the controlled object and environmental changes according to the results of model identification and parameter estimation, so as to realize the precise control of the controlled object.
3.4 Characteristics of adaptive control
Adaptive control has the ability to adapt to the characteristics of the controlled object and environmental changes, and can realize the control of nonlinear and time-varying systems. However, adaptive control also has some limitations, such as the accuracy of model identification and parameter estimation affects the control effect, and the design of control law is complicated.
4. Predictive control
Predictive control is a control method based on future prediction information, which realizes the optimal control of the controlled object by predicting the future behavior of the controlled object.
4.1 Predictive model
Predictive model is the basis of predictive control, used to describe the dynamic behavior of the controlled object. Commonly used prediction models are ARX model, BJ model and so on.
4.2 Prediction Algorithm
The prediction algorithm predicts the future behavior of the controlled object according to the prediction model and the current input and output information. Commonly used prediction algorithms include recursive least squares, Kalman filter, etc.
4.3 Optimization control
Optimal control is based on the prediction results, through the optimization algorithm to solve the optimal control law, to achieve optimal control of the controlled object. Commonly used optimization algorithms are linear programming, quadratic programming and so on.