CN105807615A - Fuzzy feedforward-feedback controller - Google Patents

Abstract

The fuzzy feed-forward feedback control algorithm includes the following steps: design a fuzzy PID controller with parameter self-tuning to replace the original PID controller, which is mainly composed of fuzzy inference engine and PID controller, and the controlled variable error e and The change rate ec of the error is used as the input of the fuzzy controller, and the changes of the PID control parameters ΔK _P , ΔK _I , ΔK _D are used as the output. Perform real-time online adjustment to meet the requirements of PID parameter self-tuning at different times. The feedforward control link is added to the parameter self-tuning fuzzy PID controller. When the disturbance occurs, the feedforward control is directly used to act on the control system before it takes effect, and most of the deviation control amount is eliminated, and the remaining deviation is fed back through PID. Control elimination, solving the problem of feedforward control's accuracy requirement on the model and untimeliness of purely using PID controller.

Description

Fuzzy Feedforward Feedback Controller

Technical field

The invention relates to an intelligent control algorithm, in particular to a control algorithm based on the combination of multiple traditional methods.

Background technique

At present, many systems have the characteristics of nonlinearity, large time delay, and large inertia. Although the traditional simple PID controller is simple in design, easy to operate, and easy to implement, it only has linear characteristics and good parameter matching. Good control effect, once the operating point is far away or the parameters change or there is interference, the controller cannot change accordingly, making it difficult for the system to maintain its dynamic quality, which has great limitations. Fuzzy control is a computer control technology based on natural language control rules and fuzzy logic reasoning. It does not depend on the mathematical model of the control system, and can use fuzzy control to realize the adaptive adjustment process of PID parameters. Nowadays, many control processes are often accompanied by a large number of disturbances, and the concept of compensation has been introduced in control theory. Feedforward control was applied to the adjustment of bubble water level as early as 1925. It is adjusted according to the magnitude of the disturbance that causes the controlled parameter to change. In the system of this control strategy, when the disturbance just appears and can be measured, the feedforward control The control sends out an adjustment signal to make the adjustment parameters change accordingly, so that the adjustment effect and the interference effect can be offset in time before the deviation of the controlled parameters occurs. The pure feedforward compensation control is an open-loop control, which can only control the specified disturbance Compensation control, and control deviation, which brings obstacles to the wide application of feedforward control. Feedback control is a technology for adjusting and compensating for the deviation of the system, and it will not appear until a period of time after the disturbance enters the system, which limits the full play of the feedback control function. At present, feedforward control and feedback control are usually combined to form a feedforward-feedback compound control system. Therefore, multiple technologies can be combined to improve the comprehensive response capability of the system. The OPT optimization strategy is used to improve the Smith predictive control and combine it with PID control, but both the PID controller and the Smith predictor are designed based on precise mathematical models, and have poor adaptability to model parameter changes. Due to the complexity of the situation, it is difficult to obtain an accurate mathematical model, so it is still difficult to achieve a satisfactory control effect for this scheme. At present, most of the optimization of PID controller is to combine it with neural network or fuzzy control, so various compound controllers are constantly emerging. Currently, there are two types of fuzzy adaptive PID controllers and neural PID controllers. The combination and introduction of various new technologies have promoted the development of intelligent control technology.

Contents of the invention

The purpose of the present invention is to provide a fuzzy feedforward feedback controller. Aiming at the problems of poor adaptability and untimely control based on the traditional PID controller model in the lack of accurate mathematical models and nonlinear, large time-delay, and large inertia systems, a combination of fuzzy control and feedforward-feedback control is proposed. A fuzzy feedforward feedback controller. Based on the conventional PID controller, the scheme replaces the traditional PID controller with a parameter self-tuning fuzzy PID controller, and introduces the concept of feedforward and feedback control, thereby ensuring the stability of the system state and enhancing the robustness of the controller. Stickiness and adaptability. The controller reduces the influence of the controller on the model parameter changes, reduces the overshoot of the system, and the controller can still achieve a good control effect no matter there is interference in the interference source or interference channel.

In order to achieve the above object, the present invention adopts following technical scheme:

The fuzzy feedforward feedback control algorithm includes the following steps:

(1) The fuzzy PID controller with parameter self-tuning is designed to replace the original PID controller, which is mainly composed of fuzzy reasoning machine and PID controller. The controlled variable error e and error change rate ec are used as fuzzy The input of the controller takes the variation of PID control parameters ΔK _P , ΔK _I , and ΔK _D as the output, and uses the fuzzy control rules summarized by experts and practice to adjust the variation of the three parameters of the PID controller online in real time to meet different requirements. Time PID parameter self-tuning requirements.

(2) The feed-forward control link is added to the fuzzy PID controller with parameter self-tuning. When the disturbance occurs, the feed-forward control is directly used to act on the control system before it takes effect, and most of the deviation control amount is eliminated, and the remaining deviation is Through the elimination of PID feedback control, the problem of the precision requirement of the feedforward control on the model and the untimeliness of simply using the PID controller are solved.

(3) The closed-loop transfer function of the disturbance to the controlled variable described in the algorithm is: Applying the non-deformation condition, P(s)≠0, Y(s)≡0, the transfer function of the feedforward controller is obtained: G _PD (s)+G _ff (s)G _PC (s)＝0, namely It can be seen from the transfer function that when the model or working conditions change, the model will also change accordingly, so it is necessary to add a feedback link to compensate. The new scheme replaces the traditional PID controller with an adaptive PID controller with fuzzy control, and introduces a feed-forward feedback control link, thereby ensuring the stability of the system state and making it have better following characteristics and suppression of disturbances. It enhances the adaptive ability of the controller to the disturbance, has good stability and robustness, and is suitable for the control of the large lagging system.

Description of drawings

Fig. 1 is a flow chart of fuzzy feedforward feedback controller formation.

Fig. 2 is a comparison diagram of fuzzy feedforward feedback controller and corresponding curve of feedforward feedback controller when step disturbance is added to the disturbance channel.

Fig. 3 is a comparison diagram of the fuzzy feedforward feedback controller and the corresponding curve of the feedforward feedback controller when an interference signal is added to the input end.

detailed description:

Fig. 1 is a flow chart of fuzzy feedforward feedback controller formation. Intelligent control algorithm of the present invention comprises the following steps:

(1) Replace the original PID controller with a parameter self-tuning fuzzy PID controller, which is mainly composed of fuzzy reasoning machine and PID controller. The controlled variable error e and error change rate ec are used as fuzzy control The input of the PID controller is based on the variation of PID control parameters ΔK _P , ΔK _I , and ΔK _D as output, and the fuzzy control rules summarized by experts and practice are used to adjust the variation of the three parameters of the PID controller online in real time to meet the needs of different time periods. PID parameter self-tuning requirements.

(2) The feed-forward control link is added to the fuzzy PID controller with parameter self-tuning. When the disturbance occurs, the feed-forward control is directly used to act on the control system before it takes effect, and most of the deviation control amount is eliminated, and the remaining deviation is Through the elimination of PID feedback control, the accuracy requirements of the feedforward control on the model and the untimely problem of using the PID controller alone are solved, and the stability and anti-interference ability of the system are improved.

Fig. 2 is a comparison diagram of fuzzy feedforward feedback controller and corresponding curve of feedforward feedback controller when step disturbance is added to the disturbance channel.

Fig. 3 is a comparison diagram of the fuzzy feedforward feedback controller and the corresponding curve of the feedforward feedback controller when an interference signal is added to the input end.

Claims (2)

1. Fuzzy feed-forward feedback control method, comprising the following steps: (1) Design parameter self-tuning fuzzy PID controller to replace the original PID controller, which is mainly composed of two parts: fuzzy inference module and PID control module, and the error e of the controlled variable and the rate of change ec of the error are used as fuzzy inference The input of the controller uses the fuzzy rules to adjust the variation of the three PID parameters online, and takes the variation of the PID control parameters ΔK _P , ΔK _I , and ΔK _D as the output, plus the initial setting value of the PID controller, and then Act on the controlled object through the PID controller to meet the requirements of PID parameter self-tuning at different times. (2) The feed-forward control link is added to the fuzzy PID controller with parameter self-tuning. When the disturbance occurs, the feed-forward control is directly used to act on the control system before it takes effect, and most of the deviation control amount is eliminated, and the remaining deviation is Through the elimination of PID feedback control, the problem of the precision requirement of the feedforward control on the model and the untimeliness of simply using the PID controller are solved. (3) The closed-loop transfer function of the disturbance to the controlled variable described in the algorithm is: Applying the non-deformation condition, P(s)≠0, Y(s)≡0, the transfer function of the feedforward controller is obtained: G _PD (s)+G _ff (s)G _PC (s)＝0, namely It can be seen from the transfer function that when the model or working conditions change, the model will also change accordingly, so it is necessary to add a feedback link to compensate. 2. fuzzy feed-forward feedback control method according to claim 1, replaces traditional PID controller with the self-adaptive PID controller of band fuzzy control, and introduces feed-forward feedback control link, and then has guaranteed the stability of system state, It has good following characteristics and restraining effect on disturbance, enhances the controller's adaptive ability to disturbance, has good stability and robustness, and is suitable for the control of large lagging systems.