CN101364082A - Human simulation PID intelligence control method of industrial process - Google Patents

Abstract

The invention relates to a human-simulated PID intelligent control method used in industrial processes, which is applicable to the automatic regulation and control of a large-delay large-inertia object in a complex industrial process. The method adopts a distributed control system, that is, a human-simulated PID intelligent controller; and achieves the effect of rapid and stable human-simulated PID intelligent control by using the combination of variable parameter, intelligent integration, opened/closed ring and nonlinear techniques, according to the intensity, the direction and velocity dynamic characteristics of the deviation change of the object. Compared with the prior PID control, the human-simulated PID intelligent control method is superior in the stability, the rapidness and the accuracy. The human-simulated PID intelligent control method omits the complex reasoning and calculation, directly adopts the manually-operated expert experience and simulates the artificial intelligent identification and intelligent decision. As for some hard-to-control characteristics in the complex system in the industrial production, the human-simulated PID intelligent control method achieves good control effect.

Description

The human simulation PID intelligence control method of industrial process

Technical field

The present invention relates to a kind of human simulation PID intelligence control method of industrial process, be applicable to that complex industrial process is delayed greatly, the automatic adjusting and the control of big inertia object (as Boiler Reheated-steam Temperature Control, boiler overheating steam temperature control, main Steam Pressure Control of Circulated).

Background technology

Industrial processes are very complicated, particularly the main heating power controlling object of some of power generation process (as main steam temperature, main vapour pressure system) usually not only have non-linear, factor such as delay greatly, and structural parameters time to time change, its Changing Pattern often are difficult to determine.Classical three controlled variable of PID control method (scale-up factor, integral time, differential coefficient) are the changeless constant of having adjusted, its effective work based on more accurately, fixing mathematical model, this makes the control that it is difficult to realize ideal in these process control.Yet industrial process for the operating mode complexity, manual control has been given enlightenment in this respect, because veteran operating personnel can rely on its understanding to the controlled process characteristic, controlled process is applied suitable manual control, still can obtain comparatively satisfied control effect.Therefore, according to the apery control features, to seek a kind of control method that does not rely on process mathematical model be highly significant.

Summary of the invention

Technical matters to be solved by this invention provides a kind of human simulation PID intelligence control method of industrial process.

The present invention adopts following technical scheme:

This method is the human-simulating PID intelligent controller by adopting scattered control system, according to size, direction and the rate dynamic feature that the object deviation changes, take variable element, Intelligent Integration, switching loops to close non-linear method and realize quick, stable human-simulating PID Based Intelligent Control; Its concrete grammar step is as follows:

(1) when $e\·\stackrel{.}{e}\≤0$ The time, and | e|≤e ₀The time, the human-simulating PID intelligent controller is output as:

u＝u _n-1

(2) when $e\·\stackrel{.}{e}>0$ The time, and | e|〉e ₀The time, the human-simulating PID intelligent controller is output as:

u＝K ₁e+∫K _iedt

(3) when $e\·\stackrel{.}{e}\≤0$ The time, and | e|〉e ₀The time, the human-simulating PID intelligent controller is output as:

u＝K ₀e

(4) when $e\·\stackrel{.}{e}>0$ The time, and | e|≤e ₀The time, the human-simulating PID intelligent controller is output as:

u＝K ₀e+∫K _iedt

In the formula: e is setting value and deviation of measuring value;

Differential for e;

e ₀Be control dead area;

U is the output of human-simulating PID intelligent controller;

u _N-1Be last one output of human-simulating PID intelligent controller constantly;

K ₀Be scale-up factor, obtain by traditional PI D setting method;

K _iBe the storage gain constant, obtain by traditional PI D setting method;

K ₁For the larger proportion coefficient, press 2*K ₀Just get, occurrence is looked the actual effect adjustment.

Described scattered control system adopts the method for model block configuration to realize described human-simulating PID Based Intelligent Control.

Foundation of the present invention, control principle and human-simulating PID intelligent controller:

1, foundation of the present invention:

The basic thought of human simulation PID intelligence control method is in control procedure, utilize computer control system, apish control behavioral function, the characteristic information of discerning to greatest extent and utilizing the control system dynamic process to be provided, inspire and the intuition reasoning, realize the object that lacks accurate model is effectively controlled.Realize Human Simulating Intelligent Control, at first should obtain the characteristic variable of reflection process feature information.Utilize the integral action among the characteristic variable control PID, make integral action meet people's controlling features more.The present invention has proposed a kind of apery non-linearity PID intelligence control method as design basis.

2, the control principle of this method:

(1) when $e\·\stackrel{.}{e}>0$ The time, adopt vast scale (can substantially exceed traditional proportionality constant), strengthen regulating action with quick inhibition overshoot, scale-up factor K=K ₁(big coefficient, desirable 2K ₀, occurrence is adjusted according to the working control effect), when embodying deviation and changing greatly, get bigger ratio, make different operating modes are had certain adaptability;

(2) when $e\·\stackrel{.}{e}\≤0$ The time, controlled volume will fall after rise to desired value, and this moment, ratio should significantly reduce to K ₀(less coefficient);

(3) take the Intelligent Integration strategy, carry out optionally integration according to the different phase imitation manual control that deviation changes; Promptly work as $e\·\stackrel{.}{e}>0$ The time, carry out integration; When $e\·\stackrel{.}{e}\≤0$ The time stop integration;

(4) when system stability and deviation hour, promptly $e\·\stackrel{.}{e}\≤0$ And | e|≤e ₀The time, system is in the open loop waiting status.

3, the described human-simulating PID intelligent controller of this method:

The human-simulating PID intelligent controller adopts scattered control system (DCS), and the DCS of main flow all can adopt the method for model block configuration to realize above-mentioned artificial intelligent pid algorithm.It is the configuration synoptic diagram for Fig. 1.

Traditional PID regulates, and ratio, integral action are crossed conference generation vibration or dispersed; And the human simulation PID intelligence control method effect that can tighten control when suppressing dynamic deviation fast, but can not bring instability, and this is because deviation when crossing limit, and system enters the open loop standby mode at once, makes the rapidity of adjusting and stability obtain unified.This method is in actual putting into operation, and parameter is easy to adjust, direction is clear and definite, the control effect is remarkable.

Beneficial effect of the present invention is as follows: human simulation PID intelligence control method changes not very sensitive to the controlling object parameter, still can guarantee controlled process dullness, non-overshoot and astatic tracking setting value when the time constant of controlled device or when changing retardation time.To having the object that big delay and super large lag behind, no matter human simulation PID intelligence control method all has obvious superiority than PID control aspect stability, rapidity and the accuracy, it has abandoned intricate reasoning and computing, directly absorb manually-operated expertise, imitate artificial Intelligent Recognition, intelligent decision, for some the difficult control characteristics in the commercial production complication system, artificial intelligent PID control method has shown excellent control effect.

Description of drawings

The DCS configuration synoptic diagram that Fig. 1 adopts for the present invention.

In Fig. 1, condition 1 is $e\·\stackrel{.}{e}>0$ And | e|〉e ₀, condition 2 is $e\·\stackrel{.}{e}>0,$ Condition 3 is $e\·\stackrel{.}{e}\≤0$ And | e|≤e ₀SUM is a summation module, and T is a handover module, and M/A is hand/automatic module.

According to shown in Figure 1, progressively each the functional block configuration among Fig. 1 is gone out, connect with signal wire, compiling dress down gets final product.

Embodiment

Get final product according to the technical scheme concrete operations in the foregoing invention content part.

Claims (2)

1, the human simulation PID intelligence control method of industrial process, it is characterized in that this method is the human-simulating PID intelligent controller by adopting scattered control system, according to size, direction and the rate dynamic feature that the object deviation changes, take variable element, Intelligent Integration, switching loops to close non-linear method and realize quick, stable human-simulating PID Based Intelligent Control; Its concrete grammar step is as follows:

(1) when $e\·\stackrel{\·}{e}\≤0$ The time, and | e|≤e ₀The time, the human-simulating PID intelligent controller is output as:

u＝u _n-1

(2) when $e\·\stackrel{\·}{e}>0$ The time, and | e|〉e ₀The time, the human-simulating PID intelligent controller is output as:

u＝K ₁e+∫K _iedt

(3) when $e\·\stackrel{\·}{e}\≤0$ The time, and | e|〉e ₀The time, the human-simulating PID intelligent controller is output as:

u＝K ₀e

(4) when $e\·\stackrel{\·}{e}>0$ The time, and | e|≤e ₀The time, the human-simulating PID intelligent controller is output as:

u＝K ₀e+∫K _iedt

In the formula: e is setting value and deviation of measuring value;

Differential for e;

e ₀Be control dead area;

U is the output of human-simulating PID intelligent controller;

u _N-1Be last one output of human-simulating PID intelligent controller constantly;

K ₀Be scale-up factor, obtain by traditional PI D setting method;

K _iBe the storage gain constant, obtain by traditional PI D setting method;

K ₁For the larger proportion coefficient, press 2*K ₀Just get, occurrence is looked the actual effect adjustment.

2, the human simulation PID intelligence control method of industrial process according to claim 1 is characterized in that described scattered control system adopts the method for model block configuration to realize described human-simulating PID Based Intelligent Control.

Images