CN103123460A - Temperature control system and temperature control method - Google Patents

Abstract

The invention relates to a temperature control system and a temperature control method. The system comprises a proportional-integral-derivative (PID) controller based on blur parameter adjustment and a PID controller based on fixed parameters. When epsilon 1 <=Iota e Iota<= epsilon 2, the PID controller based on the blur parameter adjustment is connected and the PID controller based on fixed parameters is disconnected. When Iota e Iota<= epsilon 1, the PID controller based on the blur parameter adjustment is disconnected and the PID controller based on fixed parameters is connected. The method and the system are good in quick in responding, small in overshoot and small in stable state error.

Description

Temperature control system and temperature-controlled process

Technical field

The application relates to a kind of control system and control method, particularly a kind of PID temperature control system and temperature-controlled process.

Background technology

Control system is widely used in various control object, for example is used for the control to temperature.In the use of the equipment such as single crystal growing furnace, sintering furnace, annealing furnace, heat-treatment furnace, Industrial Stoves, electric oven, all need temperature is controlled.These application scenarios are identical to temperature controlled technological requirement basically, namely require the temperature of controlled device to converge on temperature given value with the precision of regulation after the transient process that allows, and remain unchanged within a period of time; Perhaps control temperature and follow the tracks of a certain aim curve that sets in advance with certain precision.Temperature control system is originally as a large inertia, nonlinear time-delay system, and image parameter changes greatly, set up comparatively difficulty of accurate mathematical model, and this has just limited the various practical applications of modern control technology in temperature control system based on mathematical model.

On industrial control unit (ICU) market, mainly contain the control method of following several types at present:

One, PID controls: the PID control structure is simple, robustness is stronger, has occupied 95% of present industrial control unit (ICU) market.But between the Static and dynamic performance, follow the tracks of between setting value and disturbance suppression ability and exist contradiction.This is mainly because P in linear PID controlling (being the proportional control link), I (being integral element), D (being differentiation element) are linear combination, and presents the Nonlinear Mapping relation between system performance variation and controlled quentity controlled variable.

To the system of complex process especially large dead time and large inertia, such as temperature control system etc., conventional pid parameter can not be realized the real-time online adjustment, and is difficult to obtain better control effect.In recent years, intelligent control algorithm is applied to the extensive concern that various complex process controls have caused people.

Two, fuzzy control: fuzzy self-turning PI D parameter control system can detect analysis to factors such as uncertain condition, parameter, delay and interference in control procedure, the method of employing fuzzy reasoning realizes the online self-tuning of pid parameter, the advantages such as PID control system principle is simple, easy to use, robustness is stronger have not only been kept, and has greater flexibility, adaptability, control accuracy is better generally, is comparatively advanced at present a kind of control method.Fuzzy control is a kind of Based Intelligent Control based on language rule and fuzzy reasoning, it imitates the mankind with the control behavior of ambiguity, be control law with the summary of experience of operating personnel's natural language formula, and carry out the process such as fuzzy reasoning based on these rules, generate controlled quentity controlled variable.But fuzzy control is discrete to the processing of input variable, and quantification gradation is limited, and there is no integral element, and during stable state, control accuracy is controlled not as PID.Fuzzy control is combined with conventional PID control, utilize the thought of fuzzy reasoning judgement, according to different deviation (e) and deviation variation rate (ec), the parameter of PID is carried out online self-tuning, just can take into account both advantages.

Three, the Based Intelligent Control take artificial neural network, genetic algorithm, simulated annealing and swarm intelligence technology etc. as representative.Wherein representative have a genetic algorithm: genetic algorithm is based on the searching method of natural selection and gene genetics principle, in the coded strings colony that theory of biologic evolution that will " survival of the fittest; the survival of the fittest " introduces that parameter to be optimized forms, according to certain fitness function and a series of genetic manipulation, each individuality is screened, thereby make the high individuality of fitness be retained, form new colony; New colony comprises the bulk information of previous generation, and has introduced the new individuality that is better than previous generation.Go round and begin again like this, in colony, each individual fitness improves constantly, until satisfy certain maximum conditions.In this moment colony, the highest individuality of fitness is the optimum solution of problem to be optimized.Genetic algorithm can be carried out the global optimization search at complex space, has stronger robustness.In addition, what restrictive hypothesis is genetic algorithm do not need basically for the search volume, as continuously, can be little and unimodal etc.Based on the pid parameter setting method of genetic algorithm, be a kind of adaptive probability search optimization method of Finding Global Optimization, can improve the parameter optimization effect, simplify computation process.

For the control objects such as electric furnace, because temperature delay is very serious, sometimes even reach more than tens of seconds, therefore the above-mentioned existing method of simple employing is easy to cause overshoot, and after overshoot, temperature is difficult to lower, cause control aim curve and response curve to depart from for a long time, have a strong impact on control performance.

Summary of the invention

Provide hereinafter about brief overview of the present invention, in order to basic comprehension about some aspect of the present invention is provided.Should be appreciated that this general introduction is not about exhaustive general introduction of the present invention.It is not that intention is determined key of the present invention or pith, neither be intended to limit scope of the present invention.Its purpose is only that the form of simplifying provides some concept, with this as the preorder in greater detail of discussing after a while.

In order to solve the technical matters that exists in above-mentioned prior art, the present invention will provide a kind of control system and control method, thereby obtain good control accuracy and response process also can reach corresponding requirement.

According to an aspect of the present invention, a kind of temperature control system, for the temperature of controlling controlled system, described temperature control system comprises based on the PID controller of fuzzy accent ginseng and the PID controller of preset parameter;

Work as ε ₁≤ | e|≤ε ₂The time, connect based on the PID controller of fuzzy accent ginseng and disconnect the PID controller of preset parameter;

When | e|＜ε ₁The time, disconnect based on the PID controller of fuzzy accent ginseng and connect the PID controller of preset parameter;

Wherein, e is from the value of feedback of controlled system and the deviation between the control desired value; ε ₁, ε ₂Be default real number, and 0＜ε ₁＜ε ₂

Further, temperature control system also comprises proportional controller,

When | e|＞ε ₂The time, disconnect based on the PID controller of fuzzy accent ginseng and the PID controller of preset parameter, and connect described proportional controller.

Further, the PID controller based on fuzzy accent ginseng in temperature control system adopts offline mode will be input to the described deviation e of described PID controller based on fuzzy accent ginseng and the rate of change discretize of described deviation e, and the parameter that will calculate by obfuscation, fuzzy reasoning and the deblurring of off-line is stored in described temperature control system with matrix form;

When described temperature control system operation, calculate in real time described proportional control factor, integral control coefficient and derivative control coefficient by tabling look-up.

Further, proportional control factor, integral control coefficient and the derivative control coefficient in the PID controller of the preset parameter in temperature control system calculates by genetic algorithm.

Further, in temperature control system, the optimum index J of the PID controller of preset parameter is:

$J=\left\{\begin{array}{cc}\&Integral;\left({\mathrm{<figure-callout\; id="1"\; label="w"\; filenames="HDA0000110436600000021.png,HDA0000110436600000022.png"\; state="\{\{state\}\}">w</figure-callout>}}_1\right|e\left(t\right)|+w_2{u}^2\left(t\right))\mathrm{dt}+w_3{t}_u& e\left(t\right)\&GreaterEqual;0\\ \&Integral;\left({\mathrm{<figure-callout\; id="1"\; label="w"\; filenames="HDA0000110436600000021.png,HDA0000110436600000022.png"\; state="\{\{state\}\}">w</figure-callout>}}_1\right|e\left(t\right)|+w_2{u}^2\left(t\right)+{\mathrm{<figure-callout\; id="4"\; label="w"\; filenames="HDA0000110436600000011.png,HDA0000110436600000021.png"\; state="\{\{state\}\}">w</figure-callout>}}_4|e\left(t\right)\left|\right)\mathrm{dt}+w_3{t}_u& e\left(t\right)<0\end{array}\right.$

Wherein, w ₁, w ₂, w ₃, w ₄Be weighted value, and w ₄＞＞w ₁, u (t) is the output quantity of the PID controller of preset parameter, t _uBe the rise time, e (t) is the time dependent function of difference of the set-point of feedback signal and control system input; Rise time is for rising to the required time of value 90% of steady-state value from 10% of steady-state value.

Any one when further, in the PID controller of preset parameter described in temperature control system, genetic algorithm is calculated pid parameter in employing scale-of-two, Gray code, floating number or real number coding method encoded.

According to a further aspect in the invention, a kind of temperature-controlled process comprises:

Step 1: work as ε ₁≤ | e|≤ε ₂The time, carry out the PID control method based on fuzzy accent ginseng;

Step 2: when | e|＜ε ₁The time, the PID control method of execution preset parameter;

Wherein, e is from the value of feedback of controlled system and the deviation between the control desired value; ε ₁, ε ₂Be default real number, and 0＜ε ₁＜ε ₂

Further, temperature-controlled process also comprises:

Step 3: when | e|＞ε ₂The time, carry out proportional controlling means.

Further, in temperature-controlled process, the PID control method based on fuzzy accent ginseng in described step 1 adopts offline mode will be input to the described deviation e of described PID controller based on fuzzy accent ginseng and the rate of change discretize of described deviation e, and the parameter that will calculate by obfuscation, fuzzy reasoning and the deblurring of off-line is stored in described temperature control system with matrix form;

When described temperature control system operation, calculate in real time described proportional control factor, integral control coefficient and derivative control coefficient by tabling look-up.

Further, proportional control factor, integral control coefficient and the derivative control coefficient in the PID control method of the preset parameter in step 2 described in temperature-controlled process calculates by genetic algorithm.

Further, in temperature-controlled process, the optimum index J in the PID control method of the preset parameter in described step 2 is:

$J=\left\{\begin{array}{cc}\&Integral;\left({\mathrm{<figure-callout\; id="1"\; label="w"\; filenames="HDA0000110436600000021.png,HDA0000110436600000022.png"\; state="\{\{state\}\}">w</figure-callout>}}_1\right|e\left(t\right)|+w_2{u}^2\left(t\right))\mathrm{dt}+w_3{t}_u& e\left(t\right)\&GreaterEqual;0\\ \&Integral;\left({\mathrm{<figure-callout\; id="1"\; label="w"\; filenames="HDA0000110436600000021.png,HDA0000110436600000022.png"\; state="\{\{state\}\}">w</figure-callout>}}_1\right|e\left(t\right)|+w_2{u}^2\left(t\right)+{\mathrm{<figure-callout\; id="4"\; label="w"\; filenames="HDA0000110436600000011.png,HDA0000110436600000021.png"\; state="\{\{state\}\}">w</figure-callout>}}_4|e\left(t\right)\left|\right)\mathrm{dt}+w_3{t}_u& e\left(t\right)<0\end{array}\right.$

Wherein, w ₁, w ₂, w ₃, w ₄Be weighted value, and w ₄＞＞w ₁, u (t) transfers the output quantity of the PID controller of ginseng, t based on genetic algorithm _uBe the rise time, e (t) is the time dependent function of difference of the set-point of feedback signal and control system input; Rise time is for rising to the required time of value 90% of steady-state value from 10% of steady-state value.

Further, in temperature-controlled process, any one when in the PID control method of the preset parameter in described step 2, genetic algorithm is calculated pid parameter in employing scale-of-two, Gray code, floating number or real number coding method encoded.

Adopt temperature control system of the present invention and control method, have the characteristics such as the response rapidity is good, overshoot is little, steady-state error is little.

Description of drawings

Below with reference to the accompanying drawings illustrate embodiments of the invention, can understand more easily above and other objects, features and advantages of the present invention.Parts in accompanying drawing are just in order to illustrate principle of the present invention.In the accompanying drawings, same or similar technical characterictic or parts will adopt same or similar Reference numeral to represent.

Fig. 1 is the structural drawing of a kind of embodiment of temperature control system of the present invention;

Fig. 2 a be based in the PID controller of fuzzy accent ginseng from the value of feedback of controlled system and the membership function figure of a kind of embodiment of the difference e that controls desired value;

Fig. 2 b be based in the PID controller of fuzzy accent ginseng from the value of feedback of controlled system and the membership function figure of a kind of embodiment of the rate of change ec of the difference of controlling desired value;

Fig. 3 is curve map time response of the value of feedback y (t) of controlled system;

Fig. 4 is the process flow diagram of a kind of embodiment of temperature-controlled process of the present invention.

Wherein:

1--is based on the PID controller of fuzzy accent ginseng;

The PID controller of 2--preset parameter;

The 3--actuator;

The 4--controlled system;

The 5--temperature transmitter;

6--differential module;

The 7--switch.

Embodiment

Embodiments of the invention are described with reference to the accompanying drawings.The element of describing in an accompanying drawing of the present invention or a kind of embodiment and feature can combine with element and the feature shown in one or more other accompanying drawing or embodiment.Should be noted that for purpose clearly, omitted expression and the description of unrelated to the invention, parts known to persons of ordinary skill in the art and processing in accompanying drawing and explanation.

Be depicted as the structural drawing of a kind of embodiment of temperature control system of the present invention referring to accompanying drawing 1.Mainly comprise the PID controller 1 based on fuzzy accent ginseng in this control system, the PID controller 2 of preset parameter, actuator 3, controlled system 4, temperature transmitter 5, differential module 6 and switch 7.

Work as ε ₁≤ | e|≤ε ₂The time, connect based on PID controller 1 and the disconnection of fuzzy accent ginseng by switch 7 and transfer the controller 2 of ginseng based on genetic algorithm.At this moment, given temperature value is input in this control system as the input quantity of whole control system, and the PID controller 1 by based on fuzzy accent ginseng outputs to control signal u (t) in actuator 3, by the control of actuator 3 execution to controlled system 4.The output valve of controlled system forms feedback signal y (t) after temperature transmitter 5, the difference of given temperature value (namely controlling desired value) and the feedback signal value of feedback of controlled system (namely from) y (t) is as deviation signal e, then is input in PID controller 1 based on fuzzy accent ginseng.

When | e|＜ε ₁The time, disconnect based on the PID controller 1 of fuzzy accent ginseng and connect the controller 2 of preset parameter by switch 7.By the PID controller 2 of preset parameter, control signal u (t) is outputed in actuator 3, by the control of actuator 3 execution to controlled system 4.The output valve of controlled system forms feedback signal y (t) after temperature transmitter 5, the difference of given temperature value (namely controlling desired value) and the feedback signal value of feedback of controlled system (namely from) y (t) is as deviation signal e, then is input to based on genetic algorithm and transfers in the PID controller 2 of ginseng.

Wherein, ε ₁, ε ₂Be default real number, and 0＜ε ₁＜ε ₂

In one embodiment, proportional control factor, integral control coefficient and the derivative control coefficient in the PID controller 2 of preset parameter can calculate by genetic algorithm.

Below, how to carry out the parameter adjusting and adjust specifically introducing respectively based on the PID controller 1 of fuzzy accent ginseng and the PID controller 2 of preset parameter.

One, the PID based on fuzzy accent ginseng controls

The PID controller comprises proportional control link (P), integration control link (I) and differential controlling unit (D).Wherein, the characteristics of proportional control are: in a single day error produces, and controller just has control action immediately, and controlled variable is changed towards the direction that reduces error, and the power of control action depends on Proportional coefficient K p.The response speed of the larger system of Kp is faster, and the degree of regulation of system is higher, but easily produces overshoot, even can cause system unstable; The Kp value is too small, can reduce degree of regulation, makes system acting slow, extends the adjusting time.

The integration control link can be eliminated the steady-state error of system, but it makes Systems balanth decline usually.Integral action coefficient Ki is larger, and static error is eliminated faster, but Ki is excessive, can cause the larger overshoot of response process, and the number of oscillation increases, and system is with unstable.

The differential controlling unit can be improved the dynamic perfromance of system, and the main deviation that suppresses in response process is forecast change of error in advance to the variation of any direction.But derivative coefficient Kd crosses conference causes larger overshoot, makes the fierce vibration of regulated variable, and system is unstable, extends the adjusting time, reduces the interference free performance of system; If Kd is too little, the differential action too a little less than, quality of regulation is improved little.

Can find out, three parameter value sizes of PID, very large on static characteristics and the dynamic property impact of control system, adjusting of Kp, Ki and three parameters of Kd will be determined according to the mathematical model of control object or the response characteristic of real system.

Adopt the method for fuzzy control that the Kp in the PID controller, Ki and Kd are regulated, to export according to control system exactly (that is to say feedback signal y (t)) response curve to the time, adjust in real time the size of Kp, Ki and Kd, thereby improve dynamic property and the steady-state behaviour of control system.

Generally speaking, the input quantity of deviation signal e as fuzzy controller can be carried out obfuscation to its exact value, become fuzzy quantity, the fuzzy quantity of deviation e can represent with corresponding fuzzy language.So just obtained a subset { A} of the fuzzy language set of deviation.Again by { A} and fuzzy control rule R composition rule by inference carry out decision-making, obtain fuzzy control quantity U, and in continuous system, this fuzzy control quantity U can be expressed as U=Kp*e+Ki* ∫ edt+Kd* (de/dt).

Suppose to get ε ₂=10, that is to say as deviation e at [10, ε ₁] or [ε ₁, 10] time, connect based on PID controller 1 and the disconnection of fuzzy accent ginseng by switch 7 and transfer the controller 2 of ginseng based on genetic algorithm.Deviation e, deviation variation rate ec, Kp, Ki and the domain of Kd on fuzzy set all are decided to be [6,6], and quantizing factor is got Ke=6/10 so, Kec=6/10.

Utilize look-up table can draw Kp, Ki and three parameters of Kd, thereby draw controlled quentity controlled variable U.

In one embodiment, membership function can be taken as trigonometric function, the fuzzy subset of linguistic variable is " NB (negative large), NM (in negative), NS (negative little), ZO (zero), PS (just little), PM (center), PB (honest) ".Be respectively the membership function figure of a kind of embodiment of a kind of membership function figure of embodiment of deviation e and deviation variation rate ec referring to Fig. 2 a, 2b.Wherein ordinate μ is degree of membership.

Below, curve map time response of the feedback signal y (t) that reference is shown in Figure 3 illustrates how to formulate fuzzy rule.

Curve time response of feedback signal y shown in Figure 3 (t) can be divided into OA section, AB section, BC section, CD section and DE section.

1, OA section (e＞0, ec＜0): at this moment system is stable variation tendency, need system response time faster during beginning, therefore adopt larger Kp, and that Ki and Kd want is less, in order to prevent too large overshoot, at this moment will reduce Kp, Ki when near the A point, increase simultaneously Kd, in order to suppress excessive overshoot.

2, AB section (e＜0, ec＜0): at this moment larger Kd to this deviation large future development of negative bias gradually, so it is large to control the change of overshoot, get in system.

3, BC section (e＜0, ec＞0): at this moment require system to get back to as early as possible steady-state value, so Kp and Ki that will be larger when the B point, it is less that Kd wants.When soon near the C point, in order to reduce overshoot, strengthen Kd, suitably reduce Ki, with the vibration of avoiding integration to bring.

4, CD section (e＞0, ec＞0): at this moment system, equally will be as the AB section towards the development of positively biased general orientation, and the change of controlling overshoot is large, gets larger Kd, Kp and Ki.

5, DE section (e＞0, ec＜0): at this moment with the BC section, require system to get back to as early as possible steady-state value, so when the D point has just begun, get larger Kp and Ki, Kd is less.When the E point, increase Kd, suitably reduce Kp and Ki.

According to above-mentioned analysis to parameter, can obtain the fuzzy control rule table to a kind of mode of Kp, Ki and Kd, wherein table 1 is the Kp fuzzy control rule table, and table 2 is the Ki fuzzy control rule table, and table 3 is the Kd fuzzy control rule table.

Table 1Kp fuzzy control rule table

Table 2Ki fuzzy control rule table

Table 3Kd fuzzy control rule table

As a kind of embodiment, can adopt Mamdani (Buddhist nun of Mandan) method to carry out fuzzy reasoning, the fuzzy judgment of output quantity adopts " method of weighted mean ".

Can draw corresponding output for deviation e and change of error ec process Mamdani reasoning.Take Δ KP as example, according to the degree of membership of current time e and ec, can obtain the degree of membership of Δ KP all fuzzy rules under different e and ec.Then according to method of weighted mean, just can obtain in a certain sampling instant, the increment Delta Kp of COEFFICIENT K p can be expressed as:

$\mathrm{\&Delta;Kp}=\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\mathrm{\&Delta;}{\mathrm{Kp}}_i\&CenterDot;{\mathrm{\&mu;}}_{\mathrm{pi}}\left(\mathrm{\&Delta;}{K}_p\right)}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_{\mathrm{pi}}\left(\mathrm{\&Delta;}{K}_p\right)}$

Wherein, Δ Kpi is the i Kp increment size in step, μ _pi(Δ Kp) is the degree of membership of i step Kp.

Similarly, can obtain in a certain sampling instant, the expression formula of the increment Delta Kd of the expression formula of the increment Delta Ki of COEFFICIENT K i and COEFFICIENT K d is respectively:

$\mathrm{\&Delta;Ki}=\frac{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}\mathrm{\&Delta;}{\mathrm{Ki}}_j\&CenterDot;{\mathrm{\&mu;}}_{\mathrm{ij}}\left(\mathrm{\&Delta;Ki}\right)}{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_{\mathrm{ij}}\left(\mathrm{\&Delta;Ki}\right)}$

With $\mathrm{\&Delta;Kd}=\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\mathrm{\&Delta;}{\mathrm{Kd}}_i\&CenterDot;{\mathrm{\&mu;}}_{\mathrm{di}}\left(\mathrm{\&Delta;}{K}_d\right)}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_{\mathrm{di}}\left(\mathrm{\&Delta;}{K}_d\right)}.$

As a kind of embodiment, in order to satisfy the requirement of rapidity in the working control process, can adopt the calculated off-line mode inputting data for example e and ec discretize, calculated off-line is all carried out in fuzzy reasoning, synthetic, judgement etc., the result of fuzzy judgment is added up into matrix form, for tabling look-up in real time in the working control process.For example, the PID controller of fuzzy accent ginseng adopts offline mode will be input to the described deviation e of the PID controller of joining based on fuzzy accent and the rate of change discretize of described deviation e, and the parameter that will calculate by obfuscation, fuzzy reasoning and the deblurring of off-line is stored in temperature control system with matrix form; When temperature control system is moved, calculate in real time described proportional control factor, integral control coefficient and derivative control coefficient by tabling look-up.

Two, adjust based on the pid parameter of genetic algorithm

Based on the pid parameter of the genetic algorithm method for designing of adjusting, be a kind of adaptive probability search optimization method of Finding Global Optimization, need not can improve the parameter optimization effect to the objective function differential, simplify computation process.

1, genetic algorithm brief introduction: according to evolutionism, biological evolution mainly contains three reasons: i.e. heredity, variation and selection.Genetic algorithm is based on the searching method of natural selection and gene genetics principle, in the coded strings colony that theory of biologic evolution that will " survival of the fittest; the survival of the fittest " introduces that parameter to be optimized forms, according to certain fitness function and a series of genetic manipulation, each individuality is screened, thereby make the high individuality of fitness be retained, form new colony; New colony comprises the bulk information of previous generation, and has introduced the new individuality that is better than previous generation.Go round and begin again like this, in colony, each individual fitness improves constantly, until satisfy certain maximum conditions.At this moment, in colony, the highest individuality of fitness is the optimum solution of problem to be optimized.Genetic algorithm is passed through the parameter space coding, and comes the guiding search process towards more efficient future development with random the selection as instrument.Just because of the principle of work of genetic algorithm uniqueness, make it carry out the global optimization search at complex space, have stronger robustness.In addition, genetic algorithm is for the search volume, basically do not need what restrictive hypothesis (as continuously, can be little and unimodal etc.), and other optimized algorithm, as analytical method, often can only obtain locally optimal solution but not globally optimal solution, and need objective function continuously smooth and can be little.

2, the basic step of genetic algorithm:

Select coding strategy, parameter sets X and territory are converted to bit string structure space S;

Definition fitness function f (X);

Determine Genetic Strategies, comprise population size, selection, intersection, mutation operator and probability thereof.

Generate initial population P;

Calculate each individual fitness value in colony;

According to Genetic Strategies, genetic operator is acted on population, produce population of future generation;

The iteration stop technology.

Genetic algorithm relates to six large key elements: parameter coding, and the setting of initial population, the design of fitness function, the design of genetic manipulation is controlled the setting of parameter, stopping criterion for iteration.

3, adjust based on the pid parameter of genetic algorithm

Step based on the genetic algorithm PID:

(1) determine decision variable and constraint condition

Three coefficients that PID controls are Kp, Ki, Kd, the span that can determine them according to physical significance and the experience of Kp, Ki, Kd, i.e. constraint condition.

(2) set up Optimized model: for obtaining satisfied transient process dynamic perfromance, adopt Error Absolute Value time integral performance index to select the minimum target function as parameter.Excessive for preventing controlled quentity controlled variable, the quadratic term of access control input in objective function.The optimum index of selecting following formula to select as parameter:

J＝∫(w ₁|e(t)|+w ₂u ²(t))dt+w ₃t _u e(t)≥0

Wherein, e (t) is the time dependent function of difference of the set-point of feedback signal and control system input, and u (t) is controller output, t _uBe rise time (namely rising to 90% required time of steady-state value from 10% of steady-state value), w ₁, w ₂, w ₃Be weighted value.For avoiding overshoot, adopt the punishment function of genetic algorithm simultaneously, in case i.e. overshoot, with overshoot as optimum index one, this moment, optimum index was: if e (t)＜0 has:

J＝∫(w ₁|e(t)|+w ₂u ²(t)+w ₄|e(t)|)dt+w ₃t _u

Wherein, w ₄Be weighted value, and w ₄＞＞w ₁

(3) determine the Code And Decode method.Genetic algorithm has scale-of-two, Gray code, floating number and real number coding method.Need not decoding with real coding, but not too convenient when carrying out genetic manipulation; Use binary coding method, genetic manipulation is convenient, and decoding can obtain optimum solution after processing.

(4) determine the individual evaluation method, namely determine ideal adaptation degree function (Fitness Function).The general process of estimating the ideal adaptation degree is: to individual coded strings decode process after, can obtain individual phenotype; Can calculate corresponding individual target function value by phenotype; According to the type of optimal problem, can obtain individual fitness by target function value by certain transformation rule.

(5) determine the operational factor of genetic algorithm, determine group size M, genetic algebra kg, crossover probability Pc, variation probability P m and weight w according to actual conditions ₁, w ₂, w ₃, w ₄Size.

According to finding the solution of pid parameter problem of tuning, the dimension that depends on the plan variable is 3, represents respectively Kp, Ki, Kd, i.e. decision variable X=[Kp, Ki, Kd].As a kind of embodiment, the fitness function of gene individuality directly can be taken as objective function, and design fitness function F (X)=1/J.J is optimum index.

As a kind of embodiment, the span of parameter K p can be decided to be [0,20], the span of Ki, kd is [0,1].Choose binary string and represent each parameter, coded strings length elects 10 as.Be 3 length that 10 binary string couples together again, form a length and be 30 scale-of-two word string, this word string is the manipulable object of genetic algorithm.

The number of samples that uses in genetic algorithm is 30, and Evolution of Population algebraically is 100, and crossover probability and variation probability are respectively: Pc=0.6, Pm=0.001.

Initialization of population is the chromosome of generation and the same number of number of group size; Every delegation in (M, 3L) M in program * 30 arrays represents a chromosome, and whole array consists of an initial population.

Initial population can produce at random, and the random number in initial population is equally distributed random number between 0～1, and represents 1 between representing 0,0.5～1 between the random number 0～0.5 of regulation generation.

Through after the heredity of certain algebraically, make Kp, the Ki of J optimum and proportional control factor, integral control coefficient and the derivative control coefficient that Kd is final acquisition.

The controller output quantity of bringing for the randomness of effectively avoiding the genetic algorithm initial population and the fluctuation of desired value, computing drops into real control after 50 generations again.This moment, each chromosome of population obtained reasonable optimization, can not bring the large fluctuation of controlled quentity controlled variable.

Above article how to carry out P, I, D based on the PID controller 1 of fuzzy accent ginseng parameter regulate, and P, I in the PID controller 2 of preset parameter, D parameter are how to confirms.

As a kind of preferred version, temperature control system of the present invention can also comprise the proportional controller (not shown), when | e|＞ε 2, disconnects based on the PID controller 1 of fuzzy accent ginseng and the PID controller 2 of preset parameter, and connects proportional controller.Introduce proportional controller, can when control system just brings into operation, by the scale-up factor of Set scale controller, make the response speed of control system fast as much as possible.

Shown in accompanying drawing 4, be the process flow diagram of a kind of embodiment of temperature-controlled process of the present invention.

Temperature-controlled process comprises the steps:

Step 1: work as ε ₁≤ | e|≤ε ₂The time, carry out the PID control method based on fuzzy accent ginseng;

Step 2: when | e|＜ε ₁The time, the PID control method of execution preset parameter;

Wherein, | e| is the absolute value from value of feedback with the difference of controlling desired value of controlled system; ε ₁, ε ₂Be default real number, and 0＜ε ₁＜ε ₂

As a kind of preferred version, when | e|＞ε ₂The time, carry out proportional controlling means.Scale-up factor in proportional controlling means can be decided according to actual conditions.

The above is described in detail some embodiments of the present invention.Adopt temperature control system of the present invention and control method, the characteristics that have that the response rapidity is good, overshoot is little, steady-state error is little etc., thereby be particularly suitable for the large inertia controlled device such as electric furnace.

As to understand in those of ordinary skill in the art, whole or any steps or the parts of method and apparatus of the present invention, can be in the network of any computing equipment (comprising processor, storage medium etc.) or computing equipment, realized with hardware, firmware, software or their combination, this is that those of ordinary skills use their basic programming skill just can realize in the situation that understand content of the present invention, does not therefore need to illustrate at this.

In addition, it is evident that, when relating to possible peripheral operation in superincumbent explanation, will use undoubtedly any display device and any input equipment, corresponding interface and the control program that are connected with any computing equipment.Generally speaking, related hardware in computing machine, computer system or computer network, software and realize hardware, firmware, software or their combination of the various operations in preceding method of the present invention namely consist of equipment of the present invention and each building block thereof.

Therefore, based on above-mentioned understanding, purpose of the present invention can also realize by program of operation or batch processing on any messaging device.Described messaging device can be known common apparatus.Therefore, purpose of the present invention also can be only by providing the program product that comprises the program code of realizing described method or equipment to realize.That is to say, such program product also consists of the present invention, and storage or the medium that transmits such program product also consist of the present invention.Obviously, described storage or transmission medium can be well known by persons skilled in the art, and therefore storage or the transmission medium of any type that perhaps develops in the future also there is no need at this, various storages or transmission medium to be enumerated.

In equipment of the present invention and method, obviously, after can decomposing, make up and/or decompose, each parts or each step reconfigure.These decomposition and/or reconfigure and to be considered as equivalents of the present invention.The step that also it is pointed out that the above-mentioned series of processes of execution can order naturally following the instructions be carried out in chronological order, but does not need necessarily to carry out according to time sequencing.Some step can walk abreast or carry out independently of one another.Simultaneously, in the above in the description to the specific embodiment of the invention, can use in one or more other embodiment in same or similar mode for the feature that a kind of embodiment is described and/or illustrated, combined with the feature in other embodiment, or the feature in alternative other embodiment.

Should emphasize, term " comprises/comprises " existence that refers to feature, key element, step or assembly when this paper uses, but does not get rid of the existence of one or more further feature, key element, step or assembly or add.

Although described the present invention and advantage thereof in detail, be to be understood that in the situation that do not exceed the spirit and scope of the present invention that limited by appended claim and can carry out various changes, alternative and conversion.And the application's scope is not limited only to the specific embodiment of the described process of instructions, equipment, means, method and step.The one of ordinary skilled in the art will readily appreciate that from disclosure of the present invention, can use according to the present invention carry out with the essentially identical function of corresponding embodiment described herein or obtain result essentially identical with it, existing and want exploited process, equipment, means, method or step future.Therefore, appended claim is intended to comprise such process, equipment, means, method or step in their scope.

Claims (12)

1. temperature control system is used for controlling the temperature of controlled system, it is characterized in that:

Described temperature control system comprises based on the PID controller of fuzzy accent ginseng and the PID controller of preset parameter;

Work as ε ₁≤ | e|≤ε ₂The time, connect based on the PID controller of fuzzy accent ginseng and disconnect the PID controller of preset parameter;

When | e|＜ε ₁The time, disconnect based on the PID controller of fuzzy accent ginseng and connect the PID controller of preset parameter;

Wherein, e is from the value of feedback of controlled system and the deviation between the control desired value; ε ₁, ε ₂Be default real number, and 0＜ε ₁＜ε ₂

2. temperature control system according to claim 1 is characterized in that: also comprise proportional controller,

When | e|＞ε ₂The time, disconnect based on the PID controller of fuzzy accent ginseng and the PID controller of preset parameter, and connect described proportional controller.

3. temperature control system according to claim 1 and 2, it is characterized in that: described PID controller based on fuzzy accent ginseng adopts offline mode will be input to the described deviation e of described PID controller based on fuzzy accent ginseng and the rate of change discretize of described deviation e, and the parameter that will calculate by obfuscation, fuzzy reasoning and the deblurring of off-line is stored in described temperature control system with matrix form; When described temperature control system operation, calculate in real time described proportional control factor, integral control coefficient and derivative control coefficient by tabling look-up.

4. temperature control system according to claim 1 and 2, it is characterized in that: proportional control factor, integral control coefficient and derivative control coefficient in the PID controller of described preset parameter calculate by genetic algorithm.

5. temperature control system according to claim 4, it is characterized in that: the optimum index J of the PID controller of described preset parameter is:

$J=\left\{\begin{array}{cc}\&Integral;\left(w_1\right|e\left(t\right)|+w_2{u}^2\left(t\right))\mathrm{dt}+w_3{t}_u& e\left(t\right)\&GreaterEqual;0\\ \&Integral;\left(w_1\right|e\left(t\right)|+w_2{u}^2\left(t\right)+w_4|e\left(t\right)\left|\right)\mathrm{dt}+w_3{t}_u& e\left(t\right)<0\end{array}\right.$

Wherein, w ₁, w ₂, w ₃, w ₄Be weighted value, and w ₄＞＞w ₁, u (t) is the output quantity of the PID controller of preset parameter, t _uBe the rise time, e (t) is the time dependent function of difference of the set-point of feedback signal and control system input; Rise time is for rising to the required time of value 90% of steady-state value from 10% of steady-state value.

6. temperature control system according to claim 4 is characterized in that: adopt any one in scale-of-two, Gray code, floating number or real number coding method to encode when in the PID controller of described preset parameter, genetic algorithm is calculated pid parameter.

7. temperature-controlled process is used for controlling the temperature of controlled system, it is characterized in that, comprising:

Step 1: work as ε ₁≤ | e|≤ε ₂The time, carry out the PID control method based on fuzzy accent ginseng;

Step 2: when | e|＜ε ₁The time, the PID control method of execution preset parameter;

Wherein, e is from the value of feedback of controlled system and the deviation between the control desired value; ε ₁, ε ₂Be default real number, and 0＜ε ₁＜ε ₂

8. temperature-controlled process according to claim 6, is characterized in that, also comprises:

Step 3: when | e|＞ε ₂The time, carry out proportional controlling means.

9. temperature-controlled process according to claim 7 is characterized in that:

The PID control method based on fuzzy accent ginseng in described step 1 adopts offline mode will be input to the described deviation e of described PID controller based on fuzzy accent ginseng and the rate of change discretize of described deviation e, and the parameter that will calculate by obfuscation, fuzzy reasoning and the deblurring of off-line is stored in described temperature control system with matrix form;

When described temperature control system operation, calculate in real time described proportional control factor, integral control coefficient and derivative control coefficient by tabling look-up.

10. according to claim 7 or 8 described temperature-controlled process, it is characterized in that: proportional control factor, integral control coefficient and derivative control coefficient in the PID control method of the preset parameter in described step 2 calculate by genetic algorithm.

11. temperature-controlled process according to claim 10 is characterized in that:

Optimum index J in the PID control method of the preset parameter in described step 2 is:

$J=\left\{\begin{array}{cc}\&Integral;\left(w_1\right|e\left(t\right)|+w_2{u}^2\left(t\right))\mathrm{dt}+w_3{t}_u& e\left(t\right)\&GreaterEqual;0\\ \&Integral;\left(w_1\right|e\left(t\right)|+w_2{u}^2\left(t\right)+w_4|e\left(t\right)\left|\right)\mathrm{dt}+w_3{t}_u& e\left(t\right)<0\end{array}\right.$

Wherein, w ₁, w ₂, w ₃, w ₄Be weighted value, and w ₄＞＞w ₁, u (t) transfers the output quantity of the PID controller of ginseng, t based on genetic algorithm _uBe the rise time, e (t) is the time dependent function of difference of the set-point of feedback signal and control system input; Rise time is for rising to the required time of value 90% of steady-state value from 10% of steady-state value.

12. temperature-controlled process according to claim 11 is characterized in that:

Any one when in the preset parameter PID control method in described step 2, genetic algorithm is calculated pid parameter in employing scale-of-two, Gray code, floating number or real number coding method encoded.

Images