CN105511270B - A kind of PID controller parameter optimization method and system based on coevolution - Google Patents
A kind of PID controller parameter optimization method and system based on coevolution Download PDFInfo
- Publication number
- CN105511270B CN105511270B CN201610080479.3A CN201610080479A CN105511270B CN 105511270 B CN105511270 B CN 105511270B CN 201610080479 A CN201610080479 A CN 201610080479A CN 105511270 B CN105511270 B CN 105511270B
- Authority
- CN
- China
- Prior art keywords
- pid
- population
- parameter
- individual
- group
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a kind of PID controller parameter optimization methods and system based on coevolution, this method is directed to the parameter tuning of complication system PID control, it employs a kind of modified Cooperative Evolutionary to optimize PID control system parameter, makes optimum combination of tri- parameters of PID in a manner of coevolution under the automatic given performance indicator of search.The present invention is the optimum combination for searching for pid parameter automatically in a manner of coevolution under given performance indicator, and fast convergence rate, adaptivity are strong, accuracy is high, can obtain better optimal control effect, have preferable application prospect.
Description
Technical field
The present invention relates to a kind of PID controller parameter optimization methods and system based on coevolution, belong to control system
Parameter optimization technique field.
Background technology
At present for complicated controll plant, the PID control of conventional method is often difficult to meet the requirements, complication system
PID control and its parameter tuning are still the major issue for controlling engineering field that need to solve.In recent years, intelligent Computation Technology by with
It is difficult to solve complication system PID control and its Parametric optimization problem in solving conventional method, produces the PID based on intelligence computation
Controller parameter optimization method, such as the optimization method based on genetic algorithm, particle cluster algorithm.These methods, which achieve some, to be had
The effect of benefit, but also exist simultaneously that combined guided property is not strong, speed of searching optimization is slow, the deficiencies of being easy to be absorbed in local optimum.The present invention
A kind of PID controller parameter optimization method based on coevolution is proposed, can overcome the shortcomings of existing method.
Invention content
Present invention aims in view of the above shortcomings of the prior art, it is proposed that a kind of PID control based on coevolution
Device parameter optimization method, this method are directed to the parameter tuning of complication system PID control, employ a kind of modified coevolution side
Method optimizes PID control system parameter, makes tri- parameters of PID in a manner of coevolution under the automatic given performance indicator of search
Optimum combination.This method fast convergence rate, adaptivity are strong, accuracy is high, can obtain better optimal control effect, have
Preferable application prospect.
The technical scheme adopted by the invention to solve the technical problem is that:A kind of PID controller ginseng based on coevolution
Number optimization method, this method comprises the following steps:
Step 1:Initial setting up.The population of 3 certain scales is set and initializes, individual corresponds to PID controller respectively
Three parameters;Control effect is set to assess object function used;Set evolution end condition.
Step 2:It chooses assessment and represents group.Individual in 3 populations is combined sequentially into be selected group, is controlled respectively as PID
Three parameter imbedding systems of device processed, and activation system is run.It is responded according to corresponding output, is evaluated by object function and respectively treated
The fitness value of choosing group selects one group of best conduct assessment of wherein fitness value and represents group.
Step 3:Each population individual adaptation degree value assessment.With assessment represent representative individual in group respectively with population to be assessed
Individual composition parameter group, merging PID controller and activation system operation.It is responded according to corresponding output, is assessed by object function
Go out the fitness value of each population at individual;Meanwhile with currently more preferably individual update assessment representative group corresponds to individual.
Step 4:Each Evolution of Population.After all individual assessments, each population is evolved respectively according to ideal adaptation angle value,
Produce population of new generation.The various evolution algorithms mined massively can be the same or different.
Step 5:Optimize finishing control.If the not up to evolution end condition of setting, goes to step 2 continuation iteration optimizations;
If reaching the evolution end condition of setting, current assessment representative group is the pid parameter optimization knot as PID control system
Fruit.
The present invention also provides a kind of PID controller parameter optimization system based on coevolution, which includes improving
Type Co-evolution Optimization module, PID control system, control effect evaluation module.Each module concrete function is as follows:
Modified Co-evolution Optimization module:The module is based on modified Cooperative Evolutionary proposed by the present invention, realizes
Coevolution optimizing is carried out to tri- parameters of PID of PID controller, obtains the pid parameter optimum combination under given performance indicator.
PID control system:It is the application of the method for the present invention, the pid parameter of PID controller is exactly to change in the present invention
Into the optimization object of type Co-evolution Optimization module.Modified Co-evolution Optimization module is joined to PID control system output PID
Number, PID control system run out control effect.
Control effect evaluation module:The module can respond the output of specific input according to PID control system, by setting on demand
The object function put assesses the control effect of PID controller, and provides assessed value as current tri- parameter combinations of PID
Fitness value.
PID controller parameter optimization system structure the present invention is based on coevolution is as shown in Figure 1.EA1, EA2 in figure,
EA3 is three evolution algorithm modules.
First, PID control system and its parameter optimization
(1) PID control system essential characteristic
If given yd(t) control deviation e (t) is formed with reality output y (t):
E (t)=yd(t)-y(t)
The controller output u (t) of PID:
Or write as the form of transmission function:
Wherein, kp、kiAnd kdIt is the proportionality coefficient, integral coefficient and differential coefficient of PID controller respectively.PID controller is each
Correction link effect includes as follows:
1st, proportional component:Proportionally reflect the deviation signal e (t) of control system, deviation is once generate, and controller is immediately
Control action is generated to reduce deviation.
2nd, integral element:It is mainly used for eliminating static difference, the power of integral action depends on integral coefficient, and integral coefficient is got over
Small, integral action is weaker, on the contrary then stronger.
3rd, differentiation element:Reflect the change rate of deviation signal, can in systems be introduced before deviation signal becomes too big
One effective early stage revise signal so as to accelerate the movement speed of system, reduces regulating time.
2nd, pid parameter Optimized model
Common optimization object function J has an ITAE, IAE, ITSE these three:
These optimization object functions are not comprising explicit system coherence energy index, such as rise time tr, overshoot
δ %, regulating time tfDeng.
The output quantity u of PID controller is excessive in order to prevent, obtains smaller rise time trAnd smaller regulating time tf,
Object function form can be taken as follows, i.e., (by taking IAE as an example):
Wherein w1,w2,w3For weights.In order to which the overshoot of system is made (it is inclined to set its corresponding maximum overshoot no more than limit value
Difference is es), setting weight w '1:If e (t) < esThen w'1> > w1, at this time:
Finally, the optimization problem of system can be expressed as and (set minimum value as optimal solution):
min J(kp,ki,kd)
s.t.
kp∈[kpmin,kpmax]
ki∈[kimin,kimax]
kd∈[kdmin,kdmax]
3rd, modified Co-evolution Optimization module
Modified Co-evolution Optimization module of the present invention is as shown in Fig. 2, population P in figure1、P2And P3Represent parameter respectively
kp、kiAnd kd, population scale n;Individual in all populations is grouped by number, substitutes into target problem (Domain model)
In and assess corresponding fitness value, choose the best combination of fitness value, group (X represented as assessmentbest,Ybest,Zbest)。
Each population at individual Pi,j(i=1,2,3;J=1,2,3 ..., n) Fitness analysis, group (X is represented by assessmentbest,
Ybest,Zbest) completed to cooperate with.That is population P1Individual P1,jFitness analysis be by (P1,j, Ybest, Zbest) it is updated to target
Problem is assessed, and is obtained it and is corresponded to the fitness value F of individual1,j, and by wherein fitness higher than evaluation representative group most
Excellent P1,jReplace Xbest。
After each all individual assessments of population, according to the fitness value of individual each in population, all each self-evolutions of population
Obtain population of new generation.Evolution algorithm EA1, EA2 and EA3 can be the same or different.
If meeting the end condition evolved, evaluation representative group just as the optimal collaboration solution of problem, is terminated and is calculated.
Conversely, continue iteration optimizing at selection assessment representative group.
The specific steps of modified Co-evolution Optimization method of the present invention include:
The first step:Initialization.Initialize population P1(0), P2(0), P3(0);Put evolutionary generation counter initial value m:=0;If
Surely it evolves stop criterion;
Second step:Selection assessment represents group.Choose population at individual (P1,j(m),P2,j(m),P3,j(m)) (j=1,2,3 ...,
N) to be selected group is formed, assessment calculates every group of fitness value, selects best one group of current fitness value as assessment and represents group
(Xbest,Ybest,Zbest);
Third walks:The assessment of each population at individual fitness value.As shown in Fig. 2, represent group (X with assessmentbest,Ybest,
Zbest) individual in each population is assessed respectively, obtain the fitness value of all individuals;Meanwhile with currently more preferably individual updates
It assesses representative group and corresponds to individual;
4th step:Each population is evolved respectively.Each population is evolved respectively according to ideal adaptation angle value, produces new one
For population P1(m+1), P2(m+1), P3(m+1);Put m:=m+1;
5th step:It terminates and judges.If meeting evolution end condition, evaluation represents group (Xbest,Ybest,Zbest)
With regard to the optimal solution as problem, terminate and calculate;Otherwise, turn second step.
3rd, optimization system method of work of the present invention
As shown in figures 1 and 3, the method for work of optimization system of the present invention includes:
Step 1:Initial setting up
Parameter kpCorresponding population is P1, parameter kiCorresponding population is P2, parameter kdCorresponding population is P3, population scale
For n;Initialization of population is P1(0)、P2(0)、P3(0);Put evolutionary generation counter m=0;Mesh used in control effect assessment is set
Scalar functions (object function J or J as the aforementioned*);Set evolution end condition (such as J or J*Desired value to be achieved or evolve
Maximum algebraically).
Step 2:Assess the selection of representative group
Each population at individual is pressed into (P1,j(m), P2,j(m), P3,j(m)) to be selected group is formed, j takes 1,2,3 successively ..., n, respectively
By P1,j(m), P2,j(m), P3,j(m) as the k of PID controllerp,ki,kdThree parameter imbedding systems, and activation system is run,
Obtain output response.It is responded according to each output, by object function J or J*Each to be selected group of fitness value can be evaluated, selects it
One group of best conduct assessment of middle fitness value represents group (Xbest,Ybest,Zbest)。
Step 3:Each population individual adaptation degree value assessment
Group (X is represented with assessmentbest,Ybest,Zbest) in representative individual Xbest,Ybest,ZbestRespectively with population to be assessed
Individual composition kp,ki,kdParameter group, merging PID controller and activation system operation, obtain output response.It is rung according to each output
Should, by object function J or J*The fitness value of each population at individual can be evaluated.Such as, m is for P1J-th of individual P in population1,j(m)
The acquisition of fitness value:By (P1,j(m), Ybest, Zbest) as kp,ki,kdParameter group is placed in PID controller, activation system operation
Obtain output response;It is responded according to output, by object function J or J*Evaluate P1,j(m) fitness value;With better than current Xbest
P1,j(m) X is updatedbest。
Step 4:Each Evolution of Population
After all individual assessments, each population is evolved respectively according to ideal adaptation angle value, produces population of new generation
P1(m+1), P2(m+1), P3(m+1).Various evolution algorithm EA1, EA2, EA3 to mine massively can be the same or different.Put m:
=m+1.
Step 5:Optimize finishing control
If not up to evolution end condition (such as J or J of setting*Desired value or evolve maximum algebraically), then from step
2 continue iteration optimization;If reach the evolution end condition of setting, current (Xbest,Ybest,Zbest) it is used as PID control system
The k of systemp,ki,kdParameter optimization result.
Advantageous effect:
For the parameter tuning of complication system PID control, the present invention is to search for pid parameter automatically in a manner of coevolution to exist
Optimum combination under given performance indicator, fast convergence rate, adaptivity are strong, accuracy is high, can overcome the shortcomings of existing method,
Better optimal control effect is obtained, there is preferable application prospect.
Description of the drawings
Fig. 1 is the PID controller parameter optimization method system structure diagram the present invention is based on coevolution.
Fig. 2 is modified Co-evolution Optimization module diagram of the present invention.
Fig. 3 is pid parameter Co-evolution Optimization method flow diagram of the present invention.
Fig. 4 compares (1) figure for three kinds of method Tuning PID parameters system step response curves of the present invention.
Fig. 5 compares (2) figure for three kinds of method Tuning PID parameters system step response curves of the present invention.
Specific embodiment
The invention is described in further detail with reference to the accompanying drawings of the specification.
According to system shown in Figure 1 structure, the present invention combines a specific control object, provides the specific implementation of the present invention
Example.
Its PID controller parameter is optimized using the present invention, including:
1st, initial configuration
PID controller parameter kpCorresponding population is P1, parameter kiCorresponding population is P2, parameter kdCorresponding population is
P3;Population scale n takes 50, and for individual using binary coding, code length is 20;Initialization of population is P1(0)、P2(0)、P3
(0);Various to mine massively with identical genetic algorithm, crossover probability takes 0.9, and mutation probability takes 0.01;Put evolutionary generation counter m=
0;Object function used in control effect assessment is aforementioned J or J*;Evolution end condition is set as maximum evolutionary generation 50.
2nd, the selection of representative group is assessed
Each population at individual is pressed into (P1,j(m), P2,j(m), P3,j(m)) to be selected group is formed, j takes 1,2,3 successively ..., and 50, point
Not by P1,j(m), P2,j(m), P3,j(m) as the k of PID controllerp,ki,kdThree parameter imbedding systems, and activation system is transported
Row obtains output response.It is responded according to each output, by object function J or J*Each to be selected group of fitness value can be evaluated, is selected
Wherein one group of best conduct assessment of fitness value represents group (Xbest,Ybest,Zbest)。
3rd, each population individual adaptation degree value assessment
Group (X is represented with assessmentbest,Ybest,Zbest) in representative individual Xbest,Ybest,ZbestRespectively with population to be assessed
Individual composition kp,ki,kdParameter group, merging PID controller and activation system operation, obtain output response.It is rung according to each output
Should, by object function J or J*The fitness value of each population at individual is evaluated, and updates Xbest,Ybest,Zbest。
4th, each Evolution of Population
After all individual assessments, each population is evolved according to ideal adaptation angle value with genetic algorithm respectively, is produced
Population P of new generation1(m+1), P2(m+1), P3(m+1);Put m:=m+1.
5th, optimize finishing control
If the not up to evolution maximum algebraically of setting, continue iteration optimization from (2) step;If reach the evolution of setting most
Advanced algebra, then current (Xbest,Ybest,Zbest) i.e. as the k of PID control systemp,ki,kdParameter optimization result.
6th, simulation result
The J of aforementioned IAE forms and the J of (1) formula is respectively adopted*As object function, simulation run is carried out to the present embodiment.
During using the J of IAE forms as object function, the method for the present invention and Z-N methods and genetic algorithm to such as table 1 and
Shown in Fig. 4.Table 1 provides the k that three kinds of methods are adjustedp,ki,kdValue and system unit step response performance parameter comparison;Fig. 4
K is adjusted for three kinds of methodsp,ki,kdThe unit-step nsponse curve comparison of system afterwards.
Table 1:The k that three kinds of methods are adjustedp,ki,kdValue and system step response parameter comparison (1)
It can see from Fig. 4 and table 1, the rise time t of the pid parameter of the method for the present invention optimization its system step responser、
Overshoot δ % and regulating time tfBetter than other two methods (especially trAnd tfAdvantage is more notable).
Using the J of (1) formula*During as object function, the method for the present invention and Z-N methods and genetic algorithm to such as table 2 and
Shown in Fig. 5.Table 2 provides the k that three kinds of methods are adjustedp,ki,kdValue and system unit step response performance parameter comparison;Fig. 5
K is adjusted for three kinds of methodsp,ki,kdThe unit-step nsponse curve comparison of system afterwards.
J*In weights be set as:w1=0.9, w2=0.1, w3=10, w'1=100
Table 2:The k that three kinds of methods are adjustedp,ki,kdValue and system step response parameter comparison (2)
It can see from Fig. 5 and table 2, the rise time t of the pid parameter of the method for the present invention optimization its system step responser、
Overshoot δ % and regulating time tfBetter than other two methods (especially δ % and tfAdvantage is more notable).
The simulation run effect of above example shows that the present invention has preferable optimization performance, can obtain preferably excellent
Change control effect, there is preferable application prospect.It is can also be applied to the parameter optimization of other complex control systems, not
It is limited to the above embodiment, in the knowledge having in professional and technical personnel in the field, the present invention can also be not being departed from
It is made a variety of changes under the premise of objective.
Claims (5)
1. a kind of PID controller parameter optimization method based on coevolution, which is characterized in that the method includes walking as follows
Suddenly:
Step 1:Initial setting up;The population of 3 certain scales is set and initializes, individual corresponds to the three of PID controller respectively
A parameter (kp,ki,kd);Control effect is set to assess object function used;Set evolution end condition;
Step 2:It chooses assessment and represents group;Individual in 3 populations is combined sequentially into be selected group, respectively as PID controller
Three parameter imbedding systems, and activation system run;It is responded according to corresponding output, each to be selected group is evaluated by object function
Fitness value, select wherein fitness value best one group as assessment and represent group;
Step 3:Each population individual adaptation degree value assessment;With assessment represent representative individual in group respectively with population at individual to be assessed
Composition parameter group, merging PID controller and activation system operation;It is responded according to corresponding output, is evaluated respectively by object function
The fitness value of population at individual;Meanwhile with currently more preferably individual update assessment representative group corresponds to individual;
Step 4:Each Evolution of Population;After all individual assessments, each population is evolved respectively according to ideal adaptation angle value, is generated
Go out population of new generation;The various evolution algorithms mined massively are identical or different;
Step 5:Optimize finishing control;If the not up to evolution end condition of setting, goes to step 2 continuation iteration optimizations;If it reaches
To the evolution end condition of setting, then current assessment representative group is the pid parameter optimum results as PID control system;
Population P1Individual P1,jFitness analysis be by (P1,j, Ybest, Zbest) be updated to target problem and assessed, it obtains
It corresponds to the fitness value F of individual1,j, and by wherein fitness be higher than evaluation representative group optimal P1,jReplace Xbest;
M is for P1J-th of individual P in population1,j(m) acquisition of fitness value:By (P1,j(m), Ybest, Zbest) as kp,ki,kd
Parameter group is placed in PID controller, and activation system runs to obtain output response;It is responded according to output, by object function J or J*Assessment
Go out P1,j(m) fitness value, with better than current XbestP1,j(m) X is updatedbest。
2. a kind of PID controller parameter optimization method based on coevolution according to claim 1, which is characterized in that
Individual in the population is grouped by number, is substituted into target problem (Domain model) and is assessed corresponding fitness value,
The best combination of fitness value is chosen, group (X is represented as assessmentbest,Ybest,Zbest);
Each population at individual Pi,j, wherein i=1,2,3;J=1,2,3 ..., n Fitness analysis represent group (X by assessmentbest,
Ybest,Zbest) completed to cooperate with;That is population P1Individual P1,jFitness analysis be by (P1,j, Ybest, Zbest) it is updated to target
Problem is assessed, and is obtained it and is corresponded to the fitness value F of individual1,j, and by wherein fitness higher than evaluation representative group most
Excellent P1,jReplace Xbest;
After each all individual assessments of population, according to the fitness value of individual each in population, all each self-evolutions of population obtain
Population of new generation;
If meeting the end condition evolved, evaluation representative group just as the optimal collaboration solution of problem, is terminated and is calculated;Instead
It, continues iteration optimizing at selection assessment representative group.
3. a kind of PID controller parameter optimization system based on coevolution, it is characterised in that:The system comprises modified associations
With evolutionary optimization module, PID control system, control effect evaluation module;
Modified Co-evolution Optimization module realizes the k to PID controllerp、kiAnd kdThree parameters carry out coevolution and seek
It is excellent, obtain the pid parameter optimum combination under given performance indicator;
Modified Co-evolution Optimization module:The module is based on modified Cooperative Evolutionary as described in claim 1, realizes
Coevolution optimizing is carried out to tri- parameters of PID of PID controller, obtains the pid parameter optimum combination under given performance indicator;
PID control system is the application of the method, and the pid parameter of PID controller is exactly that modified coevolution is excellent
Change the optimization object of module, modified Co-evolution Optimization module exports pid parameter, PID control system fortune to PID control system
Row goes out control effect;
Control effect evaluation module is that the output of specific input is responded according to PID control system, by the target letter set on demand
Several control effects to PID controller are assessed, and provide fitness of the assessed value as current tri- parameter combinations of PID
Value.
4. a kind of PID controller parameter optimization system based on coevolution according to claim 3, which is characterized in that
The PID control system is application, and the pid parameter of PID controller is exactly the excellent of modified Co-evolution Optimization module
Change object;Modified Co-evolution Optimization module exports pid parameter to PID control system, and PID control system runs out control effect
Fruit.
5. a kind of PID controller parameter optimization system based on coevolution according to claim 3, which is characterized in that
The control effect evaluation module can respond the output of specific input according to PID control system, by the target letter set on demand
Several control effects to PID controller are assessed, and provide fitness of the assessed value as current tri- parameter combinations of PID
Value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610080479.3A CN105511270B (en) | 2016-02-04 | 2016-02-04 | A kind of PID controller parameter optimization method and system based on coevolution |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610080479.3A CN105511270B (en) | 2016-02-04 | 2016-02-04 | A kind of PID controller parameter optimization method and system based on coevolution |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105511270A CN105511270A (en) | 2016-04-20 |
CN105511270B true CN105511270B (en) | 2018-07-06 |
Family
ID=55719341
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610080479.3A Active CN105511270B (en) | 2016-02-04 | 2016-02-04 | A kind of PID controller parameter optimization method and system based on coevolution |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105511270B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106054596B (en) * | 2016-06-23 | 2018-10-26 | 东南大学 | It is a kind of that setting method is optimized based on the PID controller parameter for improving performance indicator |
CN109709795A (en) * | 2018-12-24 | 2019-05-03 | 东华大学 | A kind of PID controller parameter setting method based on longicorn palpus searching algorithm |
CN111103790A (en) * | 2019-12-05 | 2020-05-05 | 珠海格力电器股份有限公司 | Parameter setting method and device of PID controller, storage medium, terminal and system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1564416A (en) * | 2004-03-17 | 2005-01-12 | 西安交通大学 | Reactive optimizing method of power system based on coordinate evolution |
CN103105774A (en) * | 2013-01-30 | 2013-05-15 | 上海交通大学 | Fractional order proportion integration differentiation (PID) controller setting method based on improved quantum evolutionary algorithm |
CN103870696A (en) * | 2014-03-19 | 2014-06-18 | 南京邮电大学 | Multi-target optimized co-evolution method based on local-global combinational guidance |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4528984B2 (en) * | 2007-01-29 | 2010-08-25 | 国立大学法人広島大学 | PID control device and PID control method |
KR101388069B1 (en) * | 2012-05-10 | 2014-04-22 | 수원대학교산학협력단 | Method for controlling constant pressure using fuzzy pid controller |
-
2016
- 2016-02-04 CN CN201610080479.3A patent/CN105511270B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1564416A (en) * | 2004-03-17 | 2005-01-12 | 西安交通大学 | Reactive optimizing method of power system based on coordinate evolution |
CN103105774A (en) * | 2013-01-30 | 2013-05-15 | 上海交通大学 | Fractional order proportion integration differentiation (PID) controller setting method based on improved quantum evolutionary algorithm |
CN103870696A (en) * | 2014-03-19 | 2014-06-18 | 南京邮电大学 | Multi-target optimized co-evolution method based on local-global combinational guidance |
Non-Patent Citations (4)
Title |
---|
A Novel Hybrid Fuzzy PID Controller Based on Cooperative Co-evolutionary Genetic Algorithm;Farzad Fadaei,等;《Journal of Basic and Applied Scientific Research》;20131231;337-344 * |
A Uncultivated Wolf Pack Algorithm for Highdimensional Functions and Its Application in Parameters Optimization of PID Controller;Wu Husheng,等;《2014 IEEE Congress on Evolutionary Computation (CEC)》;20140711;1477-1482 * |
基于协同遗传算法的TS-PID控制器参数整定;许振凯,等;《信息技术》;20150930(第9期);16-19,23 * |
考虑非线性的水轮机调节系统协同进化模糊PID仿真;吴罗长,等;《西北农林科技大学学报(自然科学版)》;20130930;第41卷(第9期);229-234 * |
Also Published As
Publication number | Publication date |
---|---|
CN105511270A (en) | 2016-04-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105207253B (en) | Consider wind-powered electricity generation and the probabilistic AGC stochastic and dynamics Optimization Scheduling of frequency | |
CN107272403A (en) | A kind of PID controller parameter setting algorithm based on improvement particle cluster algorithm | |
CN111045326A (en) | Tobacco shred drying process moisture prediction control method and system based on recurrent neural network | |
CN107203687B (en) | Multi-target cooperative intelligent optimization control method for desulfurization process of absorption tower | |
CN109254530A (en) | MFA control method based on grinding process basis circuit | |
CN105511270B (en) | A kind of PID controller parameter optimization method and system based on coevolution | |
CN106026084B (en) | A kind of AGC power dynamic allocation methods based on virtual power generation clan | |
CN108122079A (en) | Computational methods, system and the storage medium of thermal power plant's sharing of load | |
CN104318090A (en) | Least square method support vector machine-based generalized prediction method in lysozyme fermentation process | |
CN116436033A (en) | Temperature control load frequency response control method based on user satisfaction and reinforcement learning | |
CN113110056A (en) | Heat supply intelligent decision-making method and intelligent decision-making machine based on artificial intelligence | |
CN110673482B (en) | Power station coal-fired boiler intelligent control method and system based on neural network prediction | |
CN115795992A (en) | Park energy Internet online scheduling method based on virtual deduction of operation situation | |
CN117093033A (en) | Resistance heating furnace temperature control system for optimizing PID parameters based on particle swarm optimization | |
CN112012875A (en) | Optimization method of PID control parameters of water turbine regulating system | |
CN109149558B (en) | Cross-regional power grid dispatching AGC method based on system series compensation | |
CN114909706A (en) | Secondary network balance regulation and control method based on reinforcement learning algorithm and pressure difference control | |
Löppenberg et al. | Self optimisation and automatic code generation by evolutionary algorithms in PLC based controlling processes | |
CN114384931A (en) | Unmanned aerial vehicle multi-target optimal control method and device based on strategy gradient | |
CN101833281A (en) | Control method for saving energy of aeration in sewage treatment | |
CN105955350B (en) | The fractional order predictive functional control algorithm of genetic algorithm optimization furnace temp | |
CN117375097A (en) | Photovoltaic coordination autonomous method based on multi-agent coordination control strategy and reinforcement learning | |
CN115586801B (en) | Gas blending concentration control method based on improved fuzzy neural network PID | |
CN115327890B (en) | Method for optimizing main steam pressure of PID control thermal power depth peak shaving unit by improved crowd searching algorithm | |
CN107180261A (en) | Based on the Greenhouse grape medium- and long-term forecasting method for rolling BP neural network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |