CN102968055A - Fuzzy PID (Proportion Integration Differentiation) controller based on genetic algorithm and control method thereof - Google Patents

Fuzzy PID (Proportion Integration Differentiation) controller based on genetic algorithm and control method thereof Download PDF

Info

Publication number
CN102968055A
CN102968055A CN2012105210990A CN201210521099A CN102968055A CN 102968055 A CN102968055 A CN 102968055A CN 2012105210990 A CN2012105210990 A CN 2012105210990A CN 201210521099 A CN201210521099 A CN 201210521099A CN 102968055 A CN102968055 A CN 102968055A
Authority
CN
China
Prior art keywords
fuzzy
control
genetic algorithm
controller
controlled
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.)
Pending
Application number
CN2012105210990A
Other languages
Chinese (zh)
Inventor
王海军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Dianji University
Original Assignee
Shanghai Dianji University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Dianji University filed Critical Shanghai Dianji University
Priority to CN2012105210990A priority Critical patent/CN102968055A/en
Publication of CN102968055A publication Critical patent/CN102968055A/en
Pending legal-status Critical Current

Links

Abstract

The invention discloses a fuzzy PID (Proportion Integration Differentiation) controller based on genetic algorithm and a control method thereof. The control method comprises the following steps in sequence: controlling a controlled object by a fuzzy control algorithm; dynamically generating three parameters of the PID controller in an online manner; selecting a proper fitness function by virtue of the genetic algorithm; searching based on the fitness function of each individual in the population; increasing the speed of convergence of the genetic algorithm; calculating to obtain the optimal solution; and further optimizing three basic parameters output in the fuzzy control. By adopting the control method, dynamic and static performances of a control system can be improved while the accuracy in control detection is ensured.

Description

Fuzzy controller and control method thereof based on genetic algorithm
Technical field
The present invention particularly relates to a kind of fuzzy controller based on genetic algorithm and control method thereof about a kind of PID controller and control method thereof.
Background technology
PID (proportional-integral-differential) controller is as existing more than the 70 year history of practical the earliest controller, remain now most widely used industrial control unit (ICU), the PID controller is easily understood, do not need the condition precedents such as accurate system model in the use, thereby become the controller that is most widely used.
Most popular in the at present practical application is conventional PID control strategy, its principle is simple, good stability, but because ratio, integration, differential three basic parameter in the conventional PID control, just can not change again in case adjust, therefore can not well control strong nonlinearity or the stronger system of unknown disturbances.Because fuzzy control can effectively be controlled the nonlinear system of complexity, therefore the fuzzy controller based on fuzzy control arises at the historic moment.
Yet, in general Fuzzy control system, the rule of fuzzy control is to derive from the expert of those control professional domains and technician's professional knowledge basically, has sizable artificial subjectivity, for actual control system those complex nonlinears and that stronger random perturbation is arranged, these control laws can not say entirely accurate, and sometimes even larger discrepancy may occur, and choosing of its subordinate function also has similar above-mentioned situation.
Summary of the invention
For overcoming above-mentioned the deficiencies in the prior art, the present invention's purpose is to provide a kind of fuzzy controller based on genetic algorithm and control method thereof, it utilizes FUZZY ALGORITHMS FOR CONTROL that traditional PID control is improved, adjust online three main control parameters in the PID control system by FUZZY ALGORITHMS FOR CONTROL, can adapt with real-time kinetic-control system, guaranteeing that control detects accurately on the basis, has improved the dynamic and stalic state performance of control system.
For reaching above-mentioned and other purpose, the present invention proposes a kind of fuzzy controller based on genetic algorithm, comprise the PID controller, the fuzzy control module, wherein this fuzzy controller also comprises the genetic algorithm optimization module, the FUZZY ALGORITHMS FOR CONTROL of utilizing this fuzzy control module realizes controlled device is controlled, online three parameters that dynamically generate this PID controller, this genetic algorithm optimization module is selected suitable fitness function in Optimizing Search, utilize in the population each individual fitness value to search for, increase the speed of convergence of genetic algorithm, calculate optimum solution, further optimize with these three parameters that this fuzzy control module is exported.
Further, this fuzzy control module compares exact value and the setting value of controlled volume, obtain error, and calculate error rate, then error is quantized respectively to blur into fuzzy quantity with error rate, again by error, error rate and fuzzy relation matrix by inference composition rule carry out fuzzy decision, obtain fuzzy control quantity, act on controlled device, so circulation is gone down, and realizes the control to controlled device.
Further, this genetic algorithm optimization module utilizes genetic algorithms use colony mode purpose-function space to be carried out the parallel search of multi thread.
Further, this fuzzy controller is used for the linear motor control system based on the parcel post logistics transmission.
Further, this objective function is:
J = ∫ 0 ∞ ( ω 1 | e ( t ) | + ω 2 u 2 ( t ) ) dt + ω 3 * t u
Wherein, e (t) is systematic error, and u (t) is the output of PID controller, t uBe the rise time, w 1, w 2, w 3Be weights.
Further, this fitness function is the inverse of this objective function, F=1/J.
Further, three parameters after this genetic algorithm optimization module optimization are:
Kp=Kpc+ΔKp*Pkp;
Ki=Kic+ΔKi*Pki;
Kd=Kdc+ΔKd*Pkd;
Wherein, Kpc, Kic, Kdc are the constant that the three basic parameter of this PID controller is preset first, and Δ Kp, Δ Ki, Δ Kd are the output of this fuzzy control module, Pkp, and Pki, Pkd is for being calculated the scale factor with respect to three parameters by genetic algorithm.
For reaching above-mentioned and other purpose, the present invention also provides a kind of fuzzy PID control method based on genetic algorithm, comprises the steps
Step 1 utilizes FUZZY ALGORITHMS FOR CONTROL that controlled device is controlled, online three parameters that dynamically generate the PID controller; And
Step 2, utilize genetic algorithm to select suitable fitness function, utilize in the population each individual fitness value to search for, increase the speed of convergence of genetic algorithm, calculate optimum solution, further optimize with this three basic parameter that fuzzy control is exported.
Further, in step 1, this FUZZY ALGORITHMS FOR CONTROL compares exact value and the setting value of controlled volume, obtains error, and calculates error rate, then error is quantized respectively to blur into fuzzy quantity with error rate, again by error, error rate and fuzzy relation matrix by inference composition rule carry out fuzzy decision, obtain fuzzy control quantity, act on controlled device, so circulation is gone down, and realizes the control to controlled device.
Further, the method is used for the linear motor control system based on the parcel post logistics transmission.
Compared with prior art, a kind of fuzzy controller and control method thereof based on genetic algorithm of the present invention passed through in traditional PID control system, utilize the fuzzy control module to the three basic control parameter (ratio of PID controller, integration and differentiation) carries out online dynamically adjustment, the dynamic change of adaptive control control system, in addition, for overcoming the artificial subjectivity of fuzzy control rule in the fuzzy control, the present invention also utilizes genetic algorithm that three parameters exporting in the fuzzy controller are carried out again suboptimization, make fuzzy control rule be more suitable for improving the dynamic and stalic state performance of control system in dynamic objective control object system.
Description of drawings
Fig. 1 is the system architecture diagram of a kind of fuzzy controller based on genetic algorithm of the present invention;
Fig. 2 is the system architecture diagram of the preferred embodiment of a kind of fuzzy controller based on genetic algorithm of the present invention;
Fig. 3 is the synoptic diagram that the subordinate function after quantizing factor changes among the present invention changes;
Fig. 4 is the evolution trajectory diagram of adaptive value function in the simulation process of preferred embodiment of the present invention;
Fig. 5 is step response curve comparison diagram in the simulation process of preferred embodiment of the present invention;
Fig. 6 is the flow chart of steps of a kind of fuzzy PID control method based on genetic algorithm of the present invention.
Embodiment
Below by specific instantiation and accompanying drawings embodiments of the present invention, those skilled in the art can understand other advantage of the present invention and effect easily by content disclosed in the present specification.The present invention also can be implemented or be used by other different instantiation, and the every details in this instructions also can be based on different viewpoints and application, carries out various modifications and change under the spirit of the present invention not deviating from.
Fig. 1 is the system architecture diagram of a kind of fuzzy controller based on genetic algorithm of the present invention.As shown in Figure 1, a kind of fuzzy controller based on genetic algorithm of a kind of the present invention of the present invention comprises at least: PID controller 10, fuzzy control module 11 and genetic algorithm optimization module 12.
Wherein, fuzzy control module 11 utilizes online three parameters that dynamically generate PID controller 10 of FUZZY ALGORITHMS FOR CONTROL, FUZZY ALGORITHMS FOR CONTROL compares exact value and the setting value of controlled volume, obtain error E, and calculate error rate EC, then E and EC are quantized respectively to blur into fuzzy quantity e, ec, again by e, ec and fuzzy relation matrix by inference composition rule carry out fuzzy decision, obtain fuzzy control quantity u, act on controlled device, so circulation is gone down, and realizes the control to controlled device; Genetic algorithm optimization module 12 selects suitable fitness function to utilize in the population each individual fitness value to search in Optimizing Search, increase the speed of convergence of genetic algorithm, calculate optimum solution, further optimize with this three basic parameter that fuzzy control module 11 is exported, make whole control system reach optimum control.
Fig. 2 is the system architecture diagram of the preferred embodiment of a kind of fuzzy controller based on genetic algorithm of the present invention, below will cooperate Fig. 2 to further specify the present invention.
Fuzzy controller commonly used belongs to the tactic pattern of the single output of dual input more in the at present industrial process control, as shown in Figure 3, wherein input, output variable is respectively E (error), EC (error rate) and U (control output), FUZZY ALGORITHMS FOR CONTROL compares exact value and the setting value of controlled volume, obtain error E, and calculate error rate EC, then E and EC are quantized respectively to blur into fuzzy quantity e, ec, again by e, ec and fuzzy relation matrix by inference composition rule carry out fuzzy decision, obtain fuzzy control quantity u, at last this fuzzy control quantity ambiguity solution is become accurately amount U, act on controlled device, so circulation is gone down, and realizes the control to controlled device.
The self-adjusting of quantized factor and proportional factor is that fuzzy control is applied to the most effective means in the in real time control, but controller ONLINE RECOGNITION control effect can be carried out from adjusting quantification or scale parameter according to rise time, overshoot, steady-state error and oscillation and divergence degree etc.
Fig. 3 is the synoptic diagram that the subordinate function after quantizing factor changes among the present invention changes, as seen, the adjustment of quantizing factor has produced impact for the rule output of fuzzy controller, the self-adjusting that also is quantizing factor makes fuzzy control rule and subordinate function that variation occur, then under same error and its rate of change, change has occured in the value of output U, makes by the dynamic and static characteristic of control object variation has occured.In like manner, in preferred embodiment of the present invention, directly change the value of scale factor, variation has occured in the value of then exporting U, the variation of the quantizing factor of mentioning above being equivalent to.Can use certain optimization method, rule and subordinate function according to certain system performance index is optimized then can greatly reduce artificial subjectivity, greatly improve the dynamic and static performance of control system.Therefore the present invention adopts genetic algorithm to come the Comparative Examples factor to carry out optimizing.
In fuzzy control, after fuzzy rule and subordinate function were determined, because the limited amount of control law, then the output valve of fuzzy control also was limited.And when utilizing the genetic algorithm Comparative Examples factor to be optimized adjustment, the value of output will no longer be limited in original quantitative range, and to some degree, the regular quantity that is equivalent to fuzzy control increases, and this is irrealizable in the fuzzy control of routine.
In preferred embodiment of the present invention, the genetic algorithms use colony mode that genetic algorithm optimization module 12 utilizes is carried out the parallel search of multi thread to purpose-function space, can not be absorbed in local minimum point, only need to find the solution the value of objective function, and do not need other information, and continuity, the differentiability of objective function do not required, easy to use, therefore the selection of separating and generation probabilistic manner have strong adaptive faculty and robustness.
In preferred embodiment of the present invention, simulation object is based on the linear motor control system of parcel post logistics transmission, because linear electric motors have saved the intermediate buffering links such as chain, gear, therefore has stronger non-linear.
Do not considering that in the situation that motor is turned, the present invention calculates its stress model suc as formula (1) according to its characteristics in conjunction with stressing conditions (comprising thrust, various friction force) analysis meter in practice.
(M+Δm)a c=F 11[(M+Δm)g+KF 1]
-(μ 21)exp(-av)[(M+Δm)g+KF 1]
3[M+Δm+(KF 1)/g] (1)
In following formula: first is linear motor pushing force (F1), and second is force of sliding friction, and the 3rd is stiction, and the 4th is sticking friction force.In the above Normal Force impact of all having considered motor in every, COEFFICIENT K and F1 in namely every.The electric current loop effect of considering linear electric motors is interior and the relevant parameters of system measured, can be with following order transfer function as its mathematical model:
G ( s ) = K ( Ms + d ) ( Ls + r ) - - - ( 2 )
Wherein: M is taken as 5.2 (kg) for delivery mail quality, and d is that friction factor is taken as 0.8, and L is taken as 0.0165 (H), and r is taken as 1.7 (Ω), and K is decided to be 60.
1, parameter is selected:
In preferred embodiment of the present invention, preset first certain constant for the three basic parameter of PID controller, by normal conditions, the span that limits three parameters is: Kpc=[0~20], Kic=[0~1], Kdc=[0~1], the parameter preset value is:
Kpc=5.8,
Kpi=0.0896,
Kpd=0.0451,
This group parameter proves through many experiments, when being applied to above-mentioned formula (1), its PID control performance has been better than the control performance of Fuzzy-PID, that is to say in this case, the rule of fuzzy control does not embody its superior performance (being limited under the fixed condition of empirical model), can not react well objective control status, but after scale factor is optimized, the change that is optimized of fuzzy inference rule and subordinate function, reduced wherein irrational artificial subjectivity, can press close to preferably the control target property of objective reality, last simulation results show its effective and feasible property.
The output of fuzzy control relates to the increment of Kp, Ki, Kd, i.e. Δ Kp, Δ Ki, Δ Kd, and the domain that defines respectively on the Fuzzy collection according to the variation range of each parameter is:
E、EC={-3,-2,-1,0,1,2,3}
Kp、Ki、Kd={-3,-2,-1,0,1,2,3}
Calculated with respect to Kp by genetic algorithm, Ki, the scale factor of Kd is: Pkp, Pki, Pkd.The three basic parameter of so last PID control is corrected for:
Kp=Kpc+ΔKp*Pkp
Ki=Kic+ΔKi*Pki
Kd=Kdc+ΔKd*Pkd
2, the Code And Decode of Optimal Parameters
The number of parameters that needs to optimize in genetic algorithm is 3, be respectively Pkp, Pki, Pkd, it is the scale factor of corresponding ratio, integration, three parameters of differential respectively, and the span of three parameters is: Pkp=[0.5,5], the span of Pki, Pkd all limit into: [0.5,1.5],, because in PID control, Kp is larger on the rise time in the control procedure, overshoot value, the impact of stabilization time, so that the span of its corresponding scale factor is chosen is larger.The coding method of adopting is the real coding method, and real coding has the simple and Fast Convergent equation advantage of coding.
3, adaptive value function and each other parameter are chosen
Genetic algorithm is not utilized external information substantially in Optimizing Search, only take the adaptive value function as foundation, utilize in the population each individual fitness value to search for.Therefore can choosing of fitness function directly have influence on the speed of convergence of genetic algorithm and find optimum solution, and the present invention adopts following objective function:
J = ∫ 0 ∞ ( ω 1 | e ( t ) | + ω 2 u 2 ( t ) ) dt + ω 3 * t u - - - ( 3 )
In the formula: e (t) is systematic error, and u (t) is the output of controller, and tu is the rise time, w1, and w2, w3 are weights.Excessive in order to prevent controlling energy, in objective function, added the quadratic term of control inputs.
Because genetic manipulation carries out according to the adaptive value size, and adaptive value is non-negative, and the direction that the optimization direction of objective function should increase corresponding to adaptive value, thus the inverse of selecting Fcn as fitness function, that is: F=1/J.
When the operation genetic algorithm program, get the big or small n=60 of population, evolutionary generation is 200, crossover probability Pc=0.9, variation probability P m=0.008, w1=0.999, w2=0.001, w3=2.0.
For preventing that the genetic algorithm optimizing is tending towards locally optimal solution, adopt to keep the optimum individual method, the optimum individual of parent (being made as A) is awarded reservation, when with the optimum individual (being made as B) of filial generation when comparing, if A inferior to B, then replaces A with B, if A is better than B, then still keep A.Fact proved that this way has good effect.
Numerical value through resulting three parameters behind the genetic optimization is respectively:
Pkp=-3.5737
Pkd=1.3421
Pki=0.5627
In preferred embodiment of the present invention, simulated program adopts the Matlab programming, genetic algorithm, fuzzy reasoning and pid control algorithm is combined debug.Fig. 4 is the evolution trajectory diagram of adaptive value function in the simulation process of preferred embodiment of the present invention, Fig. 5 is step response curve comparison diagram in the simulation process of preferred embodiment of the present invention, simulation result shows through the fuzzy controller behind the genetic algorithm optimization can greatly reduce artificial subjectivity than traditional fuzzy-PID control device, greatly improves the dynamic and static performance of control system.
Fig. 6 is the flow chart of steps of a kind of fuzzy PID control method based on genetic algorithm of the present invention.As shown in Figure 6, a kind of fuzzy PID control method based on genetic algorithm of the present invention comprises the steps:
Step 601 utilizes FUZZY ALGORITHMS FOR CONTROL that controlled device is controlled, online three parameters that dynamically generate the PID controller; And
Step 602, utilize genetic algorithm to select suitable fitness function, utilize in the population each individual fitness value to search for, increase the speed of convergence of genetic algorithm, calculate optimum solution, further optimize with this three basic parameter that fuzzy control is exported, make whole control system reach optimum control.
In step 601, FUZZY ALGORITHMS FOR CONTROL compares exact value and the setting value of controlled volume, obtain error E, and calculate error rate EC, then E and EC are quantized respectively to blur into fuzzy quantity e, ec, again by e, ec and fuzzy relation matrix by inference composition rule carry out fuzzy decision, obtain fuzzy control quantity u, act on controlled device, so circulation is gone down, and realizes the control to controlled device.
In step 602, genetic algorithms use colony mode is carried out the parallel search of multi thread to purpose-function space, can not be absorbed in local minimum point, only need to find the solution the value of objective function, and do not need other information, and continuity, the differentiability of objective function do not required, easy to use, therefore the selection of separating and generation probabilistic manner have strong adaptive faculty and robustness.
Equally, also be based on the linear motor control system of parcel post logistics transmission at this simulation object, do not considering to calculate its stress model suc as formula (1) according to its characteristics in conjunction with stressing conditions (comprising thrust, various friction force) analysis meter in practice in the situation that motor is turned.
(M+Δm)a c=F 11[(M+Δm)g+KF 1]
-(μ 21)exp(-av)[(M+Δm)g+KF 1]
3[M+Δm+(KF 1)/g] (1)
In following formula: first is linear motor pushing force (F1), and second is force of sliding friction, and the 3rd is stiction, and the 4th is sticking friction force.In the above Normal Force impact of all having considered motor in every, COEFFICIENT K and F 1 in namely every.The electric current loop effect of considering linear electric motors is interior and the relevant parameters of system measured, can be with following order transfer function as its mathematical model:
G ( s ) = K ( Ms + d ) ( Ls + r ) - - - ( 2 )
Wherein: M is taken as 5.2 (kg) for delivery mail quality, and d is that friction factor is taken as 0.8, and L is taken as 0.0165 (H), and r is taken as 1.7 (Ω), and K is decided to be 60.
In this control program, preset first certain constant for the three basic parameter of PID controller, by normal conditions, the span that limits three parameters is: Kpc=[0~20], Kic=[0~1], Kdc=[0~1], the parameter preset value is:
Kpc=5.8,
Kpi=0.0896,
Kpd=0.0451,
This group parameter proves through many experiments, when being applied to above-mentioned formula (1), its PID control performance has been better than the control performance of Fuzzy-PID, that is to say in this case, the rule of fuzzy control does not embody its superior performance (being limited under the fixed condition of empirical model), can not react well objective control status, but after scale factor is optimized, the change that is optimized of fuzzy inference rule and subordinate function, reduced wherein irrational artificial subjectivity, can press close to preferably the control target property of objective reality, last simulation results show its effective and feasible property.
The output of fuzzy control relates to the increment of Kp, Ki, Kd, i.e. Δ Kp, Δ Ki, Δ Kd, and the domain that defines respectively on the Fuzzy collection according to the variation range of each parameter is:
E、EC={-3,-2,-1,0,1,2,3}
Kp、Ki、Kd={-3,-2,-1,0,1,2,3}
Calculated with respect to Kp by genetic algorithm, Ki, the scale factor of Kd is: Pkp, Pki, Pkd.The three basic parameter of so last PID control is corrected for:
Kp=Kpc+ΔKp*Pkp
Ki=Kic+ΔKi*Pki
Kd=Kdc+ΔKd*Pkd
The number of parameters that needs to optimize in genetic algorithm is 3, be respectively Pkp, Pki, Pkd, it is the scale factor of corresponding ratio, integration, three parameters of differential respectively, and the span of three parameters is: Pkp=[0.5,5], the span of Pki, Pkd all limit into: [0.5,1.5],, because in PID control, Kp is larger on the rise time in the control procedure, overshoot value, the impact of stabilization time, so that the span of its corresponding scale factor is chosen is larger.The coding method of adopting is the real coding method, and real coding has the simple and Fast Convergent equation advantage of coding.
Genetic algorithm is not utilized external information substantially in Optimizing Search, only take the adaptive value function as foundation, utilize in the population each individual fitness value to search for.Therefore can choosing of fitness function directly have influence on the speed of convergence of genetic algorithm and find optimum solution, adopts objective function here:
J = ∫ 0 ∞ ( ω 1 | e ( t ) | + ω 2 u 2 ( t ) ) dt + ω 3 * t u - - - ( 3 )
In the formula: e (t) is systematic error, and u (t) is the output of controller, and tu is the rise time, w1, and w2, w3 are weights.Excessive in order to prevent controlling energy, in objective function, added the quadratic term of control inputs.
Because genetic manipulation is according to carrying out just when size, and just when being non-negative, and the optimization direction of objective function should be corresponding to the direction just when increase, thus the inverse of selecting Fcn as fitness function, that is: F=1/J.
When the operation genetic algorithm program, get the big or small n=60 of population, evolutionary generation is 200, crossover probability Pc=0.9, variation probability P m=0.008, w1=0.999, w2=0.001, w3=2.0.
For preventing that the genetic algorithm optimizing is tending towards locally optimal solution, adopt to keep the optimum individual method, the optimum individual of parent (being made as A) is awarded reservation, when with the optimum individual (being made as B) of filial generation when comparing, if A inferior to B, then replaces A with B, if A is better than B, then still keep A.Fact proved that this way has good effect.
Numerical value through resulting three parameters behind the genetic optimization is respectively:
Pkp=-3.5737
Pkd=1.3421
Pki=0.5627
In sum, a kind of fuzzy controller and control method thereof based on genetic algorithm of the present invention passed through in traditional PID control system, utilize the fuzzy control module to the three basic control parameter (ratio of PID controller, integration and differentiation) carries out online dynamically adjustment, the dynamic change of adaptive control control system, in addition, for overcoming the artificial subjectivity of fuzzy control rule in the fuzzy control, the present invention also utilizes genetic algorithm that three parameters exporting in the fuzzy controller are carried out again suboptimization, make fuzzy control rule be more suitable for improving the dynamic and stalic state performance of control system in dynamic objective control object system.
Above-described embodiment is illustrative principle of the present invention and effect thereof only, but not is used for restriction the present invention.Any those skilled in the art all can be under spirit of the present invention and category, and above-described embodiment is modified and changed.Therefore, the scope of the present invention should be listed such as claims.

Claims (10)

1. fuzzy controller based on genetic algorithm, comprise the PID controller, the fuzzy control module, it is characterized in that: this fuzzy controller also comprises the genetic algorithm optimization module, the FUZZY ALGORITHMS FOR CONTROL of utilizing this fuzzy control module realizes controlled device is controlled, online three parameters that dynamically generate this PID controller, this genetic algorithm optimization module is selected suitable fitness function in Optimizing Search, utilize in the population each individual fitness value to search for, increase the speed of convergence of genetic algorithm, calculate optimum solution, further optimize with these three parameters that this fuzzy control module is exported.
2. the fuzzy controller based on genetic algorithm as claimed in claim 1, it is characterized in that: this fuzzy control module compares exact value and the setting value of controlled volume, obtain error, and calculate error rate, then error is quantized respectively to blur into fuzzy quantity with error rate, again by error, error rate and fuzzy relation matrix by inference composition rule carry out fuzzy decision, obtain fuzzy control quantity, act on controlled device, so circulation is gone down, and realizes the control to controlled device.
3. the fuzzy controller based on genetic algorithm as claimed in claim 1, it is characterized in that: this genetic algorithm optimization module utilizes genetic algorithms use colony mode purpose-function space to be carried out the parallel search of multi thread.
4. the fuzzy controller based on genetic algorithm as claimed in claim 3 is characterized in that: this fuzzy controller is used for the linear motor control system based on the parcel post logistics transmission.
5. the fuzzy controller based on genetic algorithm as claimed in claim 4, it is characterized in that: this objective function is:
J = ∫ 0 ∞ ( ω 1 | e ( t ) | + ω 2 u 2 ( t ) ) dt + ω 3 * t u
Wherein, e (t) is systematic error, and u (t) is the output of PID controller, t uBe the rise time, w 1, w 2, w 3Be weights.
6. the fuzzy controller based on genetic algorithm as claimed in claim 5 is characterized in that: this fitness function is the inverse of this objective function, F=1/J.
7. the fuzzy controller based on genetic algorithm as claimed in claim 6 is characterized in that, three parameters after this genetic algorithm optimization module optimization are:
Kp=Kpc+ΔKp*Pkp;
Ki=Kic+ΔKi*Pki;
Kd=Kdc+ΔKd*Pkd;
Wherein, Kpc, Kic, Kdc are the constant that the three basic parameter of this PID controller is preset first, and Δ Kp, Δ Ki, Δ Kd are the output of this fuzzy control module, Pkp, and Pki, Pkd is for being calculated the scale factor with respect to three parameters by genetic algorithm.
8. the fuzzy PID control method based on genetic algorithm comprises the steps
Step 1 utilizes FUZZY ALGORITHMS FOR CONTROL that controlled device is controlled, online three parameters that dynamically generate the PID controller; And
Step 2, utilize genetic algorithm to select suitable fitness function, utilize in the population each individual fitness value to search for, increase the speed of convergence of genetic algorithm, calculate optimum solution, further optimize with this three basic parameter that fuzzy control is exported.
9. a kind of fuzzy PID control method based on genetic algorithm as claimed in claim 8, it is characterized in that: in step 1, this FUZZY ALGORITHMS FOR CONTROL compares exact value and the setting value of controlled volume, obtain error, and calculate error rate, then error is quantized respectively to blur into fuzzy quantity with error rate, again by error, error rate and fuzzy relation matrix by inference composition rule carry out fuzzy decision, obtain fuzzy control quantity, act on controlled device, so circulation is gone down, and realizes the control to controlled device.
10. a kind of fuzzy PID control method based on genetic algorithm as claimed in claim 9 is characterized in that: the method is used for the linear motor control system based on the parcel post logistics transmission.
CN2012105210990A 2012-12-07 2012-12-07 Fuzzy PID (Proportion Integration Differentiation) controller based on genetic algorithm and control method thereof Pending CN102968055A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2012105210990A CN102968055A (en) 2012-12-07 2012-12-07 Fuzzy PID (Proportion Integration Differentiation) controller based on genetic algorithm and control method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2012105210990A CN102968055A (en) 2012-12-07 2012-12-07 Fuzzy PID (Proportion Integration Differentiation) controller based on genetic algorithm and control method thereof

Publications (1)

Publication Number Publication Date
CN102968055A true CN102968055A (en) 2013-03-13

Family

ID=47798246

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2012105210990A Pending CN102968055A (en) 2012-12-07 2012-12-07 Fuzzy PID (Proportion Integration Differentiation) controller based on genetic algorithm and control method thereof

Country Status (1)

Country Link
CN (1) CN102968055A (en)

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103143298A (en) * 2013-03-27 2013-06-12 桂林电子科技大学 Power modulation control method of artificial diamond hydraulic machine optimized on the basis of genetic algorithm
CN103312249A (en) * 2013-07-09 2013-09-18 南昌航空大学 Self-adaptation excitation control method for synchronous generator
CN103592852A (en) * 2013-11-29 2014-02-19 西南交通大学 PID (Proportion Integration Differentiation) controller optimizing design method based on particle swarm membrane algorithm
CN104155877A (en) * 2014-08-19 2014-11-19 江苏科技大学 Brushless DC motor fuzzy control system based on genetic algorithm and control method thereof
CN104834329A (en) * 2015-04-27 2015-08-12 重庆工商职业学院 Method adopting fuzzy control to adjust genetic algorithm so as to optimize parameters and application of method
CN105807607A (en) * 2016-05-11 2016-07-27 杭州电子科技大学 Method for optimizing predictive fuzzy-PID coking furnace temperature control through genetic algorithm
CN106054596A (en) * 2016-06-23 2016-10-26 东南大学 PID (Proportion Integration Differentiation) controller parameter optimization setting method based on improved performance indexes
CN106094843A (en) * 2016-08-02 2016-11-09 哈尔滨工程大学 A kind of adaptive fuzzy submarine navigation device control method using genetic algorithm optimizing
CN108549208A (en) * 2018-03-14 2018-09-18 重庆邮电大学 A kind of quadrotor attitude control method based on factor adaptive fuzzy
CN109213178A (en) * 2018-11-15 2019-01-15 武汉南华工业设备工程股份有限公司 A kind of course heading control method and device
CN109606382A (en) * 2018-12-24 2019-04-12 河南理工大学 Power transmission system for electric vehicle control method
CN112096649A (en) * 2020-08-28 2020-12-18 武汉理工大学 Vehicle-mounted air conditioner fan control method, storage medium and system
CN112526879A (en) * 2020-11-23 2021-03-19 珠海格力电器股份有限公司 Parameter determination method, device, control method, system and medium for temperature control system
CN112968353A (en) * 2021-02-03 2021-06-15 吉林大学 Driving system and driving method for pumping source of ultrashort pulse laser
CN108549208B (en) * 2018-03-14 2021-12-17 重庆邮电大学 Four-rotor aircraft attitude control method based on factor self-adaptive fuzzy PID

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE4215959A1 (en) * 1991-05-15 1992-11-19 Toyoda Automatic Loom Works GAIN FACTOR ADJUSTMENT FOR PID CONTROLLERS
CN101751049A (en) * 2008-11-28 2010-06-23 上海电机学院 Method for adjusting temperature of subway carriage

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE4215959A1 (en) * 1991-05-15 1992-11-19 Toyoda Automatic Loom Works GAIN FACTOR ADJUSTMENT FOR PID CONTROLLERS
CN101751049A (en) * 2008-11-28 2010-06-23 上海电机学院 Method for adjusting temperature of subway carriage

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
白羽等: "基于遗传算法和模糊PID 的励磁控制器", 《PLC技术应用200例》 *
程大方等: "遗传算法优化模糊PID 的喷雾干燥温度控制应用", 《工业控制计算机》 *
訾兴建: "模糊PID 控制的遗传算法优化", 《淮北职业技术学院学报》 *

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103143298A (en) * 2013-03-27 2013-06-12 桂林电子科技大学 Power modulation control method of artificial diamond hydraulic machine optimized on the basis of genetic algorithm
CN103312249A (en) * 2013-07-09 2013-09-18 南昌航空大学 Self-adaptation excitation control method for synchronous generator
CN103592852A (en) * 2013-11-29 2014-02-19 西南交通大学 PID (Proportion Integration Differentiation) controller optimizing design method based on particle swarm membrane algorithm
CN104155877A (en) * 2014-08-19 2014-11-19 江苏科技大学 Brushless DC motor fuzzy control system based on genetic algorithm and control method thereof
CN104834329A (en) * 2015-04-27 2015-08-12 重庆工商职业学院 Method adopting fuzzy control to adjust genetic algorithm so as to optimize parameters and application of method
CN105807607A (en) * 2016-05-11 2016-07-27 杭州电子科技大学 Method for optimizing predictive fuzzy-PID coking furnace temperature control through genetic algorithm
CN106054596A (en) * 2016-06-23 2016-10-26 东南大学 PID (Proportion Integration Differentiation) controller parameter optimization setting method based on improved performance indexes
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
CN106094843A (en) * 2016-08-02 2016-11-09 哈尔滨工程大学 A kind of adaptive fuzzy submarine navigation device control method using genetic algorithm optimizing
CN106094843B (en) * 2016-08-02 2019-04-12 哈尔滨工程大学 A kind of adaptive fuzzy submarine navigation device control method using genetic algorithm optimizing
CN108549208A (en) * 2018-03-14 2018-09-18 重庆邮电大学 A kind of quadrotor attitude control method based on factor adaptive fuzzy
CN108549208B (en) * 2018-03-14 2021-12-17 重庆邮电大学 Four-rotor aircraft attitude control method based on factor self-adaptive fuzzy PID
CN109213178A (en) * 2018-11-15 2019-01-15 武汉南华工业设备工程股份有限公司 A kind of course heading control method and device
CN109606382B (en) * 2018-12-24 2020-04-03 河南理工大学 Control method for power transmission system of electric automobile
CN109606382A (en) * 2018-12-24 2019-04-12 河南理工大学 Power transmission system for electric vehicle control method
CN112096649A (en) * 2020-08-28 2020-12-18 武汉理工大学 Vehicle-mounted air conditioner fan control method, storage medium and system
CN112526879A (en) * 2020-11-23 2021-03-19 珠海格力电器股份有限公司 Parameter determination method, device, control method, system and medium for temperature control system
CN112968353A (en) * 2021-02-03 2021-06-15 吉林大学 Driving system and driving method for pumping source of ultrashort pulse laser

Similar Documents

Publication Publication Date Title
CN102968055A (en) Fuzzy PID (Proportion Integration Differentiation) controller based on genetic algorithm and control method thereof
Tai et al. Review of recent type-2 fuzzy controller applications
Li et al. Offset-free fuzzy model predictive control of a boiler–turbine system based on genetic algorithm
CN104533701A (en) Automatic setting method for control parameters of water turbine speed regulating system
CN104698842B (en) A kind of LPV model nonlinear forecast Control Algorithms based on interior point method
CN104776446A (en) Combustion optimization control method for boiler
CN104154635A (en) Variable air volume room temperature control method based on fuzzy PID and prediction control algorithm
CN103439887B (en) PI attitude conirol method and system optimal lower order system ITAE
CN106054596B (en) It is a kind of that setting method is optimized based on the PID controller parameter for improving performance indicator
CN104898426A (en) Room temperature loop control method based on gradient descent method and generalized prediction control
CN102393645A (en) Control method of high-speed electro-hydraulic proportional governing system
CN104199299A (en) Multivariable limited generalized prediction control method of gas turbine load regulation performance
CN103399492A (en) Quick non-linear predictive control method for voltage of solid oxide fuel cell
Malikopoulos et al. A real-time computational learning model for sequential decision-making problems under uncertainty
CN104500336B (en) A kind of Wind turbines invariable power generalized forecast control method based on Hammerstein Wiener models
CN105334730A (en) Heating furnace oxygen content IGA optimization T-S fuzzy ARX modeling method
CN104808486A (en) Predictive control method and predictive control device for piezoelectric ceramic actuator based on fuzzy TS (Takagi-Sugeno) model
CN106971548A (en) The Short-time Traffic Flow Forecasting Methods of optimizable multi-core adaptive SVMs
CN104076686A (en) Method for controlling aluminum oxide production process dynamic cost
Liang et al. Freeway ramp control based on genetic PI and fuzzy logic
Hu et al. A short-term wind speed forecasting model based on improved QPSO optimizing LSSVM
CN108333947A (en) Dull integral coefficient Predictive function control parameter tuning method based on intelligent optimizing
Chi et al. Comparison of two multi-step ahead forecasting mechanisms for wind speed based on machine learning models
Lu et al. The temperature control of electric furnace based on PID genetic algorithm
CN106444389A (en) Method for optimizing PI control by fuzzy RBF neural network based on system of pyrolysis of waste plastic temperature

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20130313

C05 Deemed withdrawal (patent law before 1993)