CN110320796A - A kind of electric control method based on PID controller, device and equipment - Google Patents
A kind of electric control method based on PID controller, device and equipment Download PDFInfo
- Publication number
- CN110320796A CN110320796A CN201910722233.5A CN201910722233A CN110320796A CN 110320796 A CN110320796 A CN 110320796A CN 201910722233 A CN201910722233 A CN 201910722233A CN 110320796 A CN110320796 A CN 110320796A
- Authority
- CN
- China
- Prior art keywords
- pid controller
- value
- function
- objective function
- dimensional variable
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 230000006870 function Effects 0.000 claims description 131
- 230000006399 behavior Effects 0.000 claims description 47
- 238000004590 computer program Methods 0.000 claims description 10
- 230000001351 cycling effect Effects 0.000 claims description 2
- 238000004064 recycling Methods 0.000 claims description 2
- 238000005457 optimization Methods 0.000 abstract description 13
- 238000004422 calculation algorithm Methods 0.000 description 38
- 238000010586 diagram Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 230000004044 response Effects 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- 230000002068 genetic effect Effects 0.000 description 3
- 210000002569 neuron Anatomy 0.000 description 3
- 239000002245 particle Substances 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000010355 oscillation Effects 0.000 description 2
- 238000004886 process control Methods 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 238000010187 selection method Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005684 electric field Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
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
- G05B11/00—Automatic controllers
- G05B11/01—Automatic controllers electric
- G05B11/36—Automatic controllers electric with provision for obtaining particular characteristics, e.g. proportional, integral, differential
- G05B11/42—Automatic controllers electric with provision for obtaining particular characteristics, e.g. proportional, integral, differential for obtaining a characteristic which is both proportional and time-dependent, e.g. P. I., P. I. D.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a kind of electric control method based on PID controller, device, equipment and computer readable storage mediums, it include: the objective function for constructing PID controller parameter problem of tuning, wherein, the undetermined parameter of the objective function includes N number of one-dimensional variable；After carrying out discretization to N number of one-dimensional variable, according to nitrification enhancement, N number of one-dimensional variable is learnt respectively using N number of agency, determines the target value of N number of one-dimensional variable；According to the target value of N number of one-dimensional variable, determines the optimal value of the objective function, complete the parameter tuning of the PID controller；Using the PID controller after completion parameter tuning, the control object in electric control system is controlled.Method, apparatus, equipment and computer readable storage medium provided by the present invention, improve the control performance of PID controller parameter optimization efficiency, convergence rate and PID controller.
Description
Technical field
The present invention relates to process control technology field, more particularly to a kind of electric control method based on PID controller,
Device, equipment and computer readable storage medium.
Background technique
As the process control technology in electric field achieves biggish development in recent decades.Worldwide
Persons have investigated different control methods, including self adaptive control, artificial neural-network control, fuzzy control etc..Wherein, most base
Sheet, most widely used is single-circuit PID controller.The PID being made of ratio (P), integral (I) and differential (D) unit
Controller architecture is simple, and can keep preferable robustness when service condition variation range is larger.Therefore, how excellent
Ratio, integral and the differential parameter for changing Tuning PID Controller are one of the emphasis of control problem research.
In the prior art, parameter optimization method includes two classifications: traditional adjusting method and intelligence adjust method.Firstly, traditional
Adjusting method includes Ziegler-Nichols algorithm, based on time-domain integration index (Integral Square Time Error
Criterion, ISTE) optimum PID parameter adjust method.Its adjustment process is complex, and is difficult to avoid that oscillation and surpasses greatly
It adjusts, it is more difficult to obtain optimum PID parameter.Therefore, researcher is dedicated to developing the intelligent PID parameter based on various heuritic approaches whole
Determine method.Genetic algorithm (Genetic Algorithm, GA), particle swarm algorithm (Particle Swarm Optimization,
PSO), the artificial intelligence technologys such as Fuzzy Logic Reasoning Algorithm and artificial neural network are subsequently used in the tuning process of pid parameter.These
Technology can enhance the control performance of PID controller efficiently against the disadvantages mentioned above of traditional adjusting method.However, these technologies
There is also respective defects.For example, GA needs first to handle cumbersome cataloged procedure, while GA and PSO all rely on the general of population
It reads, convergence time is longer, and rate of convergence is slower；The method that fuzzy reasoning is difficult to find system completes the choosing of algorithm inherent parameters
It selects；It include the neuron of multilayer in neural network, the initial weight of the number and neuron that how to determine hidden neuron is also very
Difficulty finds specific method.
In summary as can be seen that how to improve PID controller parameter optimization efficiency is current problem to be solved.
Summary of the invention
The object of the present invention is to provide a kind of electric control method based on PID controller, device, equipment and computers
Readable storage medium storing program for executing, to solve, the parameter adjusting method process complexity of PID controller in the prior art, convergence time is longer, receives
Hold back the slower problem of rate.
In order to solve the above technical problems, the present invention provides a kind of electric control method based on PID controller, comprising: structure
Build the objective function of PID controller parameter problem of tuning, wherein the undetermined parameter of the objective function includes N number of one-dimensional variable；
After carrying out discretization to N number of one-dimensional variable, according to nitrification enhancement, N number of one-dimensional is become respectively using N number of agency
Amount is learnt, and determines the target value of N number of one-dimensional variable；According to the target value of N number of one-dimensional variable, the mesh is determined
The optimal value of scalar functions completes the parameter tuning of the PID controller；Using the PID controller after completion parameter tuning, to electricity
Control object in gas control system is controlled.
Preferably, it is described building PID controller parameter problem of tuning objective function, wherein the objective function to
Determining parameter includes that N number of one-dimensional variable includes:
Construct the objective function of PID controller parameter problem of tuning:
Wherein, e (t) is the tracking error of the PID controller；U (t) is the output of the PID controller；tuIt is described
The output signal y (t) of electric control system from the 10% of steady-state value rise to 90% used in the rise time；Ey (t)=y (t)-
Y (t-1) is overshoot penalty term, as ey (t) >=0, ω4=0；As ey (t) < 0, ω4≠ 0 and ω4> > ω1；The mesh
The undetermined parameter of scalar functions includes the first weights omega1, the second weights omega2, third weights omega3And the 4th weights omega4。
Preferably, each agency's the step of learning to each one-dimensional variable, includes:
S1: (i=1,2 ..., N) a agency (i=1,2 ..., N) a one-dimensional variable i-th takes behavior i-th
After choosing current behavior in set, the current solution of the objective function is determined；
S2: according to the current solution of the computation rule of preset reward function and the objective function, the current behavior is determined
Corresponding reward function value；
S3: updating the corresponding value function of the current behavior according to the reward function value, acts on behalf of root so as to described i-th
Next behavior is chosen according to updated value function；
S4: different disturbances are added to all dimensions currently solved；
S5: circulation executes the S1 to the S4, until cycle-index reaches preset times, completes i-th of one-dimensional
The study of variable.
Preferably, the current solution of the computation rule according to preset reward function and the objective function, determine described in
The corresponding reward function value of current behavior includes:
According toDetermine the reward function value R of the current behavior kth step of i-th of agencyk；Its
In, JkFor the current solution of the objective function；JbestFor the initial optimal solution of the objective function.
Preferably, described to include: according to the corresponding value function of the reward function value update current behavior
According to Vk+1(i, j)=(1- α) Vk(i,j)+α[Rk+(1-λ2)Lmax(i,j)+λ2Lmin(i, j)] to the current line
It is updated for corresponding value function；
Wherein, Vk(i, j) is the corresponding value function；Ll(i, j) is path values, and l=1 indicates path to the left, l=2
Indicate path to the right；λ1For the value function VkThe weight of (i, j)；α is learning rate；Lmax(i, j) and Lmin(i, j) difference
For maximum and the smallest two path values；λ2For the maximum weight with the smallest path values, (1- λ2) > λ2。
The present invention also provides a kind of electrical control gears based on PID controller, comprising:
Module is constructed, for constructing the objective function of PID controller parameter problem of tuning, wherein the objective function
Undetermined parameter includes N number of one-dimensional variable；
Intensified learning module, after carrying out discretization to N number of one-dimensional variable, according to nitrification enhancement, using N
A agency respectively learns N number of one-dimensional variable, determines the target value of N number of one-dimensional variable；
It adjusts module and determines the optimal value of the objective function for the target value according to N number of one-dimensional variable, it is complete
At the parameter tuning of the PID controller；
Electric control module, for utilizing the PID controller after completing parameter tuning, to the control in electric control system
Object is controlled.
Preferably, the building module is specifically used for:
Construct the objective function of PID controller parameter problem of tuning:
Wherein, e (t) is the tracking error of the PID controller；U (t) is the output of the PID controller；tuIt is described
The output signal y (t) of electric control system from the 10% of steady-state value rise to 90% used in the rise time；Ey (t)=y (t)-
Y (t-1) is overshoot penalty term, as ey (t) >=0, ω4=0；As ey (t) < 0, ω4≠ 0 and ω4> > ω1；The mesh
The undetermined parameter of scalar functions includes the first weights omega1, the second weights omega2, third weights omega3And the 4th weights omega4。
Preferably, the intensified learning module includes:
Selection unit is used in a agency of i-th (i=1,2 ..., N) in i-th (i=1,2 ..., N) a one-dimensional variable
It can take after choosing current behavior in behavior set, determine the current solution of the objective function；
Determination unit, for determining institute according to the computation rule of preset reward function and the current solution of the objective function
State the corresponding reward function value of current behavior；
Updating unit, for updating the corresponding value function of the current behavior according to the reward function value, so as to described
I-th of agency chooses next behavior according to updated value function；
Unit is disturbed, for different disturbances to be added to all dimensions currently solved；
Cycling element, for recycling described in execution in a agency of i-th (i=1,2 ..., N) in i-th (i=1,2 ..., N)
A one-dimensional variable take in behavior set choose current behavior after, determine the current solution of the objective function；According to default
The current solution of the computation rule of reward function and the objective function determines the corresponding reward function value of the current behavior；Root
The corresponding value function of the current behavior is updated according to the reward function value, so that described i-th agency is according to updated value
Function chooses next behavior；The step of different disturbances are added to all dimensions currently solved, until cycle-index reaches
Preset times complete the study of i-th of one-dimensional variable.
The present invention also provides an electrical control equipments based on PID controller, comprising:
Memory, for storing computer program；Processor realizes above-mentioned one kind when for executing the computer program
The step of electric control method based on PID controller.
The present invention also provides a kind of computer readable storage medium, meter is stored on the computer readable storage medium
Calculation machine program, the computer program realize a kind of above-mentioned electric control method based on PID controller when being executed by processor
The step of.
Electric control method provided by the present invention based on PID controller, using nitrification enhancement to PID controller
After parameter is adjusted, the control object in electric control system is controlled using the PID controller after completion parameter tuning
System.The electric control system is made of PID controller with by control electrical system.Wherein, using nitrification enhancement to PID
Controller carry out parameter tuning when, first to N number of one-dimensional variable in the objective function of the parameter tuning problem of PID controller into
Row discretization.Then, according to nitrification enhancement, using N number of agency respectively to N number of one-dimensional variable after discretization
It practises, the target value of N number of unit variance is determined, so that it is determined that the optimal value of the objective function, completes the PID controller
Parameter tuning.Method provided by the present invention is adjusted, not based on parameter of the nitrification enhancement to PID controller
Dependent on population, but the thought of " trial and error repeatedly " is used, it is whole with the interaction completion parameter of circumstances not known by acting on behalf of
Fixed, when circumstances not known variation is by control system dynamic time-varying, nitrification enhancement can be joined with on-line optimization PID controller
Number carries out tracing control to system.The present invention improves PID controller parameter optimization efficiency, convergence rate and PID controller
Control performance, be easy to implement simultaneously, have practicability；And nitrification enhancement has certain randomness, can jump out office
Portion is optimal.
Detailed description of the invention
It, below will be to embodiment or existing for the clearer technical solution for illustrating the embodiment of the present invention or the prior art
Attached drawing needed in technical description is briefly described, it should be apparent that, the accompanying drawings in the following description is only this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the stream of the first specific embodiment of the electric control method provided by the present invention based on PID controller
Cheng Tu；
Fig. 2 is the structural schematic diagram of electric control system；
Fig. 3 is the step response comparison diagram of GA algorithm, PSO algorithm system corresponding with three kinds of algorithms of RL algorithm；
Fig. 4 is the average target function convergence result that GA algorithm, PSO algorithm and three kinds of algorithms of RL algorithm are separately optimized 10 times
Comparison diagram；
Fig. 5 is that each agency controls the method flow diagram that variable is learnt to each one-dimensional；
Fig. 6 is a kind of structural block diagram of the electrical control gear based on PID controller provided in an embodiment of the present invention.
Specific embodiment
Core of the invention is to provide a kind of electric control method based on PID controller, device, equipment and computer
Readable storage medium storing program for executing improves the control performance of PID controller parameter optimization efficiency, convergence rate and PID controller.
In order to enable those skilled in the art to better understand the solution of the present invention, with reference to the accompanying drawings and detailed description
The present invention is described in further detail.Obviously, described embodiments are only a part of the embodiments of the present invention, rather than
Whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, shall fall within the protection scope of the present invention.
Referring to FIG. 1, Fig. 1 is specific for the first of the electric control method provided by the present invention based on PID controller
The flow chart of embodiment；Specific steps are as follows:
Step S101: building PID controller parameter problem of tuning objective function, wherein the objective function it is undetermined
Parameter includes N number of one-dimensional variable；
The electric control system being made of a PID controller and a control object is as shown in Fig. 2, wherein C (s) is institute
State the transmission function of PID controller, G (s) is the transmission function of the control object, the input of entire electric control system and defeated
Respectively r (t) and y (t) out, reference of the input signal of the electric control system as control object output signal, they
Between difference be the PID controller tracking error e (t), the output u (t) of the PID controller is the defeated of control object
Enter.In control process, input signal is given, i.e., reference signal and control object, the PID controller can be by tracking
Error is handled, and the output of control object is made to approach input signal, and specific processing method is passed by following Laplce
Delivery function indicates:
Wherein, Kp、KiAnd KdRatio, integral and differential parameter respectively undetermined.
Ambient condition is by target function value quantization means.The objective function of the PID controller parameter problem of tuning is expressed
Formula is as follows:
Wherein, e (t) is the tracking error of the PID controller；U (t) is the output of the PID controller；tuIt is described
The output signal y (t) of electric control system from the 10% of steady-state value rise to 90% used in the rise time；In order to avoid super greatly
Item is provided with an overshoot penalty term in objective function；Ey (t)=y (t)-y (t-1) is overshoot penalty term, when ey (t) >=0
When, ω4=0；As ey (t) < 0, ω4≠ 0 and ω4> > ω1；The undetermined parameter of the objective function includes the first weight
ω1, the second weights omega2, third weights omega3And the 4th weights omega4。
Step S102: after carrying out discretization to N number of one-dimensional variable, according to nitrification enhancement, using N number of agency point
It is other that N number of one-dimensional variable is learnt, determine the target value of N number of one-dimensional variable；
Assuming that the dimension of undetermined parameter X is N, it is represented by X=[x1,x2,…,xN].Nitrification enhancement uses N number of generation
Reason, each to act on behalf of the optimization for being responsible for an one-dimensional variable, N number of agency carries out a study step to respective one-dimensional variable in turn
Suddenly.D is turned to by the feasible zone of i-th of one-dimensional of variable is discretei(i=1,2 ..., N) a grid, the behavior that i-th of agency can take
Collection is combined into Ai=1,2 ..., Di}。
Step S103: it according to the target value of N number of one-dimensional variable, determines the optimal value of the objective function, completes institute
State the parameter tuning of PID controller；
Control object in the electric control system is set asUsing based on nitrification enhancement
PID controller controlled, input signal be a unit step signal.
PSO algorithm and GA algorithm are chosen in the present embodiment and the pid parameter in the present embodiment based on nitrification enhancement is whole
The method of determining compares, the PID controller parameter in three kinds of algorithm optimization same systems.The parameter setting of nitrification enhancement
Are as follows: n=4, λ1=0.5, λ2=0.25, α=1, D=10.The parameter setting of PSO algorithm are as follows: accelerator coefficient c1=c2=2, population
Scale is 100.The parameter setting of GA algorithm are as follows: intersect and mutation rate is respectively 0.9 and 0.01, population scale 100.Target letter
The weighted value setting that number includes are as follows: ω1=0.999, ω2=0.001, ω3=2, ω4=100.
Fig. 3 illustrates the step response of the corresponding system of three kinds of algorithms, and RL algorithm is that the intensified learning in the present embodiment is calculated
Method.At 0 moment, input signal sports 1 from 0, can be seen that from result, and three kinds of algorithms can be eliminated in system response, output
Signal reaches the oscillation and overshoot before steady-state value 1.And the performance of three kinds of algorithms is close, and step is completed in 0.1 second and is rung
It answers.The system response corresponding with PSO of RL algorithm is almost overlapped, and the corresponding system response of GA algorithm is slightly rapid in ascent stage,
But it is slightly slower than other two kinds of algorithms to enter stationary value.Fig. 4 illustrates the average target function receipts that three kinds of algorithms are separately optimized 10 times
Hold back result.PSO and RL algorithmic statement is to target function value more smaller than GA algorithm, but RL convergence speed of the algorithm ratio PSO is promoted
One times.
Step S104: using the PID controller after completion parameter tuning, the control object in electric control system is carried out
Control.
Method provided by the present embodiment is carried out for the parameter tuning problem of PID controller using nitrification enhancement
Parameter optimization.Nitrification enhancement can be avoided the introducing in genetic algorithm and particle swarm algorithm to population, and be the introduction of generation
Reason optimizes objective function, therefore improves the convergence rate in optimization process.Meanwhile nitrification enhancement has centainly
Randomness, local optimum can be jumped out；It is easy to implement, there is practicability.
Step S102 in based on the above embodiment, provide in the present embodiment it is each agency to each one-dimensional variable into
The step of row study.Referring to FIG. 5, Fig. 5 is that each agency controls the method flow diagram that variable is learnt, tool to each one-dimensional
Body Optimization Steps include:
S501: (i=1,2 ..., N) a agency (i=1,2 ..., N) a one-dimensional variable i-th takes row i-th
After choosing current behavior in set, the current solution of the objective function is determined；
Ambient condition is by target function value quantization means.
S502: according to the current solution of the computation rule of preset reward function and the objective function, the current line is determined
For corresponding reward function value；
Environmental feedback gives agency one reward function, and to characterize whether agency takes advantageous behavior, environment is turned
Become better state.According toDetermine the reward function of the current behavior kth step of i-th of agency
Value Rk；Wherein, JkFor the current solution of the objective function；JbestFor the initial optimal solution of the objective function.
S503: the corresponding value function of the current behavior is updated according to the reward function value, so as to described i-th agency
Next behavior is chosen according to updated value function；
The corresponding value function of j-th of behavior of i-th of agency is V (i, j).Agency updates according to reward function and path values
The corresponding value function of the behavior currently taken.Path values refer to i-th of agency in i-th of dimension of variable, from current j-th
Grid selection continues the value of leftward or rightward route searching, is expressed as Ll(i, j), l=1 indicate path to the left, l=2
Indicate path to the right.Path values to the left and to the right are corresponding by the n grid for closing on left and right side of j-th of grid
Value function calculates, and calculation method is as follows:
Wherein,For will close on n value function descending arrangement after m-th of element, λ1For the weight of value function, and it is full
Foot
To sum up, the update rule of the value function is as follows:
Vk+1(i, j)=(1- α) Vk(i,j)+α[Rk+(1-λ2)Lmax(i,j)+λ2Lmin(i,j)]
Wherein, Vk(i, j) is the corresponding value function；Ll(i, j) is path values, and l=1 indicates path to the left, l=2
Indicate path to the right；λ1For the value function VkThe weight of (i, j)；α is learning rate, to characterize new information [Rk+(1-
λ2)Lmax(i,j)+λ2Lmin(i, j)] influence to value function；Lmax(i, j) and Lmin(i, j) is respectively maximum and two the smallest
Path values；λ2For the maximum weight with the smallest path values, (1- λ2) > λ2。
Agency selects next behavior according to updated value function.Before this, agency needs to select a path first,
Selection method is as follows:
Wherein, τkFor temperature, value range is 0≤τk≤1.Work as τkNumerical value is larger, what remaining non-most advantageous behavior was selected
Probability is close；Work as τkFor numerical value close to 0, the probability that these behaviors are selected can be different according to the difference of value function size.τkNumber
Value is gradually reduced with study number, it may be assumed that
Then, it acts on behalf of on selected path, from j-th of grid, a row is selected in the n grid closed on
For selection method is as follows:
And the numerical value of next one-dimensional variable determines at random from selected grid.
S504: different disturbances are added to all dimensions currently solved；
In order to increase the diversity of solution, and also to which algorithm is avoided to fall into local optimum, algorithm is acted on behalf of in n-th and is completed
After one learning procedure, different disturbances is added to all dimensions currently solved, specific practice is as follows:
X ← X+ Δ, Δ=[Δ1,Δ2,…,ΔN]
Wherein, disturbance quantity Δ is generated according to covariance evolution algorithm.
S505: circulation executes the S501 to the S504, until cycle-index reaches preset times, completes described i-th
The study of a one-dimensional variable.
S504 described in repeating said steps S501, after i-th of agency completes a learning process, counter k
Add 1.When k reaches preset threshold value kmaxWhen, algorithm terminates.
A kind of PID controller parameter setting method based on nitrification enhancement is provided in the present embodiment, this method is not
Dependent on population, but the thought of " trial and error repeatedly " is used, it is whole with the interaction completion parameter of circumstances not known by acting on behalf of
It is fixed, when circumstances not known variation i.e. by control system dynamic time-varying when, intensified learning algorithm on-line optimization pid parameter, to being
System carries out tracing control.
Referring to FIG. 6, Fig. 6 is a kind of knot of the electrical control gear based on PID controller provided in an embodiment of the present invention
Structure block diagram；Specific device may include:
Module 100 is constructed, for constructing the objective function of PID controller parameter problem of tuning, wherein the objective function
Undetermined parameter include N number of one-dimensional variable；
Intensified learning module 200, according to nitrification enhancement, is adopted after carrying out discretization to N number of one-dimensional variable
N number of one-dimensional variable is learnt respectively with N number of agency, determines the target value of N number of one-dimensional variable；
It adjusts module 300 and determines the optimal value of the objective function for the target value according to N number of one-dimensional variable,
Complete the parameter tuning of the PID controller；
Electric control module 400, for utilizing the PID controller after completing parameter tuning, in electric control system
Control object is controlled.
The present embodiment based on the electrical control gear of PID controller for realizing the electricity above-mentioned based on PID controller
Pneumatic control method, thus specific embodiment in the electrical control gear based on PID controller it is visible hereinbefore based on PID
The embodiment part of the electric control method of controller, for example, building module 100, intensified learning module 200 adjust module
300, electric control module 400 is respectively used to realize step S101 in the above-mentioned electric control method based on PID controller,
S102, S103 and S104, so, specific embodiment is referred to the description of corresponding various pieces embodiment, herein not
It repeats again.
The specific embodiment of the invention additionally provides a kind of electrical control equipment based on PID controller, comprising: memory,
For storing computer program；Processor is realized above-mentioned a kind of based on PID controller when for executing the computer program
The step of electric control method.
The specific embodiment of the invention additionally provides a kind of computer readable storage medium, the computer readable storage medium
On be stored with computer program, the computer program realizes a kind of above-mentioned electricity based on PID controller when being executed by processor
The step of pneumatic control method.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.For being filled disclosed in embodiment
For setting, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method part
Explanation.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond the scope of this invention.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
It above can to the electric control method based on PID controller, device, equipment and computer provided by the present invention
Storage medium is read to be described in detail.Specific case used herein explains the principle of the present invention and embodiment
It states, the above description of the embodiment is only used to help understand the method for the present invention and its core ideas.It should be pointed out that for this skill
For the those of ordinary skill in art field, without departing from the principle of the present invention, several change can also be carried out to the present invention
Into and modification, these improvements and modifications also fall within the scope of protection of the claims of the present invention.
Claims (10)
1. a kind of electric control method based on PID controller characterized by comprising
Construct the objective function of PID controller parameter problem of tuning, wherein the undetermined parameter of the objective function includes N number of list
Tie up variable；
After carrying out discretization to N number of one-dimensional variable, according to nitrification enhancement, using N number of agency respectively to N number of list
Dimension variable is learnt, and determines the target value of N number of one-dimensional variable；
According to the target value of N number of one-dimensional variable, determines the optimal value of the objective function, complete the PID controller
Parameter tuning；
Using the PID controller after completion parameter tuning, the control object in electric control system is controlled.
2. the method as described in claim 1, which is characterized in that the target letter of the building PID controller parameter problem of tuning
Number, wherein the undetermined parameter of the objective function includes that N number of one-dimensional variable includes:
Construct the objective function of PID controller parameter problem of tuning:
Wherein, e (t) is the tracking error of the PID controller；U (t) is the output of the PID controller；tuIt is described electrical
The output signal y (t) of control system from the 10% of steady-state value rise to 90% used in the rise time；Ey (t)=y (t)-y (t-
It 1) is overshoot penalty term, as ey (t) >=0, ω4=0；As ey (t) < 0, ω4≠ 0 and ω4> > ω1；The target letter
Several undetermined parameters includes the first weights omega1, the second weights omega2, third weights omega3And the 4th weights omega4。
3. method according to claim 2, which is characterized in that the step of each agency learns each one-dimensional variable packet
It includes:
S1: (i=1,2 ..., N) a agency (i=1,2 ..., N) a one-dimensional variable i-th takes behavior set i-th
After middle selection current behavior, the current solution of the objective function is determined；
S2: according to the current solution of the computation rule of preset reward function and the objective function, determine that the current behavior is corresponding
Reward function value；
S3: the corresponding value function of the current behavior is updated according to the reward function value, so that i-th of agency is according to more
Value function after new chooses next behavior；
S4: different disturbances are added to all dimensions currently solved；
S5: circulation executes the S1 to the S4, until cycle-index reaches preset times, completes i-th of one-dimensional variable
Study.
4. method as claimed in claim 3, which is characterized in that the computation rule according to preset reward function and the mesh
The current solution of scalar functions determines that the corresponding reward function value of the current behavior includes:
According toDetermine the reward function value R of the current behavior kth step of i-th of agencyk；Wherein, Jk
For the current solution of the objective function；JbestFor the initial optimal solution of the objective function.
5. method as claimed in claim 4, which is characterized in that described to update the current behavior according to the reward function value
Corresponding value function includes:
According to Vk+1(i, j)=(1- α) Vk(i,j)+α[Rk+(1-λ2)Lmax(i,j)+λ2Lmin(i, j)] to the current behavior pair
The value function answered is updated；
Wherein, Vk(i, j) is the corresponding value function；Ll(i, j) is path values, and l=1 indicates path to the left, and l=2 is indicated
Path to the right；λ1For the value function VkThe weight of (i, j)；α is learning rate；Lmax(i, j) and Lmin(i, j) is respectively most
Greatly with the smallest two path values；λ2For the maximum weight with the smallest path values, (1- λ2) > λ2。
6. a kind of electrical control gear based on PID controller characterized by comprising
Construct module, for constructing the objective function of PID controller parameter problem of tuning, wherein the objective function it is undetermined
Parameter includes N number of one-dimensional variable；
Intensified learning module, after carrying out discretization to N number of one-dimensional variable, according to nitrification enhancement, using N number of generation
Reason respectively learns N number of one-dimensional variable, determines the target value of N number of one-dimensional variable；
Module is adjusted, for the target value according to N number of one-dimensional variable, the optimal value of the objective function is determined, completes institute
State the parameter tuning of PID controller；
Electric control module, for utilizing the PID controller after completing parameter tuning, to the control object in electric control system
It is controlled.
7. device as claimed in claim 6, which is characterized in that the building module is specifically used for:
Construct the objective function of PID controller parameter problem of tuning:
Wherein, e (t) is the tracking error of the PID controller；U (t) is the output of the PID controller；tuIt is described electrical
The output signal y (t) of control system from the 10% of steady-state value rise to 90% used in the rise time；Ey (t)=y (t)-y (t-
It 1) is overshoot penalty term, as ey (t) >=0, ω4=0；As ey (t) < 0, ω4≠ 0 and ω4> > ω1；The target letter
Several undetermined parameters includes the first weights omega1, the second weights omega2, third weights omega3And the 4th weights omega4。
8. device as claimed in claim 7, which is characterized in that the intensified learning module includes:
Selection unit, for being adopted in a agency of i-th (i=1,2 ..., N) in i-th (i=1,2 ..., N) a one-dimensional variable
It takes after choosing current behavior in behavior set, determines the current solution of the objective function；
Determination unit, for the current solution according to the computation rule of preset reward function and the objective function, determine described in work as
It moves ahead as corresponding reward function value；
Updating unit, for updating the corresponding value function of the current behavior according to the reward function value, so as to described i-th
Agency chooses next behavior according to updated value function；
Unit is disturbed, for different disturbances to be added to all dimensions currently solved；
Cycling element, for recycling described in execution in a agency of i-th (i=1,2 ..., N) in i-th (i=1,2 ..., N) a list
Dimension variable take in behavior set choose current behavior after, determine the current solution of the objective function；According to preset reward
The current solution of the computation rule of function and the objective function determines the corresponding reward function value of the current behavior；According to institute
It states reward function value and updates the corresponding value function of the current behavior, so that described i-th agency is according to updated value function
Choose next behavior；The step of different disturbances are added to all dimensions currently solved, until cycle-index reaches default
Number completes the study of i-th of one-dimensional variable.
9. a kind of electrical control equipment based on PID controller characterized by comprising
Memory, for storing computer program；
Processor is realized a kind of based on PID control as described in any one of claim 1 to 5 when for executing the computer program
The step of electric control method of device processed.
10. a kind of computer readable storage medium, which is characterized in that be stored with computer on the computer readable storage medium
Program is realized a kind of based on PID control as described in any one of claim 1 to 5 when the computer program is executed by processor
The step of electric control method of device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910722233.5A CN110320796A (en) | 2019-08-06 | 2019-08-06 | A kind of electric control method based on PID controller, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910722233.5A CN110320796A (en) | 2019-08-06 | 2019-08-06 | A kind of electric control method based on PID controller, device and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110320796A true CN110320796A (en) | 2019-10-11 |
Family
ID=68125626
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910722233.5A Pending CN110320796A (en) | 2019-08-06 | 2019-08-06 | A kind of electric control method based on PID controller, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110320796A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118011783A (en) * | 2024-04-09 | 2024-05-10 | 天津仁爱学院 | Construction environment PID control method based on improved barrel jellyfish algorithm |
| CN118011783B (en) * | 2024-04-09 | 2024-06-04 | 天津仁爱学院 | Construction environment PID control method based on improved barrel jellyfish algorithm |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104200096A (en) * | 2014-08-29 | 2014-12-10 | 中国南方电网有限责任公司超高压输电公司昆明局 | Lightning arrester grading ring optimization method based on differential evolutionary algorithm and BP neural network |
| CN105911868A (en) * | 2016-06-15 | 2016-08-31 | 南京工业大学 | Multi-batch intermittent reactor two-dimension iterative learning feedback control method |
| CN106896716A (en) * | 2017-04-17 | 2017-06-27 | 华北电力大学（保定） | Micro-capacitance sensor alternating current-direct current section transverter pid parameter optimization method based on grey wolf algorithm |
| EP3357651A2 (en) * | 2017-02-06 | 2018-08-08 | Seiko Epson Corporation | Control device, robot, and robot system |
-
2019
- 2019-08-06 CN CN201910722233.5A patent/CN110320796A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104200096A (en) * | 2014-08-29 | 2014-12-10 | 中国南方电网有限责任公司超高压输电公司昆明局 | Lightning arrester grading ring optimization method based on differential evolutionary algorithm and BP neural network |
| CN105911868A (en) * | 2016-06-15 | 2016-08-31 | 南京工业大学 | Multi-batch intermittent reactor two-dimension iterative learning feedback control method |
| EP3357651A2 (en) * | 2017-02-06 | 2018-08-08 | Seiko Epson Corporation | Control device, robot, and robot system |
| CN106896716A (en) * | 2017-04-17 | 2017-06-27 | 华北电力大学（保定） | Micro-capacitance sensor alternating current-direct current section transverter pid parameter optimization method based on grey wolf algorithm |
Non-Patent Citations (1)
| Title |
|---|
| X. Y. SHANG ET. AL.: "Parameter Optimization of PID Controllers by Reinforcement Learning", 《5TH COMPUTER SCIENCE AND ELECTRONIC ENGINEERING CONFERENCE (CEEC)》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118011783A (en) * | 2024-04-09 | 2024-05-10 | 天津仁爱学院 | Construction environment PID control method based on improved barrel jellyfish algorithm |
| CN118011783B (en) * | 2024-04-09 | 2024-06-04 | 天津仁爱学院 | Construction environment PID control method based on improved barrel jellyfish algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Khooban et al. | A new intelligent online fuzzy tuning approach for multi-area load frequency control: Self Adaptive Modified Bat Algorithm | |
| Valdez et al. | Comparative study of the use of fuzzy logic in improving particle swarm optimization variants for mathematical functions using co-evolution | |
| Mohamed et al. | Optimal power flow using moth swarm algorithm | |
| Panda et al. | Comparison of particle swarm optimization and genetic algorithm for FACTS-based controller design | |
| Shabanpour-Haghighi et al. | A modified teaching–learning based optimization for multi-objective optimal power flow problem | |
| Modares et al. | Parameter estimation of bilinear systems based on an adaptive particle swarm optimization | |
| Ghoshal et al. | Application of GA based optimal integral gains in fuzzy based active power-frequency control of non-reheat and reheat thermal generating systems | |
| Ghoshal | Application of GA/GA-SA based fuzzy automatic generation control of a multi-area thermal generating system | |
| Coban | A fuzzy controller design for nuclear research reactors using the particle swarm optimization algorithm | |
| Lamamra et al. | A new technique with improved control quality of nonlinear systems using an optimized fuzzy logic controller | |
| CN106779177A (en) | Multiresolution wavelet neutral net electricity demand forecasting method based on particle group optimizing | |
| Surmann et al. | Learning feed-forward and recurrent fuzzy systems: A genetic approach | |
| CN104050505A (en) | Multilayer-perceptron training method based on bee colony algorithm with learning factor | |
| Sharma et al. | A review on scale factor strategies in differential evolution algorithm | |
| Serra et al. | Multiobjective evolution based fuzzy PI controller design for nonlinear systems | |
| Chadi et al. | Understanding Reinforcement Learning Algorithms: The Progress from Basic Q-learning to Proximal Policy Optimization | |
| CN113919217B (en) | Adaptive parameter setting method and device for active disturbance rejection controller | |
| Narendra et al. | Improving the speed of response of learning algorithms using multiple models | |
| CN110320796A (en) | A kind of electric control method based on PID controller, device and equipment | |
| Lu et al. | Multivariable self-organizing fuzzy logic control using dynamic performance index and linguistic compensators | |
| Ikemoto et al. | Continuous deep Q-learning with a simulator for stabilization of uncertain discrete-time systems | |
| CN108089442B (en) | PI controller parameter self-tuning method based on prediction function control and fuzzy control | |
| Goswami et al. | Performance Analysis of the AVR Using An Artificial Neural Network and Genetic Algorithm Optimization Technique | |
| Chopra et al. | Neural network tuned fuzzy controller for MIMO system | |
| Fathi et al. | Artificial Intelligence and Evolutionary Algorithms-Based Optimization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191011 |
|
| WD01 | Invention patent application deemed withdrawn after publication |