CN112994047A - Adaptive fractional order PID load frequency control method based on two-dimensional cloud model - Google Patents
Adaptive fractional order PID load frequency control method based on two-dimensional cloud model Download PDFInfo
- Publication number
- CN112994047A CN112994047A CN202110441824.2A CN202110441824A CN112994047A CN 112994047 A CN112994047 A CN 112994047A CN 202110441824 A CN202110441824 A CN 202110441824A CN 112994047 A CN112994047 A CN 112994047A
- Authority
- CN
- China
- Prior art keywords
- cloud
- fractional order
- deviation
- order pid
- cloud model
- 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 37
- 230000003044 adaptive effect Effects 0.000 title claims description 18
- 230000008859 change Effects 0.000 claims description 23
- 238000012937 correction Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 10
- 238000013507 mapping Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 4
- 238000013459 approach Methods 0.000 claims description 2
- 230000004069 differentiation Effects 0.000 claims 1
- 230000002349 favourable effect Effects 0.000 abstract 1
- 230000000694 effects Effects 0.000 description 13
- 230000004044 response Effects 0.000 description 11
- 230000006870 function Effects 0.000 description 5
- 238000004088 simulation Methods 0.000 description 5
- 238000007599 discharging Methods 0.000 description 4
- 230000001105 regulatory effect Effects 0.000 description 4
- 238000012546 transfer Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000004146 energy storage Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000013016 damping Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/24—Arrangements for preventing or reducing oscillations of power in networks
- H02J3/241—The oscillation concerning frequency
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
Abstract
The invention discloses a self-adaptive fractional order PID load frequency control method based on a two-dimensional cloud model, which is used for carrying out parameter self-tuning on a traditional fractional order PID controller by adopting a two-dimensional cloud model parameter tuning method to obtain a two-dimensional cloud model self-adaptive fractional order PID controller; the two-dimensional cloud model self-adaptive fractional order PID controller mainly comprises a two-dimensional cloud processor and a fractional order PID controller, converts input quantity into adjusting parameters of the fractional order PID controller through the steps of clouding, cloud rule base query, cloud reasoning, inverse clouding and the like, and is mainly applied to the conditions that load disturbance and parameter perturbation occur to a multi-region interconnected power system containing an electric automobile power exchanging station, and the system frequency fluctuates and has large deviation. The method can eliminate the frequency deviation of the system by virtue of good control performance, effectively inhibit frequency fluctuation, quickly recover the system frequency to a rated value, and is favorable for maintaining normal and stable operation of a power system and improving the power quality.
Description
Technical Field
The invention relates to the technical field of intelligent power grid frequency control, in particular to a two-dimensional cloud model-based adaptive fractional order PID load frequency control method.
Background
The power grid frequency is an important index for measuring the quality of electric energy, and maintaining the stability of the power grid frequency is a necessary condition for ensuring the safety and stability of a power system. Power systems are required to maintain system frequency within an allowable range in order to meet performance requirements for safety and stability. The huge power demand brought by the development of information technology puts higher requirements on stable power supply, and meanwhile, the scale of a modern power system is continuously enlarged, and the wide application of a new energy power supply in a power grid also becomes a possible factor causing the frequency of the power grid to be more unstable.
The off-line setting method is generally suitable for scenes with low real-time requirements, and is generally practical in occasions requiring batch processing and high time delay, one time period is manually specified to be a period for off-line setting, and the controller parameters are re-set in each period. The setting mode is generally complicated, usually, in order to reduce the setting task amount, the period setting is long, most controllers are controlled by parameters which are manually set in advance, the parameters of the controllers cannot be adjusted in real time according to the real-time change of the frequency in the control process, and the control effect of the controllers is usually greatly reduced when the power system is complicated and changeable; the relative online setting method, also called real-time setting, can collect the frequency deviation in the power system at every moment, and adjust the parameters of the controller in real time through real-time calculation, and has the advantages of low time delay and high performance.
The cloud model controller combines the cloud model module and the fractional order PID controller together, the cloud model realizes the conversion between the qualitative concept and the quantitative representation by using the linguistic variable, compared with the traditional fuzzy control, the cloud model adds the randomness concept in the parameter setting process, and randomly represents the qualitative concept by using the normally distributed random number, so that the cloud model controller has a better control effect on the random frequency fluctuation in the power system, and is more suitable for the frequently occurring small fluctuation and small condition in the power system. Meanwhile, compared with the traditional PID controller, the cloud model controller has two more degrees of freedom in parameter adjustment, can control the controlled object more flexibly and more accurately, and has more excellent control effect after observing the dynamic behavior of the power system. The problems that the uncertainty of the power system causes the difficulty of adjusting the parameters of the controller to rise and the control effect to be poor are the key and difficult points of the research of the invention.
Disclosure of Invention
The invention aims to solve the problem that parameters of a conventional load frequency controller are difficult to set on line, and the method is a self-adaptive fractional order PID load frequency control method based on a two-dimensional cloud model, wherein the method comprises the steps of respectively establishing the two-dimensional cloud model controller and the fractional order PID model, carrying out clouding, cloud rule base query, inference and inverse clouding on input quantity by using the two-dimensional cloud model controller, inputting the output quantity of the two-dimensional cloud model controller into the fractional order PID controller, and respectively setting 5 parameters of the fractional order PID controller; the method is suitable for the conditions of load disturbance and frequency fluctuation in a multi-region interconnected power grid, and simulation experiments prove that the method can effectively improve the control performance of the frequency controller, so that the system frequency is quickly recovered to a normal value, and the robustness of the system is improved.
The technical scheme adopted by the invention for solving the technical problem is as follows:
a self-adaptive fractional order PID load frequency control method based on a two-dimensional cloud model comprises the following steps:
sampling tie line power deviation, regional frequency deviation and frequency bias constant of any region of the interconnected power grid to obtain corresponding regional control deviation, taking the regional control deviation as an error signal E, and taking the output of the regional control deviation after a differential term as a deviation change rate Ec；
Error signal E and deviation rate of change EcRespectively carrying out normalization processing and then taking the normalized values as the input of a two-dimensional cloud model controller, wherein the two-dimensional cloud model controller converts the input numerical values into corresponding state languages;
state speech and bias rate of change E based on error signal EcState language of (1), query cloud pushA rule management table, which is used for acquiring the state language of each adjusting parameter of the fractional order PID controller so as to obtain the expected value of each adjusting parameter; carrying out inverse clouding processing to obtain correction quantities of all the adjusting parameters; obtaining the adjustment quantity of each adjusting parameter based on the expected value of each adjusting parameter and the correction quantity of each adjusting parameter;
mapping the adjustment quantity of each adjustment parameter from the normalized discourse domain to the actual discourse domain to obtain the actual value of the adjustment quantity;
and adjusting the adjusting parameters based on the actual values of the adjusting parameters to realize the real-time control of the load frequency.
Preferably, the tie line power deviation, the area frequency deviation and the frequency offset constant of any area of the interconnected power grid are sampled to obtain the corresponding area control deviation, which specifically comprises the following steps:
ACE＝ΔPtie+B·Δf
where ACE denotes the area control deviation, Δ PtieRepresents the tie line power deviation, Δ f represents the regional frequency deviation, and B represents the frequency offset constant.
Preferably, the status language is divided into seven cloud sets, the cloud sets include "negative large NB", "negative medium NM", "negative small NS", "zero ZE", "positive small PS", "positive medium PM", and "positive large PB", the maximum membership of each cloud set is "1", and each cloud set is the desired ExEntropy EnAnd entropy HeThree numerical characteristics are expressed;
in the two-dimensional cloud model controller, error signal E and deviation change rate E are correctedcSetting value delta K of proportionality coefficientpIntegral coefficient setting value delta KiDifferential coefficient setting value delta KdThe integral term index setting value delta lambda and the differential term index setting value delta mu are subjected to cloud set division; the setting value delta K of the proportionality coefficientpIntegral coefficient setting value delta KiDifferential coefficient setting value delta KdAnd the integral term index setting value delta lambda and the differential term index setting value delta mu are five adjusting parameters of the fractional order PID controller.
Preferably, the association of the seven cloud sets and their corresponding digital signatures is represented as follows:
negative large (-0.6,0.1,0.01), negative medium (-0.4, 0.1,0.01), negative small (-0.2, 0.1,0.01), zero (0, 0.1,0.01), positive small (0.2, 0.1,0.01), medium (0.4, 0.1,0.01) and positive large (0.6, 0.1, 0.01).
Preferably, the cloud reasoning rule table comprises five cloud reasoning rule tables, and the five cloud reasoning rule tables are respectively a proportional coefficient setting value delta KpCloud reasoning rule table and integral coefficient setting value delta KiCloud reasoning rule table and differential coefficient setting value delta KdThe cloud reasoning rule table of the integral term index setting value delta lambda, the cloud reasoning rule table of the differential term index setting value delta mu and the cloud reasoning rule table of the differential term index setting value delta mu are adopted; dividing based on cloud set to obtain different input quantity error signals E and deviation change rate EcAnd respectively converting the words into the words of the corresponding state languages, respectively using the words of the corresponding state languages as the rows and columns of each cloud inference rule table, and finding out the words of the state languages of the output variables corresponding to the words as expected values of each adjusting parameter.
Preferably, the error signal E and the rate of change of deviation E are combinedcAnd respectively carrying out normalization processing and then taking the normalized data as the input of a two-dimensional cloud model controller, wherein the output is a normal random number x which is a cloud droplet in space.
Preferably, the membership degree is calculated as follows:
wherein R represents a degree of membership;
is represented by EnTo be expected, with HeNormal random number E of standard deviationn *(ii) a x represents by ExTo expectation, with En *Is a normal random number of standard deviation.
Preferably, the inverse clouding process is performed to obtain the correction amount of each adjustment parameter, specifically as follows:
where Δ c represents the correction amount.
Preferably, the obtaining the adjustment amount of each adjustment parameter based on the expected value of each adjustment parameter and the correction amount of each adjustment parameter specifically includes:
and respectively carrying out sum operation on the expected value of each adjusting parameter and the correction quantity of each adjusting parameter to obtain the adjustment quantity of each adjusting parameter.
Preferably, the adaptive fractional order PID load frequency control method based on the two-dimensional cloud model further includes:
repeating the steps S1 to S5 until the regional frequency deviation approaches the preset range.
The invention has the following beneficial effects:
1) according to the method, cloud model control and fractional order PID control are combined, and aiming at load frequency disturbance of a multi-region interconnected power system, wave crests can be effectively reduced, overshoot is reduced, adjusting time is shortened, and the safety of the power system is maintained;
2) the cloud set and the cloud reasoning rule concept reflect the uncertainty in the event set, and the possibility of system frequency imbalance when accidental situations occur is reduced;
3) the two-dimensional cloud model is adopted to carry out adaptive parameter adaptation on the fractional order PID controller parameters, the difficulty of parameter setting is reduced, the fractional order PID controller parameters are continuously adjusted to be optimal according to the response condition of an actual system, and the effect of guaranteeing the stability of the system is achieved.
The invention is described in further detail below with reference to the drawings and embodiments, but the adaptive fractional order PID load frequency control network based on the two-dimensional cloud model is not limited to the embodiments.
Drawings
FIG. 1 is a schematic flow chart of a self-adaptive fractional order PID load frequency control method based on a two-dimensional cloud model according to the invention;
FIG. 2 is a model schematic diagram of the adaptive fractional order PID load frequency control method based on the two-dimensional cloud model;
FIG. 3 is a schematic diagram of a cloud collection;
FIG. 4 is a schematic diagram of a model of a two-region interconnected power system
FIG. 5 is a comparison graph of response curves of a cloud model fractional order PID controller and a conventional controller to frequency deviation of a region one under the condition that a load disturbance of 0.02p.u occurs in a region 1 of a 5 second and a load disturbance of-0.01 p.u occurs in a region 2 of a 15 second;
FIG. 6 is a comparison graph of response curves of a cloud model fractional order PID controller and a conventional controller for frequency deviation of a second region under the condition that a load disturbance of 0.02p.u occurs in a 5 second region 1 and a load disturbance of-0.01 p.u occurs in a 15 second region 2;
FIG. 7 is a comparison graph of response curves of a cloud model fractional order PID controller and a conventional controller to power deviation of a tie line under the condition that the load disturbance of 0.02p.u occurs in the 5 th second region 1 and the load disturbance of-0.01 p.u occurs in the 15 th second region 2;
FIG. 8 is a comparison graph of response curves of a cloud model fractional order PID controller and a conventional controller for frequency deviation of region one in the case of region 1 experiencing random load disturbances with amplitude between-0.1 p.u and 0.1 p.u;
FIG. 9 is a comparison graph of response curves of a cloud model fractional order PID controller and a conventional controller for frequency deviation of region two in the case of region 1 experiencing random load disturbances with amplitude between-0.1 p.u and 0.1 p.u;
FIG. 10 is a graph comparing response curves of a cloud model fractional order PID controller and a conventional controller to tie-line power deviation in the case of region 1 with random load disturbances of amplitude between-0.1 p.u and 0.1 p.u;
FIG. 11 is a comparison graph of the response curves of a cloud model fractional order PID controller and a conventional controller for frequency deviation of region one in the case of region 2 experiencing random load disturbances with amplitude between-0.1 p.u and 0.1 p.u;
FIG. 12 is a comparison graph of response curves of a cloud model fractional order PID controller and a conventional controller for frequency deviation of region two in the case of region 2 with random load disturbance having an amplitude between-0.1 p.u and 0.1 p.u;
FIG. 13 is a graph comparing response curves of a cloud model fractional order PID controller and a conventional controller to tie-line power deviation in the case of region 2 with a random load disturbance of amplitude between-0.1 p.u and 0.1 p.u.
Detailed Description
The technical scheme of the invention is specifically explained below with reference to the accompanying drawings.
The invention relates to a self-adaptive fractional order PID load frequency control method based on a two-dimensional cloud model, which comprises the following steps of:
for deviation E and deviation change rate EcSetting value delta K of proportionality coefficientpIntegral coefficient setting value delta KiDifferential coefficient setting value delta KdThe integral term index setting value delta lambda and the differential term index setting value delta mu are clouded and expressed by three characteristic values of a cloud model, a cloud control rule is established, and E to delta K are respectively establishedp、ΔKi、ΔKdΔ λ, Δ μ, and EcTo Δ Kp、ΔKi、ΔKdAnd delta lambda and delta mu are subjected to cloud mapping to carry out cloud reasoning and output inverse clouding, so that the online setting of the fractional order PID controller parameters is realized.
Specifically, referring to fig. 1 and fig. 2, the fractional order PID load frequency control method based on the two-dimensional cloud model in the present embodiment includes the following steps:
step 1: design of designing a two-dimensional cloud model processor
Frequency deviation delta f to an area by a sampling systemiRegional deviation coefficient BiExchange power delta P of tie linetie,i(where i denotes the ith region), calculating the ACE value (denoted E) and the rate of change of deviation (denoted E) by differentiating ACEc) And performing clouding as an input of the two-dimensional cloud model processor. The cloud processing steps are as follows:
step 101: to effect translation of input quantities from numeric values to state language, input quantities E, E are appliedcDividing the image into seven regions from small to large, and marking the seven regions as E1、E2……E7，Ec,1、Ec,2……Ec,7The control rule base of the cloud model is established according to linguistic variables, the linguistic variables are valued into seven sets of 'negative large NB', 'negative middle NM', 'negative small NS', 'zero ZE', 'positive small PS', 'positive middle PM' and 'positive large PB', the maximum membership degree of each linguistic value is '1', and three numerical characteristics (expected E) of the cloud model are usedxEntropy EnEntropy of He) And (4) performing representation.
Step 102: generating as E from the digital characteristics of the imagined cloudnTo be expected, with HeNormal random number E of standard deviationn *Further generated as ExTo expect, En *And a normal random number x which is the standard deviation is a cloud droplet in space.
The deviation E and the deviation change rate E are calculated as followscFor the purpose of illustration, the numerical characteristics of both are shown in table 1. Repeating the operation for N times to obtain a set A of N cloud droplets, wherein A is the cloud model of a certain state language. Entropy E of seven State languagesnAnd entropy HeSame, expect ExThe cloud droplet collection obtained from small to large is-0.6, -0.4, -0.2, 0, 0.2, 0.4, 0.6 respectively, as shown in figure 3.
TABLE 1 three features of the cloud model
To sum up, the deviations E are divided into 7 sets: "negative big NB", "negative medium NM", "negative small NS", "zero ZE", "positive small PS", "positive PM", "positive big PB", expressed by the numerical characteristics of the cloud model: (-0.6,0.1,0.01), (-0.4, 0.1,0.01), (-0.2, 0.1,0.01), (0, 0.1,0.01), (0.2, 0.1,0.01), (0.4, 0.1,0.01), (0.6, 0.1, 0.01);
deviation change rate EcThe division into 7 sets: "negative big NB", "negative medium NM", "negative small NS", "zero ZE", "positive small PS", "positive PM", "positive big PB", expressed by the numerical characteristics of the cloud model: (-0.6,0.1,0.01) (-0.4，0.1，0.01)、(-0.2，0.1，0.01)、(0，0.1，0.01)、(0.2，0.1，0.01)、(0.4，0.1，0.01)、(0.6，0.1，0.01)；
Setting the proportional coefficient to be delta KpThe division into 7 sets: "negative big NB", "negative medium NM", "negative small NS", "zero ZE", "positive small PS", "positive PM", "positive big PB", expressed by the numerical characteristics of the cloud model: (-0.6,0.1,0.01), (-0.4, 0.1,0.01), (-0.2, 0.1,0.01), (0, 0.1,0.01), (0.2, 0.1,0.01), (0.4, 0.1,0.01), (0.6, 0.1, 0.01);
setting the integral coefficient to be delta KiThe division into 7 sets: "negative big NB", "negative medium NM", "negative small NS", "zero ZE", "positive small PS", "positive PM", "positive big PB", expressed by the numerical characteristics of the cloud model: (-0.6,0.1,0.01), (-0.4, 0.1,0.01), (-0.2, 0.1,0.01), (0, 0.1,0.01), (0.2, 0.1,0.01), (0.4, 0.1,0.01), (0.6, 0.1, 0.01);
setting the differential coefficient to be delta KdThe division into 7 sets: "negative big NB", "negative medium NM", "negative small NS", "zero ZE", "positive small PS", "positive PM", "positive big PB", expressed by the numerical characteristics of the cloud model: (-0.6,0.1,0.01), (-0.4, 0.1,0.01), (-0.2, 0.1,0.01), (0, 0.1,0.01), (0.2, 0.1,0.01), (0.4, 0.1,0.01), (0.6, 0.1, 0.01);
dividing the integral term index setting value delta lambda into 7 sets: "negative big NB", "negative medium NM", "negative small NS", "zero ZE", "positive small PS", "positive PM", "positive big PB", expressed by the numerical characteristics of the cloud model: (-0.6,0.1,0.01), (-0.4, 0.1,0.01), (-0.2, 0.1,0.01), (0, 0.1,0.01), (0.2, 0.1,0.01), (0.4, 0.1,0.01), (0.6, 0.1, 0.01);
the differential term index setting value delta mu is divided into 7 sets: "negative big NB", "negative medium NM", "negative small NS", "zero ZE", "positive small PS", "positive PM", "positive big PB", expressed by the numerical characteristics of the cloud model: (-0.6,0.1,0.01), (-0.4, 0.1,0.01), (-0.2, 0.1,0.01), (0, 0.1,0.01), (0.2, 0.1,0.01), (0.4, 0.1,0.01), (0.6, 0.1, 0.01);
step 103: judging which state language the input quantity falls in, and according to En *And x calculates the qualitative concept of the input quantity: degree of membership R is calculated by the formula
And repeating the steps until the membership degrees of the N cloud droplets are calculated. Under the condition that the computing resource allows, the larger the value of N, the better, generally the value is 3000.
Step 2: establishing a two-dimensional cloud processor rule base
And after the cloud is finished, obtaining the adjustment quantity of the fractional order PID parameters according to the established cloud inference rule base. Establishing a correct inference rule base is extremely important for adjusting the control performance of the controller, and inputting the regional control deviation E of the frequency in the power grid system and the differential quantity E thereof into the cloud processorcThe two-dimensional cloud model controller firstly converts the input into the corresponding state language, outputs the state language of the parameter by inquiring the control rule table of the corresponding parameter, converts the state language into the adjustment quantity and completes the conversion from the qualitative mode to the quantitative mode. The fractional order PID controller receives the adjustment amount and modifies the corresponding parameter, thus completing one adjustment.
The cloud inference rule base is established by the following steps:
step 201: the state language is divided into seven cloud sets, and the cloud sets and the corresponding numerical characteristics are minus large (-0.6,0.1,0.01), minus medium (-0.4, 0.1,0.01) minus small (-0.2, 0.1,0.01), zero (0, 0.1,0.01), plus small (0.2, 0.1,0.01), and plus large (0.6, 0.1,0.01) in the middle (0.4, 0.1, 0.01). By regulating quantity delta KpFor example, the area control deviation E and the differential EcIn the rule table corresponding to the regulating variable Δ KpA cloud set of (1), representing EcTo Δ KpA mapping of (2). Likewise make Δ Ki、ΔKdAnd the mapping of delta lambda and delta mu are integrated into corresponding control rules to be stored in a rule base.
Step 202: according to experience, writing out individual conditional statements and setsRule table of synthetic reasoning, different input quantities E, EcAnd converting the vocabulary into corresponding language variable state vocabulary which is respectively used as the row and the column of the rule table, and finding the corresponding output variable state vocabulary.
For example: the state of the frequency deviation E in a certain time zone is "plus or minus", and the change rate E of the deviationcIs 'big' and can be obtained by inquiring a rule table, and the proportional term adjustment quantity delta K of the fractional order controllerpShould be "positive small", integral term adjustment Δ KiThe integral term index adjustment amount Δ λ should be "positive small", the differential term index adjustment amount Δ μ should be "negative small".
Step 203: the fractional order PID controller has five parameters to be set, so that five cloud inference rule tables need to be established in advance.
The inference rule tables established according to the selection principle of the control rule are shown in tables 2 to 6.
TABLE 2. DELTA.KpControl rule table of
TABLE 3. DELTA.KiControl rule table of
TABLE 4 control rule Table for Δ Kd
TABLE 5 control rule Table of Δ λ
TABLE 6 control rule Table of Δ μ
And step 3: and obtaining the adjustment quantity of the final control parameter through the inverse clouding processor.
Step 301: the expected value of the corresponding regulating parameter in the inference rule table is c0The cloud droplet thus output is drop (c)0,R),drop(c0R) is the desired value c0And a set of membership degrees R. And sending the obtained N cloud drops to an inverse clouding processor, carrying out inverse clouding operation on the cloud drops and calculating an average value to obtain the adjustment quantity of the final control parameter.
Step 302: the inverse clouding processor converts the fuzzy quantity of the clouding concept into an accurate quantity of control. Firstly, substituting the membership parameter R in the cloud drop into a formula
The correction quantity delta c is obtained, and the adjustment quantity c is c ═ c0C, integrating the c values of N cloud droplets and taking the average value to obtain
I.e. the adjustment of the final control parameter.
Step 303: adjustment of control parameters obtained under different inference rules
In a different way, will
Replacing with corresponding control parameters to obtain the adjustment delta K of the parameters of the fractional order PID controller processed by the two-dimensional cloud modelp、ΔKi、ΔKd、Δλ、Δμ。
And 4, step 4: design of fractional order PID controller
The fractional order PID transfer function formula is
λ and μ are the weights of the controller integral and differential terms, respectively. The controller has five degrees of freedom which can be set, and compared with other common integral-order PID controllers, the controller has two more degrees of freedom, and the effect of adjusting frequency fluctuation is better.
The fractional order PID controller is arranged behind the two-dimensional cloud processor, receives the adjusting information from the two-dimensional cloud processor, and adjusts the internal parameters of the controller in real time.
Further, in specific application, the fractional order PID load frequency control method based on the two-dimensional cloud model in this embodiment includes:
and 5: error sampling
Given load disturbance in a multi-region interconnected power system simulation model, and controlling the deviation ACE (area control error) to be delta Ptie+ B · Δ f as the error signal E at time t, where Δ PtieFor the tie line power deviation, Δ f is the regional frequency deviation, and B is the frequency offset constant. The output of ACE via differential terms is taken as the rate of change of deviation Ec。
Step 6: error and rate of change normalization
From the actual universe of discourse [ -Emin,Emax]Mapping to normalized discourse field [ -1,1 []To obtain E ', wherein E' ═ E/Emax。
Will EcFrom the actual universe of discourse [ -Ec.min,Ec.max]Mapping to normalized discourse field [ -1,1 []To obtain Ec', wherein Ec’＝Ec/Ec.max。
And 7: obtaining a fractional order PID controller proportionality coefficient setting value delta K based on a two-dimensional cloud model through the steps 1-3pIntegral coefficient setting value delta KiDifferential coefficient setting value delta KdIntegral term index setting value delta lambda and differential term index setting value delta mu.
Step 701: the E 'and Ec' are taken as the input of a cloud model processor based on a two-dimensional cloud model, and the output is the digital characteristic (expecting E)xEntropy EnEntropy of He) The cloud droplet collection of (1).
Step 702: and obtaining a cloud droplet set of 5 fractional order PID controller control parameters after rule base reasoning.
Step 703: and the cloud drop set of the control parameters is used as the input of the inverse clouding processor, and the adjustment quantity of the control parameters is obtained through inverse clouding processing.
Step 704: adjustment Δ KpFrom the normalized universe of discourse [ -1,1]Mapping to the actual universe of discourse [ - Δ Kp.min,ΔKp.max]The adjustment amount Δ K is obtained in the same manneri、ΔKdActual values of Δ λ, Δ μ.
Step 705: and the fractional PID control parameter at the t moment is the fractional PID control parameter at the t-1 moment plus the control parameter adjustment quantity at the t moment. The real-time dynamic calculation formula of the control parameters of the fractional order PID controller is as follows:
wherein, Kp0、Ki0、Kd0、λ0、μ0Respectively, the respective control parameters of the t-1 fractional order PID controller, and Δ Kp、ΔKi、ΔKdAnd delta lambda and delta mu are adjustment quantities of control parameters obtained by calculation and analysis of the cloud model processor at the time t, and are summed to obtain real-time parameters of the fractional order PID controller.
And 8: and (5) repeating the steps 5-7, enabling the frequency deviation delta f to be close to a set value (wherein the set value can be set to be 0, and the close range is +/-0.2 Hz), enabling the control to be free of deviation, and realizing the self-adaptive fractional order PID load frequency control method based on the two-dimensional cloud model.
Simulation experiment
Step 1: establishing load frequency control model of interconnected power grid
Referring to fig. 4, the interconnected power grid is established in two areas, each area comprises a controller, a speed governor, a prime mover and a generator, and the purpose of control is to eliminate frequency deviation Δ f existing in the power grid. The input of the controller is a regional control deviation ACE which is composed of a regional frequency deviation delta f, a regional deviation coefficient B and a junctor exchange power delta PtieBy the formula ACE ═ Δ PtieAnd + B. delta. f.The objective function of the controller is selected as absolute value of error and time integral ITAE, which is expressed by formula
And (4) calculating.
Step 101: establishing a generator model
When the generator works, the prime motor is driven by the bearing to rotate, the mechanical energy of the prime motor is converted into electric energy to be output, and the relation between the torque and the power of the generator is delta Pm-△Pe＝(△Tm-△Te)ω0In which Δ Pe(s) are divided into two categories, one typical active load varying with system frequency, with motor load, using Δ PLAnd (4) showing. Another type of active load is represented by a purely resistive load, which is independent of frequency variations, denoted by Δ PDAnd (4) showing. Delta PDCan be expressed as Δ PDD · Δ ω, where D is referred to as the load damping coefficient. According to Δ PDCan be given a transfer function of
Step 102: establishing a governor model
The speed regulator is a basic component in an electric power system, and mainly plays a role of primary frequency modulation in the system. It can output regulating signal according to the motor speed, power system frequency and load set value, and control the position of prime mover air inlet valve so as to change the prime mover speed and output mechanical power. Where the inertia link is used
And (4) showing.
Step 103: establishing a prime mover model
The prime motor models used in the invention are non-reheat steam turbines. The position of an air inlet valve of the steam turbine is controlled, the air inlet quantity of the valve is changed, the pressure in the cylinder is increased or weakened, and the output mechanical power of the steam turbine is adjusted. Due to the machinery of the steam turbineThe change of inertia and mechanical power is relatively lagged, so that the transfer function model of the steam turbine can be equivalent by using a first-order inertia link, namely the transfer function model can be used
And (4) showing.
Step 104: establishing energy storage model of electric automobile battery replacement station
Along with the wide application of new energy automobiles, the electric automobile can be used as a novel controllable load which can be connected to a network, and can transmit energy to the power grid through the discharge of a storage battery in a stop state, so that the bidirectional circulation of the energy between the electric automobile and the power grid is realized. An energy storage model of an electric automobile power changing station is shown in figure 4, TeIs a time constant, ± mueFor inverter capacity constraints, ± δeFor power increment change rate constraint, the input quantity is the voltage deviation delta U of the micro-gridESetting two constraint links +/-mue、±δeThe method is used for limiting the charging and discharging power of the electric automobile.
The charge-discharge State (SOC) of the electric automobile is related to the percentage of the total energy occupied by the battery energy of the electric automobile, and a scheduling strategy of a single automobile is formulated according to the battery state information of the electric automobile. In the power grid V2G mode, the charging and discharging power of the electric vehicle in a certain period can be represented as:
PE.i(k)＝μepE.i(k)fi(k)
in the formula, PE.i(k) The power after optimization has the characteristic of continuous adjustability; mu.seIs a constraint condition; p is a radical ofE.i(k) The exchange power between the electric automobile and the power grid at the moment k; f. ofi(k) The charging and discharging state of the electric automobile is set; emaxIs the upper limit of capacity; eminThe lower limit of capacity. When the energy of the battery is lower than the lower limit of the capacity, the controlled electric automobile is in a charging state; when the energy of the battery is higher than the upper limit of the capacity, the electric automobile is in a discharging state; at the same timeThe constraint link set in advance ensures that the discharge rate of the electric automobile does not exceed mueThe charging rate is not lower than mue。
Step 2: performing simulation verification
And (3) building a two-area power grid model in Matlab/Simulink, and selecting a fractional order PID controller and an integer solution PID controller based on a two-dimensional cloud model for a controller to perform a comparison experiment. In the two-area interconnected power grid model, disturbance experiments under different conditions are carried out: 1) the load disturbance of 0.02p.u occurs in the area 1 at the time of 5 seconds, and the load disturbance of-0.01 p.u occurs in the area 2 at the time of 15 seconds; 2) region 1 experiences random load disturbances with amplitudes between-0.1 p.u and 0.1 p.u; 3) region 2 experiences random load disturbances with an amplitude between-0.1 p.u and 0.1 p.u.
And the simulation analysis compares the control effects of the adaptive fractional order PID controller based on the two-dimensional cloud model, the integer order PID controller based on the two-dimensional cloud model and the common integer order PID controller. The resulting regional grid dynamic response is shown in fig. 5-13.
Compared with the three conditions, the control effect of the PID controller set by using the two-dimensional cloud model is obviously better than that of the traditional controller under the condition of load disturbance, the speed of the two-dimensional cloud model fractional order PID controller for suppressing disturbance is higher, the amplitude of frequency deviation is smaller, the time required for recovering stability is shorter, and the control effect is better. The two-dimensional cloud model is used for self-adaptively setting the control parameters, the control effect of the controller is improved, and the frequency deviation of the model can be well inhibited.
Experiments and theoretical analysis show that the adaptive fractional order PID load frequency control method based on the two-dimensional cloud model can effectively realize load frequency control of the interconnected power system, and compared with the traditional integer order PID controller and the traditional fractional order PID controller, the fractional order PID controller based on the two-dimensional cloud model has better control effect when participating in secondary frequency control of a power grid, and has more excellent performance in wave crest reduction, overshoot reduction and regulation time shortening.
The above is only one preferred embodiment of the present invention. However, the present invention is not limited to the above embodiments, and any equivalent changes and modifications made according to the present invention, which do not bring out the functional effects beyond the scope of the present invention, belong to the protection scope of the present invention.
Claims (10)
1. A self-adaptive fractional order PID load frequency control method based on a two-dimensional cloud model is characterized by comprising the following steps:
s1, sampling the tie line power deviation, the area frequency deviation and the frequency offset constant of any area of the interconnected network to obtain the corresponding area control deviation, taking the area control deviation as an error signal E, and taking the output of the area control deviation after the differentiation item as a deviation change rate Ec；
S2, the error signal E and the deviation change rate EcRespectively carrying out normalization processing and then taking the normalized values as the input of a two-dimensional cloud model controller, wherein the two-dimensional cloud model controller converts the input numerical values into corresponding state languages;
s3, state language and deviation change rate E based on error signal EcInquiring a cloud inference rule table to obtain the state language of each adjusting parameter of the fractional order PID controller so as to obtain the expected value of each adjusting parameter; carrying out inverse clouding processing to obtain correction quantities of all the adjusting parameters; obtaining the adjustment quantity of each adjusting parameter based on the expected value of each adjusting parameter and the correction quantity of each adjusting parameter;
s4, mapping the adjustment quantity of each adjustment parameter from the normalized domain to the actual domain to obtain the actual value of the adjustment quantity;
and S5, adjusting the adjusting parameters based on the actual values of the adjusting parameters, and realizing the real-time control of the load frequency.
2. The two-dimensional cloud model-based adaptive fractional order PID load frequency control method according to claim 1, wherein the tie line power deviation, the area frequency deviation and the frequency offset constant of any area of the interconnected network are sampled to obtain the corresponding area control deviation, specifically as follows:
ACE＝ΔPtie+B·Δf
where ACE denotes the area control deviation, Δ PtieRepresents the tie line power deviation, Δ f represents the regional frequency deviation, and B represents the frequency offset constant.
3. The adaptive fractional order PID load frequency control method based on the two-dimensional cloud model according to claim 1, wherein the state language is divided into seven cloud sets in total, the cloud sets include "negative big NB", "negative middle NM", "negative small NS", "zero ZE", "positive small PS", "positive middle PM" and "positive big PB", the maximum membership of each cloud set is "1", and each cloud set is used with the desired ExEntropy EnAnd entropy HeThree numerical characteristics are expressed;
in the two-dimensional cloud model controller, error signal E and deviation change rate E are correctedcSetting value delta K of proportionality coefficientpIntegral coefficient setting value delta KiDifferential coefficient setting value delta KdThe integral term index setting value delta lambda and the differential term index setting value delta mu are subjected to cloud set division; the setting value delta K of the proportionality coefficientpIntegral coefficient setting value delta KiDifferential coefficient setting value delta KdAnd the integral term index setting value delta lambda and the differential term index setting value delta mu are five adjusting parameters of the fractional order PID controller.
4. The adaptive fractional order PID load frequency control method based on the two-dimensional cloud model according to claim 3, wherein the association of seven cloud sets and their corresponding digital features is represented as follows:
negative large (-0.6,0.1,0.01), negative medium (-0.4, 0.1,0.01), negative small (-0.2, 0.1,0.01), zero (0, 0.1,0.01), positive small (0.2, 0.1,0.01), medium (0.4, 0.1,0.01) and positive large (0.6, 0.1, 0.01).
5. The adaptive fractional order PID load frequency control method based on the two-dimensional cloud model according to claim 3, wherein the cloud inference rule table comprises five,respectively a proportional coefficient setting value delta KpCloud reasoning rule table and integral coefficient setting value delta KiCloud reasoning rule table and differential coefficient setting value delta KdThe cloud reasoning rule table of the integral term index setting value delta lambda, the cloud reasoning rule table of the differential term index setting value delta mu and the cloud reasoning rule table of the differential term index setting value delta mu are adopted; dividing based on cloud set to obtain different input quantity error signals E and deviation change rate EcAnd respectively converting the words into the words of the corresponding state languages, respectively using the words of the corresponding state languages as the rows and columns of each cloud inference rule table, and finding out the words of the state languages of the output variables corresponding to the words as expected values of each adjusting parameter.
6. The adaptive fractional order PID load frequency control method based on the two-dimensional cloud model according to claim 3, wherein an error signal E and a deviation change rate E are usedcAnd respectively carrying out normalization processing and then taking the normalized data as the input of a two-dimensional cloud model controller, wherein the output is a normal random number x which is a cloud droplet in space.
7. The adaptive fractional order PID load frequency control method based on the two-dimensional cloud model according to claim 6, wherein the membership degree is calculated as follows:
wherein R represents a degree of membership;
is represented by EnTo be expected, with HeNormal random number E of standard deviationn *(ii) a x represents by ExTo expectation, with En *Is a normal random number of standard deviation.
8. The adaptive fractional order PID load frequency control method based on the two-dimensional cloud model according to claim 7, wherein the inverse cloud processing is performed to obtain correction amounts of each adjusting parameter, specifically as follows:
where Δ c represents the correction amount.
9. The adaptive fractional order PID load frequency control method based on the two-dimensional cloud model according to claim 1, wherein the obtaining of the adjustment amount of each regulation parameter based on the expected value of each regulation parameter and the correction amount of each regulation parameter specifically comprises:
and respectively carrying out sum operation on the expected value of each adjusting parameter and the correction quantity of each adjusting parameter to obtain the adjustment quantity of each adjusting parameter.
10. The adaptive fractional order PID load frequency control method based on the two-dimensional cloud model according to claim 1, further comprising:
repeating the steps S1 to S5 until the regional frequency deviation approaches the preset value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110441824.2A CN112994047A (en) | 2021-04-23 | 2021-04-23 | Adaptive fractional order PID load frequency control method based on two-dimensional cloud model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110441824.2A CN112994047A (en) | 2021-04-23 | 2021-04-23 | Adaptive fractional order PID load frequency control method based on two-dimensional cloud model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112994047A true CN112994047A (en) | 2021-06-18 |
Family
ID=76340020
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110441824.2A Pending CN112994047A (en) | 2021-04-23 | 2021-04-23 | Adaptive fractional order PID load frequency control method based on two-dimensional cloud model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112994047A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205569A (en) * | 2015-10-15 | 2015-12-30 | 华侨大学 | Draught fan gear box state on-line evaluation model building method and on-line evaluation method |
| US20170084197A1 (en) * | 2015-09-23 | 2017-03-23 | ValueCorp Pacific, Incorporated | Systems and methods for automatic distillation of concepts from math problems and dynamic construction and testing of math problems from a collection of math concepts |
| CN108306340A (en) * | 2018-02-05 | 2018-07-20 | 河北工业大学 | Interconnected electric power system LOAD FREQUENCY Planar clouds control method containing new energy |
| CN109240082A (en) * | 2018-08-25 | 2019-01-18 | 南京理工大学 | A kind of caterpillar mobile robot sliding formwork cloud model cross-coupling control method |
| CN112350316A (en) * | 2020-11-03 | 2021-02-09 | 河北工业大学 | Variable-discourse-domain cloud PI load frequency control method based on cloud scaling factor |
-
2021
- 2021-04-23 CN CN202110441824.2A patent/CN112994047A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170084197A1 (en) * | 2015-09-23 | 2017-03-23 | ValueCorp Pacific, Incorporated | Systems and methods for automatic distillation of concepts from math problems and dynamic construction and testing of math problems from a collection of math concepts |
| CN105205569A (en) * | 2015-10-15 | 2015-12-30 | 华侨大学 | Draught fan gear box state on-line evaluation model building method and on-line evaluation method |
| CN108306340A (en) * | 2018-02-05 | 2018-07-20 | 河北工业大学 | Interconnected electric power system LOAD FREQUENCY Planar clouds control method containing new energy |
| CN109240082A (en) * | 2018-08-25 | 2019-01-18 | 南京理工大学 | A kind of caterpillar mobile robot sliding formwork cloud model cross-coupling control method |
| CN112350316A (en) * | 2020-11-03 | 2021-02-09 | 河北工业大学 | Variable-discourse-domain cloud PI load frequency control method based on cloud scaling factor |
Non-Patent Citations (2)
| Title |
|---|
| 左剑等: "计及时滞的互联电网负荷频率控制最优分数阶PID控制器设计", 《控制理论与应用》 * |
| 李笑: "基于云理论的含新能源互联电网系统的负荷频率控制", 《万方数据》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Oshnoei et al. | Disturbance observer and tube-based model predictive controlled electric vehicles for frequency regulation of an isolated power grid | |
| CN112186811B (en) | AGC unit dynamic optimization method based on deep reinforcement learning | |
| CN111697597B (en) | Fire storage combined AGC frequency modulation control method based on particle swarm optimization | |
| CN109611274B (en) | LQG (Linear quadratic glass) optimization control method for high wind speed area of wind generating set | |
| CN112928763B (en) | Virtual synchronous generator fuzzy control method based on auxiliary inertia power adjustment | |
| CN113448248A (en) | Intelligent control method for flexibility and deep peak regulation of thermal power generating unit | |
| Arya et al. | Fuzzy logic based load frequency control of multi-area electrical power system considering non-linearities and boiler dynamics | |
| Luan et al. | Load-following control of nuclear reactors based on Takagi-Sugeno fuzzy model | |
| CN112994047A (en) | Adaptive fractional order PID load frequency control method based on two-dimensional cloud model | |
| CN111064228B (en) | Wind turbine generator droop control method and system considering wind speed and load change scene and computer equipment | |
| CN112865137B (en) | Load frequency control method for variable frequency bias coefficient | |
| CN113359468B (en) | Wind turbine generator fault-tolerant control method based on robust self-adaption and sliding mode variable structure control | |
| Yang et al. | Tracking-dispatch of a combined wind-storage system based on model predictive control and two-layer fuzzy control strategy | |
| CN115760178A (en) | Multi-region dynamic economic scheduling method with real-time power balance constraint | |
| CN108278597B (en) | Coal feeding control system of circulating fluidized bed boiler and fuzzy control method thereof | |
| CN113690950A (en) | Electric automobile auxiliary traditional unit combined frequency modulation control method based on stable economic model predictive control | |
| CN110233486B (en) | Energy storage auxiliary fuzzy PID control method based on active prediction SOC | |
| Li et al. | AGC fuzzy PID control strategy based on Parameter Optimization | |
| Lopatin et al. | An adaptive control system of a polymerizer based on fuzzy supervisor | |
| CN114944659B (en) | Automatic control method and system for wind-solar-energy-storage combined power station | |
| Li et al. | Research on pressure and water level control of the pressurizer for marine nuclear power plant based on multivariable MPC | |
| Rao et al. | Load Frequency Control (LFC) of three area interconnected power system using adaptive neuro fuzzy interface system | |
| Yao et al. | Optimized Design of Steam Turbine Control System | |
| Shan et al. | Research on Improved Droop Control Strategy for Primary Frequency Regulation of Doubly-fed Wind Turbine Based on Gray Wolf Algorithm | |
| Cheng et al. | Fuzzy model based control of steam generation in drum-boiler power plant |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210618 |
|
| RJ01 | Rejection of invention patent application after publication |