CN117277422B - Method, system, terminal and medium for evaluating stability of direct-drive wind farm - Google Patents
Method, system, terminal and medium for evaluating stability of direct-drive wind farm Download PDFInfo
- Publication number
- CN117277422B CN117277422B CN202311552495.4A CN202311552495A CN117277422B CN 117277422 B CN117277422 B CN 117277422B CN 202311552495 A CN202311552495 A CN 202311552495A CN 117277422 B CN117277422 B CN 117277422B
- Authority
- CN
- China
- Prior art keywords
- power plant
- wind power
- stability
- sample
- stability margin
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 36
- 238000013145 classification model Methods 0.000 claims abstract description 31
- 238000012706 support-vector machine Methods 0.000 claims abstract description 23
- 238000001514 detection method Methods 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000013097 stability assessment Methods 0.000 claims description 11
- 238000010276 construction Methods 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000006073 displacement reaction Methods 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000010355 oscillation Effects 0.000 abstract description 11
- 238000012482 interaction analysis Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 15
- 238000011156 evaluation Methods 0.000 description 10
- 230000001965 increasing effect Effects 0.000 description 9
- 238000004422 calculation algorithm Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000004088 simulation Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 239000013598 vector Substances 0.000 description 5
- 230000003068 static effect Effects 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000013106 supervised machine learning method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- 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
- H02J3/002—Flicker reduction, e.g. compensation of flicker introduced by non-linear load
-
- 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
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/381—Dispersed generators
-
- 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
- H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
-
- 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]
-
- 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
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/28—The renewable source being wind energy
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Power Engineering (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Nonlinear Science (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to the field of wind power plant state detection, and particularly discloses a method, a system, a terminal and a medium for evaluating stability of a direct-driven wind power plant, which are used for acquiring stable state parameter samples under a plurality of operating conditions, constructing lumped impedance of a wind power plant grid-connected system according to the acquired stable state parameter samples, and constructing a stability margin data set by taking the maximum real part of a lumped impedance determinant zero point as a stability margin index; training the stability margin data set by using a fuzzy support vector machine to obtain a wind power plant state classification model; acquiring a real-time steady state parameter sample of a wind power plant; and inputting the real-time steady state parameter sample into a trained wind power plant state classification model to obtain the current running state of the wind power plant. The method and the device perform grid-connected interaction analysis by using the parameters under a plurality of operation conditions, are suitable for detecting the state of the wind power plant with nonlinear and multi-mode oscillation characteristics, and improve the accuracy of detecting the state of the wind power plant.
Description
Technical Field
The invention relates to the field of wind farm state detection, in particular to a method, a system, a terminal and a medium for evaluating stability of a direct-drive wind farm.
Background
The direct-drive wind power plant is a wind power plant formed by direct-drive fans, and generally comprises a plurality of branches, and each branch is connected with a plurality of direct-drive fans. The whole wind farm is connected into an alternating current power grid through a step-up transformer. Meanwhile, a static reactive power generator is arranged on a collecting bus of the wind power plant and is used for compensating reactive power.
At present, the direct-driven wind turbine generator has the advantages of low cost and high efficiency, and gradually becomes a main stream machine type of a wind power installation machine. With the continuous improvement of the permeability of wind power, the direct-drive wind power plant has a great influence on the stability and reliability of a power grid, so that the problem of small disturbance stability frequently occurs. Therefore, the analysis of grid-connected small disturbance stability and the quantification of a stable domain of the direct-driven wind power plant are the problems which are urgent to be solved at present. However, current large-scale wind farm small disturbance stability studies focus mainly on grid-connected interaction analysis under a single operating condition. The related stability margin improvement measures are only aimed at a single oscillation mode, and cannot be applied to the nonlinear and multi-mode oscillation characteristics.
Disclosure of Invention
In order to solve the problems, the invention provides a method, a system, a terminal and a medium for evaluating stability of a direct-driven wind farm, which are used for performing grid-connected interaction analysis by using parameters under a plurality of operation conditions, are suitable for evaluating the stability of the wind farm with nonlinear and multi-mode oscillation characteristics, and improve the accuracy of evaluating the stability of the wind farm.
In a first aspect, the present invention provides a method for evaluating stability of a direct-drive wind farm, including the following steps:
obtaining steady state parameter samples under a plurality of operation conditions, wherein the steady state parameters comprise active power output by fans of each branch of a wind power plant and reactive power output by a reactive generator;
according to the obtained stable state parameter samples, building lumped impedance of the wind power plant grid-connected system, and taking the maximum real part of the zero point of the lumped impedance determinant as a stability margin index to build a stability margin data set;
training the stability margin data set by using a fuzzy support vector machine to obtain a wind power plant state classification model; the wind power plant state classification model divides a stable state parameter sample into a stable sample and an unstable sample by a stability margin boundary, and corresponds to a stable running state and an unstable running state of the wind power plant respectively;
acquiring a real-time steady state parameter sample of a wind power plant;
and inputting the real-time steady state parameter sample into a trained wind power plant state classification model to obtain the current steady state of the wind power plant.
In an alternative embodiment, the method for constructing the stability margin data set according to the acquired several stable state parameter samples specifically comprises the following steps:
Step 1, the acquired steady state parameter sample is recorded as;
Representing Co-acquisition->Steady state parameter samples at various operating conditions,
is the firsthSteady state parameters for each operating condition,
representing the number of wind farm branches,/->Representing real space, +.>For the active power output by the branch fan, Q SVG Reactive power output by the reactive generator;
step 2, under each operation condition, obtaining the single machine impedance corresponding to each branch of the wind power plant;
Step 3, impedance of the single branch circuitPerforming series-parallel connection calculation with the impedance of the reactive generator to construct a wind power plant impedance model +.>;
Step 4, calculating lumped impedance of wind power plant grid-connected system,
Wherein,is the impedance of the power grid;
step 5, calculating lumped impedanceIs a determinant zero point of (2);
step 6, defining the maximum real part of the determinant zero point as a stability margin indexThereby constructing a stability margin dataset +>。
In an alternative embodiment, the training of the stability margin dataset by using the fuzzy support vector machine specifically comprises:
step 1, determining a stability margin demarcation value epsilon, wherein epsilon is a number greater than or equal to zero;
step 2, re-expressing the stability margin data set as a training sample,/>Representation->According to the sample category attribute corresponding to the current stability margin demarcation value epsilon, M G >Epsilon ∈ ->= 1,M G <Epsilon ∈ ->= -1;
For fuzzy membership variable, express +.>Membership in category attribute->The extent of (3);
step 3, using mapping functionThe original parameter space->Hilbert space mapping to higher dimensions,/>Representing the dimension of Gao Weixi erbet space;
step 4, bySubstitute for the input variable +.>At Gao Weide HillHyperplane in the bert space looking for decision boundaries +.>,/>Classifying interface vectors->Is a displacement term;
step 5, defining a kernel function;
Step 6, solving the following optimal problem
,
Obtaining the optimal hyperplane;
Wherein alpha is the intermediate coefficient of the fuzzy support vector machine,Cis a penalty factor.
In an alternative embodiment, the method further comprises the steps of:
and determining a plurality of stability margin boundary values epsilon, training for each stability margin boundary value epsilon, and constructing a gradient stability domain.
In an alternative embodiment, the training of the stability margin dataset using a fuzzy support vector machine further comprises determining membershipSpecifically comprises the following steps:
step 1, useCharacteristic weights representing the elements in the steady state parameters;
step 2, for stability margin datasetSampling k times, and taking the average value of the k times of calculation results as the final characteristic weight; wherein the ith sample takes the value of randomly extracting a sample +. >Find q and +.>Nearest neighbor samples with the same category are constructed to form the same-category nearest neighbor sample set +.>From->Searching q nearest neighbor samples in each corresponding heterogeneous y, and constructing a heterogeneous nearest neighbor sample set +.>The feature weights are calculated according to the following formula:
wherein,indicating that the sample belongs to the category->Probability of->Representation sample->And sample->The difference in features is calculated as follows:
;
step 3, constructing a characteristic weighting matrix;
Step 4, calculatingSample and class center->The calculation formula is as follows:
;
s representsA total covariance matrix of the samples;
step 5, calculating membership degreeThe formula is as follows:
wherein,representing class radius.
In an alternative embodiment, the method further comprises the steps of:
and adjusting a steady state parameter of the wind farm in response to the current operating state of the wind farm being an unstable operating state.
In a second aspect, the present invention provides a system for evaluating stability of a direct-drive wind farm, comprising,
historical parameter sample acquisition module: obtaining steady state parameter samples under a plurality of operation conditions, wherein the steady state parameters comprise active power output by fans of each branch of a wind power plant and reactive power output by a reactive generator;
Training data set construction module: according to the obtained stable state parameter samples, building lumped impedance of the wind power plant grid-connected system, and taking the maximum real part of the zero point of the lumped impedance determinant as a stability margin index to build a stability margin data set;
the classification model training module: training the stability margin data set by using a fuzzy support vector machine to obtain a wind power plant state classification model; the wind power plant state classification model divides a stable state parameter sample into a stable sample and an unstable sample by a stability margin boundary, and corresponds to a stable running state and an unstable running state of the wind power plant respectively;
the real-time parameter sample acquisition module: acquiring a real-time steady state parameter sample of a wind power plant;
the real-time state detection module is used for: and inputting the real-time steady state parameter sample into a trained wind power plant state classification model to obtain the current running state of the wind power plant.
In a third aspect, a technical solution of the present invention provides a terminal, including:
the storage is used for storing a direct-drive wind farm stability evaluation program;
and the processor is used for realizing the steps of the direct-drive wind farm stability evaluation method according to any one of the above steps when executing the direct-drive wind farm stability evaluation program.
In a fourth aspect, the present invention provides a computer readable storage medium, where a direct-drive wind farm stability assessment program is stored, where the direct-drive wind farm stability assessment program when executed by a processor implements the steps of the direct-drive wind farm stability assessment method according to any one of the above.
Compared with the prior art, the method, the system, the terminal and the medium for evaluating the stability of the direct-drive wind farm have the following beneficial effects: based on steady state parameter samples under different operation conditions, a supervision type machine learning method is used for constructing a steady domain of the wind power plant on line, the constructed steady domain comprises a stability margin boundary for measuring steady states, and can specifically comprise a plurality of stability margin boundaries for measuring stability degrees to form a gradient boundary, so that the grid-connected operation stability of the wind power plant can be evaluated on line, the absolute stability can be the absolute stability and the relative stability can be the relative stability.
Drawings
For a clearer description of embodiments of the invention or of the prior art, the drawings that are used in the description of the embodiments or of the prior art will be briefly described, it being apparent that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained from them without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a method for evaluating stability of a direct-drive wind farm according to an embodiment of the present invention.
FIG. 2 is a sample of two types of samples, tight and sparseAnd->Schematic of distances to respective class centers.
FIG. 3 is a stand-alone unitAnd->Is a stable domain schematic of (a).
FIG. 4 is a schematic diagram of verification results of a time domain simulation model for the stability domain shown in FIG. 3.
FIG. 5 is a schematic diagram of a wind farm stability domain including SVG.
FIG. 6 is a schematic diagram of verification results of a time domain simulation model for the stability domain shown in FIG. 5.
Fig. 7 is a schematic block diagram of a direct-drive wind farm stability evaluation system according to an embodiment of the present invention.
Fig. 8 is a schematic structural diagram of a terminal according to an embodiment of the present invention.
Detailed Description
In order to better understand the aspects of the present invention, the present invention will be described in further detail with reference to the accompanying drawings and detailed description. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.
The following explains key terms appearing in the present invention.
Direct-drive fan: the generator is directly driven by wind power, namely a gearless wind power engine, and the generator adopts a mode of directly connecting a multipolar motor with an impeller for driving, so that a traditional part of a gear box is omitted.
Direct-drive wind farm: the wind power plant is composed of direct-drive fans, the wind power plant generally comprises a plurality of branches, and each branch is connected with a plurality of direct-drive fans. The whole wind farm is connected into an alternating current power grid through a step-up transformer. Meanwhile, a static reactive power generator is arranged on a collecting bus of the wind power plant and is used for compensating reactive power.
Impedance model: an impedance model is a model that describes the electrical characteristics of a device, line, node, or other component in a power system. It is based on the concept of electrical impedance to treat a device or node as a combination of equivalent resistance and reactance and to link it to other components in the system.
Wind farm stability domain: refers to a set of operating parameter ranges within which a wind farm can operate stably in a given power system. These parameters may include the wind farm output, wind speed, generator speed, etc. If the wind farm is operating at its maximum output, or at lower power under system disturbances, it can cause the system to be unstable or even collapse.
Fig. 1 is a schematic flow chart of a method for evaluating stability of a direct-drive wind farm according to an embodiment of the present invention. The execution body of fig. 1 may be a direct-drive wind farm stability evaluation system. The direct-drive wind farm stability evaluation method provided by the embodiment of the invention is executed by computer equipment, and correspondingly, the direct-drive wind farm stability evaluation system is operated in the computer equipment. The order of the steps in the flow chart may be changed and some may be omitted according to different needs.
As shown in fig. 1, the method includes the following steps.
S1, obtaining steady state parameter samples under a plurality of operation conditions.
The steady state parameters of the embodiment comprise the active power output by each branch fan of the wind power plant and the reactive power output by the reactive generator.
S2, constructing lumped impedance of the wind power plant grid-connected system according to the obtained plurality of steady state parameter samples, and constructing a stability margin data set by taking the maximum real part of the zero point of the lumped impedance determinant as a stability margin index.
Firstly, defining a small disturbance stability domain of a direct-drive wind power plant.
The impedance of the direct-drive wind farm is mainly related to the operating point of the system, circuit parameters and control parameters. The embodiment mainly focuses on analyzing the influence of active power of a fan and SVG reactive power on system stability. Active power output by the branch fan and reactive power output by the SVG are selected as state variables, and a parameter space is defined:
wherein,representation ofThe number of wind farm branches, < >>Representing the real space of the real number,P br for the active power output by the branch fan, Q SVG And the reactive power is output by the reactive generator.
At a given operating pointUnder this, the direct drive wind farm impedance can be expressed as +.>,sRepresenting S-domain after Law transformation, ubiquitous in transfer function, combined with grid impedance +.>Lumped impedance of grid-connected system of direct-drive wind farm can be calculated and obtained>:
By calculating lumped impedanceThe determinant zero point of the (2) can judge the stability of the system under a certain operating condition. The determinant zero of the lumped impedance is the system characteristic value. Thus, the system is stable at the current operating conditions if and only if the real part of all the zeros of the lumped impedance determinant is less than zero. Otherwise unstable. And the smaller the lumped impedance determinant zero point maximum real part, the more stable the system. Thus, it is possible to define a value corresponding to the current state +. >Stability index of->:
Wherein σ refers to the real part of the zero of the lumped impedance determinant, whenThe system is stable; when (when)The system is unstable. The small disturbance stability domain may be represented as a set of all operating conditions that result in a system stability indicator greater than zero. The small disturbance stability domain of the direct-driven wind farm grid-connected system is as follows:
。
on the basis of the stability domain definition, a stability margin data set is constructed, wherein the stability margin data set is an input training sample set for training a classification model in the subsequent step.
The embodiment provides a stable index data set construction method based on direct-drive wind farm impedance, and the construction process mainly comprises the following steps.
Step 1, the acquired steady state parameter sample is recorded as。
First, an operating condition parameter is determinedIs described. The range of variation is typically limited by fan output, converter capacity, static stability limits, and the like. Secondly, randomly selecting +.>Multiple samples of operating conditions, i.e.)>。
Step 2, at each operatorUnder the condition, obtaining the single machine impedance corresponding to each branch of the wind power plant。
Step 3, impedance of the single branch circuitPerforming series-parallel connection calculation with the impedance of the reactive generator to construct a wind power plant impedance model +.>。
Step 4, calculating lumped impedance of wind power plant grid-connected system ,
Wherein,is the grid impedance.
Step 5, calculating lumped impedanceIs a determinant zero point of (c).
Step 6, defining the maximum real part of the determinant zero point as a stability margin indexThereby constructing a stability margin dataset +>。
For a determined wind farm state variableObtaining the single machine impedance corresponding to each branch of the wind power plant>. Further performing series-parallel calculation on the branch impedance and SVG impedance to construct a direct-drive wind power plant impedance model +.>. Finally, the corresponding stability index is calculated>Thereby constructing a wind farm grid-connected stable sample data set.
To determine the boundary between stable and unstable, i.eIs a boundary of the stability index sample set +.>All->Is defined as an unstable sample, all +.>Is defined as a stable sample. Thus, the determination of the stability domain boundaries can be generalized to a classification problem. Meanwhile, the sample can be re-divided according to the requirement, and the +.>Is defined by the boundary of (a). Thereby forming a plurality of stable index boundaries and constructing a gradient stable domain. Where ε is any number greater than zero.
And S3, training the stability margin data set by using a fuzzy support vector machine to obtain a wind power plant state classification model.
In the wind power plant state classification model of the embodiment, the stability margin boundary divides the stability state parameter sample into a stable sample and an unstable sample, which correspond to the stable running state and the unstable running state of the wind power plant respectively.
Based on the stability index data set, a machine learning intelligent algorithm can be adopted to process the classification problem of the stability data and the instability data division, and the stability boundary of the direct-driven wind power plant is identified. The traditional support vector machine has a relatively slow training speed for a sample set with a very large data volume, is difficult to reduce noise, and is difficult to meet the calculation speed requirement of online construction of a stable domain. Therefore, the embodiment adopts a fuzzy support vector machine algorithm, and each sample point is endowed with a membership degree to realize the importance distinction of the samples. The algorithm enables different training samples to play different roles in training the optimal classification decision surface, and reduces the influence of noise on the optimal decision surface.
The fuzzy support vector machine is used for training the stability margin data set, and the method specifically comprises the following steps.
And step 1, determining a stability margin demarcation value epsilon, wherein epsilon is a number greater than or equal to zero.
Step 2, re-expressing the stability margin data set as a training sample,/>Representation->M is calculated according to the sample category attribute corresponding to the current stability margin demarcation value epsilon G >Epsilon ∈ ->= 1,M G <Epsilon ∈ ->= -1;
For fuzzy membership variable, express +.>Membership in category attribute->To a degree of (3).
This embodimentIn order to improve the calculation efficiency and the classification precision, the stability margin data set is classified according to the selected stability margin demarcation value and re-expressed 。
Step 3, using mapping functionThe original parameter space->Hilbert space mapping to higher dimensions,/>Representing the dimension of Gao Weixi erbet space.
Because the mapping relation between the stability margin and the operation condition parameters is complex, the stability margin and the operation condition parameters are difficult to be in the original parameter spaceThe original samples are classified linearly. Use of mapping function->Can be used for spatial->Hilbert space mapped to higher dimensions +.>。
Step 4, bySubstitute for the input variable +.>Hyperplane for finding decision boundaries in Gao Weide Hilbert space +.>,/>Classifying interface vectors->Is a displacement term.
It should be noted that, according to different values of the stability indicator, a plurality of decision boundaries may be established.
Step 5, defining a kernel function。
Step 6, solving the following optimal problem
,
Obtaining the optimal hyperplane。
To obtain the optimal hyperplane for proper classification of all samples, the following optimal problem needs to be solved:
wherein,for classifying interface vectors, ++>For punishment factors->Is a relaxation factor. />Corresponding to the assignment of different weights to the classification errors. The smaller the product is, the smaller the representationThe less likely it is to be classified as erroneous, and the greater the likelihood of correct. From the above, it can be seen that when +.>The smaller the slack factor, the less the effect that the slack factor plays in the optimization process. Therefore, training sample during training +. >The smaller the effect that is played. />The sample inner product after the high-dimensional mapping is introduced. If it is directly in the high-dimensional space +.>The complexity of computing the sample inner product will be high. Thus, to calculate the inner product of the sample, a kernel function is defined. Thereby +.>Middle calculation of inner product
The dual form of the original optimization problem after the kernel function is adopted is as follows:
wherein alpha is the intermediate coefficient of the fuzzy support vector machine, and finally, the stable boundary under the original dimension can be obtained。
Fuzzy membership variableIs determined by (a)Is a difficulty of the fuzzy support vector machine, which affects the accuracy of the algorithm classification. The membership function commonly used at present is mainly based on the distance from the sample to the class center to determine the membership. Class centers are typically chosen as the average of the class samples. However, this method only considers the distance of the samples from the class center, and does not consider the degree of tightness between samples. The exception vector and the support vector cannot be effectively distinguished. As shown in FIG. 2, given a tight and sparse sample, sample +.>And->The distances D to the respective class centers are equal. />Possibly a support vector, but +.>It is more likely to be an outlier vector. Therefore, the present embodiment proposes a membership function based on weighted mahalanobis distance. The mahalanobis distance considers both the distance of the sample from the center of the class and the distribution of the sample. Simultaneously, calculating sample state variable +. >The characteristic weights of the parameters distinguish the influence degree of the parameters of the state variables on the stable boundary, so that the classification effect is improved. For fig. 2, it is possible to effectively give +.>Is a conclusion of (2).
(1) Relief-F feature weighting
The Relief-F algorithm judges the importance degree of each parameter of the state variable on the stability boundary according to the characteristic distance difference between similar and dissimilar neighbor samples. When the distance between the adjacent samples of the same class is very small and the distance between the nearest samples of different classes is very different, the parameter is beneficial to classification, and the parameter is given a larger weight. And conversely, the feature weight is reduced. The embodiment adopts the algorithm to calculate the characteristic weight of each parameter of the sample state variable.
For the state variable of the direct-drive wind farmIs common->And parameters. />Characteristic weights representing the parameters. Calculation of feature weights requires +.>Subsampling. />The initial value of (2) is set to 0, < ->The sub-sampling value is calculated as follows: randomly extracting a sample for the stability margin dataset>Find->Person and->Nearest neighbor samples with the same category are constructed to form the same-category nearest neighbor sample set +.>. Similarly from->Corresponding each of the different classes- >Search for->Constructing a heterogeneous nearest neighbor sample set +.>. The Relief-F algorithm allows for more than one heterogeneous.
Wherein,indicating that the sample belongs to the category->Probability of->Representation sample->And sample->The difference in characteristics is calculated as follows:
and finally, taking the average value of the k times of calculation results as the final characteristic weight.
(2) Weighted mahalanobis distance
The mahalanobis distance is calculated based on global information of the sample, representing the covariance distance of the data. The distance can effectively remove the association interference among the attributes, is independent of the measurement scale of the sample data, and eliminates dimensionality. Thus, the mahalanobis distance is particularly suitable for state variable stability domain analysis having a plurality of different dimensions in this embodiment. The mahalanobis distance between any two samples is calculated as follows
Wherein,is the overall covariance matrix of the sample.
Characteristic weights of parameters of sample state variables obtained according to Relief-F algorithmConstructing a feature weighting matrix +.>
。
The sample is thenAnd class center->The weighted mahalanobis distance calculation of (a) is:
。
and determining the importance of the sample according to the weighted Markov distance from the sample point to the class center of the sample point, wherein the smaller the distance is, the larger the membership value is given to the sample. Membership function expression based on weighted mahalanobis distance is as follows
Wherein,representing class radius.
S4, acquiring a real-time steady state parameter sample of the wind power plant.
S5, inputting the real-time steady state parameter sample into a trained wind power plant state classification model to obtain the current running state of the wind power plant.
The embodiment detects the real-time running state of the wind power plant based on the training model, and adjusts the stable state parameters of the wind power plant in response to the current running state of the wind power plant being an unstable running state, so as to guide the actual running regulation.
In order to comprehensively reflect grid-connected operation characteristics of the direct-driven wind power plant, the system safety stability boundary under the time-varying working condition needs to be considered in the wind power plant small disturbance stability analysis, a quantized small disturbance stability domain with considerable operation working conditions is constructed on line in the embodiment, definition of the small disturbance stability domain of the direct-driven wind power plant is given in the section, and an on-line quantization method of the small disturbance stability domain based on a fuzzy support vector machine is provided. The method establishes a stability margin dataset based on determining wind farm impedance at an operating point. Then, a stable domain and a stable boundary of the system are established by adopting a fuzzy support vector machine, and the small disturbance stable domain boundary can be accurately and efficiently quantized in a high-dimensional parameter space.
Based on the proposed small disturbance stability domain, the operational stability is enhanced by changing the operational operating point. The following provides a specific embodiment for considering the scene that the power of a single fan of a wind power plant is changed independently and the power of a whole fan is changed cooperatively, establishing a stable domain of the wind power plant, and researching the stability of the wind power plant.
Set up stand aloneAnd->Is represented by a gradient color>Size of the product. As can be seen from FIG. 3, the stability margin is varied>And->Is a common influence of (a) and (b). Specifically, in the whole stable domain, < > is>The larger the system is, the more stable the system is>The larger the system, the more unstable the system tends to be. In State1 in FIG. 3, < > therein>0.9, MW->5 MW, the operating point is at +.>>0. Is stable. In the process from State1 to State2, the total output power value of the wind farm is changed, +.>The power of a single fan 1 is unchanged when the power is increased to 5.28 and MW, the power set values of other fans are increased proportionally, the stability of a wind farm is gradually deteriorated, and the wind farm enters +.><0. Is not stable. Changing the power of the blower 1, +.>Increasing to 1.2 MW, the total power set point of the wind farm is unchanged, the power of other fans is reduced proportionally, the system goes from state2 to state 3 in fig. 3, and the system can return to stability.
Subsequently, a time domain simulation model is designed for the wind farm stability domain shown in FIG. 3 for verification. As shown in fig. 4, the system was running steadily at State1 before 7 seconds. Changing power in simulation process, increasing total power of wind power plant in 7 seconds, causing instability by small interference of system, and generating wind power plant And->The oscillation is gradually increased, and the power of a single fan is +.>And->The oscillations diverge as well, enter State2, and the system is unstable. In order to suppress unstable oscillations of the wind farm, the power distribution of the stand-alone 1 is varied at 11 seconds such that +.>Increasing to 1.2 MW, the system stabilizes after a short oscillation, entering State3. The simulation results are shown in fig. 4. As can be seen from the figure, when the total active power of the wind field increases, the system transitions from a steady state to an unstable state, and at this time, the power set point of the fan 1 is increased, the distributed power in the field is changed, and the system is restored to steady state again. The power data of the wind farm in the example of fig. 4 is measured by the power grid side, and the power value is slightly lower than the power set value of the wind farm in fig. 3 due to the fact that the power value is slightly lost due to the influence of the power grid impedance. In State3, because the power of each fan of the wind power plant is redistributed, the output power of different fans is changed, and the total power loss of the wind power plant caused by the impedance of the power grid is also changed, so that the total power value of the wind power plant in State2 and State3 is>There is a small difference in the values of (c). And the overall result of the time domain calculation example is consistent with the stability analysis result of the wind power plant, so that the correctness of the conclusion is verified.
And (3) considering SVG, establishing a wind power plant stability domain comprising SVG, and analyzing the influence of the reactive compensation device on the stability of the wind power plant. Establishing a wind farm stability domain considering SVG. The larger the QSVG, the more stable the system will be, The larger the system, the more unstable the system tends to be. In FIG. 5, +.1 for State>At 5 MW, SVG does not generate reactive power, and the working point is located at MGP>0. Is stable. Wind farm power->Changing to 5.55 and MW, proportionally increasing the power set value of a single fan, still providing reactive output by SVG, changing the working point from State1 to State2, and controlling the system MGP<0, the system enters an unstable region. At the moment, the total power set value of the wind power plant is unchanged, the SVG is enabled to send out 0.4 MVar reactive power, the system enters into the State3 working point, and at the moment, the MGP>0, it can be seen that by letting the SVG emit the proper reactive power, the unstable system can be brought back to a steady state.
The conclusion of the wind farm stability domain analysis considering SVG is verified through time domain simulation, and the simulation result is shown in FIG. 6. Before 7 seconds, the system was running steadily at State1,at 5 MW, QSVG at 0 MVar. At 7 seconds, the total power of the wind farm increases, < >>At 5.55 MW, QSVG is still 0 MVar, instability caused by small disturbance of the system occurs, the oscillation amplitude of each variable in the graph is gradually increased, and the system enters State2. In order to inhibit the unstable oscillation of the power grid from continuously expanding, the SVG is enabled to emit reactive power of 0.4 MVar at 9 seconds, the system enters State3, the oscillation amplitude is gradually reduced, and the system enters a stable area. Whereby the stable domain results of fig. 6 are verified.
Based on the branch impedance and the stable sample data thereof under different operation conditions, the steady domain on-line construction method based on the supervised machine learning method is provided in the embodiment. The fuzzy support vector machine based on the weighted mahalanobis distance is adopted, so that the calculated amount of model training can be obviously reduced, and the requirement of constructing a stable domain on line is met. The obtained stability domain comprises gradient boundaries for measuring stability, and absolute stability and relative stability of grid connection of the wind power plant can be evaluated on line. Finally, based on the proposed stability domain, the present embodiment proposes a practical method for enhancing stability by changing the operating point. By adjusting the active power and SVG reactive power of the wind farm, the grid-connected stability of the wind farm can be obviously affected, and the result can be used for guiding the actual operation regulation.
The embodiment of the method for evaluating the stability of the direct-drive wind farm is described in detail above, and the embodiment of the invention further provides a system for evaluating the stability of the direct-drive wind farm, which corresponds to the method, based on the method for evaluating the stability of the direct-drive wind farm described in the embodiment.
Fig. 7 is a schematic block diagram of a direct-drive wind farm stability assessment system according to an embodiment of the present invention, where the direct-drive wind farm stability assessment system 200 may be divided into a plurality of functional modules according to the functions performed by the direct-drive wind farm stability assessment system, as shown in fig. 7. The functional module may include: a historical parameter sample acquisition module 710, a training data set construction module 720, a classification model training module 730, a real-time parameter sample acquisition module 740, and a real-time status detection module 750. The module referred to in the present invention refers to a series of computer program segments capable of being executed by at least one processor and of performing a fixed function, stored in a memory.
Historical parameter sample acquisition module 710: and obtaining steady state parameter samples under a plurality of operation conditions, wherein the steady state parameters comprise active power output by fans of each branch of the wind power plant and reactive power output by a reactive generator.
Training data set construction module 720: and constructing lumped impedance of the wind power plant grid-connected system according to the acquired plurality of steady state parameter samples, and taking the maximum real part of the zero point of the lumped impedance determinant as a stability margin index to construct a stability margin data set.
Classification model training module 730: training the stability margin data set by using a fuzzy support vector machine to obtain a wind power plant state classification model; the wind power plant state classification model divides a stable state parameter sample into a stable sample and an unstable sample by a stability margin boundary, and the stable operation state and the unstable operation state of the wind power plant are respectively corresponding to each other.
Real-time parameter sample acquisition module 740: and acquiring a real-time steady state parameter sample of the wind power plant.
Real-time status detection module 750: and inputting the real-time steady state parameter sample into a trained wind power plant state classification model to obtain the current running state of the wind power plant.
The direct-drive wind farm stability evaluation system of the embodiment is used for realizing the direct-drive wind farm stability evaluation method, so that the specific implementation of the device can be seen from the foregoing example part of the direct-drive wind farm stability evaluation method, and therefore, the specific implementation of the device can be referred to the description of the examples of the corresponding parts, and will not be further described herein.
In addition, since the direct-drive wind farm stability evaluation system of the embodiment is used for implementing the foregoing direct-drive wind farm stability evaluation method, the function thereof corresponds to that of the foregoing method, and the description thereof is omitted herein.
Fig. 8 is a schematic structural diagram of a terminal 800 according to an embodiment of the present invention, including: processor 810, memory 820, and communication unit 830. The processor 810 is configured to implement the following steps when implementing the direct-drive wind farm stability evaluation program stored in the memory 820:
obtaining steady state parameter samples under a plurality of operation conditions, wherein the steady state parameters comprise active power output by fans of each branch of a wind power plant and reactive power output by a reactive generator;
according to the obtained stable state parameter samples, building lumped impedance of the wind power plant grid-connected system, and taking the maximum real part of the zero point of the lumped impedance determinant as a stability margin index to build a stability margin data set;
training the stability margin data set by using a fuzzy support vector machine to obtain a wind power plant state classification model; the wind power plant state classification model divides a stable state parameter sample into a stable sample and an unstable sample by a stability margin boundary, and corresponds to a stable running state and an unstable running state of the wind power plant respectively;
Acquiring a real-time steady state parameter sample of a wind power plant;
and inputting the real-time steady state parameter sample into a trained wind power plant state classification model to obtain the current running state of the wind power plant.
The terminal 800 includes a processor 810, a memory 820, and a communication unit 830. The components may communicate via one or more buses, and it will be appreciated by those skilled in the art that the configuration of the server as shown in the drawings is not limiting of the invention, as it may be a bus-like structure, a star-like structure, or include more or fewer components than shown, or may be a combination of certain components or a different arrangement of components.
The memory 820 may be implemented by any type of volatile or non-volatile memory terminal or combination thereof, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk, among other things, for storing instructions for execution by the processor 810. The execution of the instructions in memory 820, when executed by processor 810, enables terminal 800 to perform some or all of the steps in the method embodiments described below.
The processor 810 is a control center of the storage terminal, connects various parts of the entire electronic terminal using various interfaces and lines, and performs various functions of the electronic terminal and/or processes data by running or executing software programs and/or modules stored in the memory 820, and invoking data stored in the memory. The processor may be comprised of an integrated circuit (Integrated Circuit, simply referred to as an IC), for example, a single packaged IC, or may be comprised of a plurality of packaged ICs connected to the same function or different functions. For example, the processor 810 may include only a central processing unit (Central Processing Unit, simply CPU). In the embodiment of the invention, the CPU can be a single operation core or can comprise multiple operation cores.
A communication unit 830, configured to establish a communication channel, so that the storage terminal may communicate with other terminals. Receiving user data sent by other terminals or sending the user data to other terminals.
The invention also provides a computer storage medium, which can be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (random access memory, RAM) and the like.
The computer storage medium stores a direct-drive wind farm stability evaluation program, which when executed by the processor, realizes the following steps:
obtaining steady state parameter samples under a plurality of operation conditions, wherein the steady state parameters comprise active power output by fans of each branch of a wind power plant and reactive power output by a reactive generator;
according to the obtained stable state parameter samples, building lumped impedance of the wind power plant grid-connected system, and taking the maximum real part of the zero point of the lumped impedance determinant as a stability margin index to build a stability margin data set;
training the stability margin data set by using a fuzzy support vector machine to obtain a wind power plant state classification model; the wind power plant state classification model divides a stable state parameter sample into a stable sample and an unstable sample by a stability margin boundary, and corresponds to a stable running state and an unstable running state of the wind power plant respectively;
acquiring a real-time steady state parameter sample of a wind power plant;
and inputting the real-time steady state parameter sample into a trained wind power plant state classification model to obtain the current running state of the wind power plant.
It will be apparent to those skilled in the art that the techniques of embodiments of the present invention may be implemented in software plus a necessary general purpose hardware platform. Based on such understanding, the technical solution in the embodiments of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium such as a U-disc, a mobile hard disc, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk or an optical disk, etc. various media capable of storing program codes, including several instructions for causing a computer terminal (which may be a personal computer, a server, or a second terminal, a network terminal, etc.) to execute all or part of the steps of the method described in the embodiments of the present invention.
In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The foregoing disclosure is merely illustrative of the preferred embodiments of the invention and the invention is not limited thereto, since modifications and variations may be made by those skilled in the art without departing from the principles of the invention.
Claims (8)
1. The method for evaluating the stability of the direct-drive wind farm is characterized by comprising the following steps of:
obtaining steady state parameter samples under a plurality of operation conditions, wherein the steady state parameters comprise active power output by fans of each branch of a wind power plant and reactive power output by a reactive generator;
according to the obtained stable state parameter samples, building lumped impedance of the wind power plant grid-connected system, and taking the maximum real part of the zero point of the lumped impedance determinant as a stability margin index to build a stability margin data set;
training the stability margin data set by using a fuzzy support vector machine to obtain a wind power plant state classification model; the wind power plant state classification model divides a stable state parameter sample into a stable sample and an unstable sample by a stability margin boundary, and corresponds to a stable running state and an unstable running state of the wind power plant respectively;
acquiring a real-time steady state parameter sample of a wind power plant;
Inputting the real-time steady state parameter sample into a trained wind power plant state classification model to obtain the current steady state of the wind power plant;
the method comprises the steps of constructing a stability margin data set according to a plurality of acquired stability state parameter samples, and specifically comprises the following steps:
step 1, the acquired steady state parameter sample is recorded as;
Representing Co-acquisition->Steady state parameter samples at various operating conditions,
is the firsthSteady state parameters for each operating condition,
representing the number of wind farm branches,/->Representing real space, +.>For the active power output by the branch fan, Q SVG Reactive power output by the reactive generator;
step 2, under each operation condition, obtaining the single machine impedance corresponding to each branch of the wind power plant;
Step 3, impedance of the single branch circuitPerforming series-parallel calculation with the impedance of the reactive generator to construct a wind power plant impedance model;
Step 4, calculating lumped impedance of wind power plant grid-connected system,
Wherein,is the impedance of the power grid;
step 5, calculating lumped impedanceIs a determinant zero point of (2);
step 6, defining the maximum real part of the determinant zero point as a stability margin indexThereby constructing a stability margin dataset +>。
2. The direct drive wind farm stability assessment method of claim 1, wherein training the stability margin dataset using a fuzzy support vector machine, specifically comprises:
Step 1, determining a stability margin demarcation value epsilon, wherein epsilon is a number greater than or equal to zero;
step 2, re-expressing the stability margin data set as a training sample,/>Representation ofAccording to the sample category attribute corresponding to the current stability margin demarcation value epsilon,M G >epsilon ∈ -> = 1,M G <Epsilon is then = -1;
For fuzzy membership variable, express +.>Membership in category attribute->The extent of (3);
step 3, using mapping functionThe original parameter space->Hilbert space mapping to higher dimensions,/>Representing the dimension of Gao Weixi erbet space;
step 4, bySubstitute for the input variable +.>Hyperplane for finding decision boundaries in Gao Weide Hilbert space +.>,/>Classifying interface vectors->Is a displacement term;
step 5, defining a kernel function;
Step 6, solving the following optimal problem
,
Obtaining the optimal hyperplane;
Wherein alpha is the intermediate coefficient of the fuzzy support vector machine,Cis a penalty factor.
3. The direct drive wind farm stability assessment method of claim 2, further comprising the steps of:
and determining a plurality of stability margin boundary values epsilon, training for each stability margin boundary value epsilon, and constructing a gradient stability domain.
4. A method of direct drive wind farm stability assessment according to claim 2 or 3, wherein the stability margin dataset is trained using a fuzzy support vector machine, further comprising determining membership Specifically comprises the following steps:
step 1, useCharacteristic weights representing the elements in the steady state parameters;
step 2, for stability margin datasetSampling k times, and taking the average value of the k times of calculation results as the final characteristic weight; wherein the ith sample takes the value of randomly extracting a sample +.>Find q and +.>Nearest neighbor samples with the same category are constructed to form the same-category nearest neighbor sample set +.>From->Searching q nearest neighbor samples in each corresponding heterogeneous y, and constructing a heterogeneous nearest neighbor sample set +.>The feature weights are calculated according to the following formula:
wherein,indicating that the sample belongs to the category->Probability of->Representation sample->And sample->The difference in features is calculated as follows:
;
step 3, constructing featuresWeighting matrix;
Step 4, calculatingSample and class center->The calculation formula is as follows:
;
s representsA total covariance matrix of the samples;
step 5, calculating membership degreeThe formula is as follows:
wherein,representing class radius.
5. The direct drive wind farm stability assessment method of claim 1, further comprising the steps of:
and adjusting a steady state parameter of the wind farm in response to the current operating state of the wind farm being an unstable operating state.
6. A direct-drive wind farm stability assessment system is characterized by comprising,
historical parameter sample acquisition module: obtaining steady state parameter samples under a plurality of operation conditions, wherein the steady state parameters comprise active power output by fans of each branch of a wind power plant and reactive power output by a reactive generator;
training data set construction module: according to the obtained stable state parameter samples, building lumped impedance of the wind power plant grid-connected system, and taking the maximum real part of the zero point of the lumped impedance determinant as a stability margin index to build a stability margin data set;
the classification model training module: training the stability margin data set by using a fuzzy support vector machine to obtain a wind power plant state classification model; the wind power plant state classification model divides a stable state parameter sample into a stable sample and an unstable sample by a stability margin boundary, and corresponds to a stable running state and an unstable running state of the wind power plant respectively;
the real-time parameter sample acquisition module: acquiring a real-time steady state parameter sample of a wind power plant;
the real-time state detection module is used for: inputting the real-time steady state parameter sample into a trained wind power plant state classification model to obtain the current running state of the wind power plant;
The training data set construction module constructs a stability margin data set according to a plurality of acquired steady state parameter samples, and specifically comprises the following steps:
step 1, the acquired steady state parameter sample is recorded as;
Representing Co-acquisition->Steady state parameter samples at various operating conditions,
is the firsthSteady state parameters for each operating condition,
representing the number of wind farm branches,/->Representing real space, +.>For the active power output by the branch fan, Q SVG Reactive power output by the reactive generator;
step 2, under each operation condition, obtaining the single machine impedance corresponding to each branch of the wind power plant;
Step 3, impedance of the single branch circuitPerforming series-parallel calculation with the impedance of the reactive generator to construct a wind power plant impedance model;
Step 4, calculating lumped impedance of wind power plant grid-connected system,
Wherein,is the impedance of the power grid;
step 5, calculating lumped impedanceIs a determinant zero point of (2);
step 6, defining the maximum real part of the determinant zero point as a stability margin indexThereby constructing a stability margin dataset +>。
7. A terminal, comprising:
the storage is used for storing a direct-drive wind farm stability evaluation program;
the processor is used for implementing the steps of the direct-drive wind farm stability assessment method according to any one of claims 1-5 when the direct-drive wind farm stability assessment program is executed.
8. A computer readable storage medium, characterized in that the readable storage medium has stored thereon a direct drive wind farm stability assessment program, which when executed by a processor, implements the steps of the direct drive wind farm stability assessment method according to any of claims 1-5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311552495.4A CN117277422B (en) | 2023-11-21 | 2023-11-21 | Method, system, terminal and medium for evaluating stability of direct-drive wind farm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311552495.4A CN117277422B (en) | 2023-11-21 | 2023-11-21 | Method, system, terminal and medium for evaluating stability of direct-drive wind farm |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117277422A CN117277422A (en) | 2023-12-22 |
CN117277422B true CN117277422B (en) | 2024-03-15 |
Family
ID=89203046
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311552495.4A Active CN117277422B (en) | 2023-11-21 | 2023-11-21 | Method, system, terminal and medium for evaluating stability of direct-drive wind farm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117277422B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110224392A (en) * | 2019-05-10 | 2019-09-10 | 广西电网有限责任公司电力科学研究院 | It is a kind of for analyzing the Unscented transform method of the probability of voltage stabilization containing wind power system |
CN113644689A (en) * | 2021-08-16 | 2021-11-12 | 清华大学 | Method and device for constructing wind power system operation stable domain, electronic equipment and readable storage medium thereof |
CN113725910A (en) * | 2021-10-08 | 2021-11-30 | 南通大学 | Stability analysis and quantitative evaluation method for wind power plant grid-connected system |
CN113991651A (en) * | 2021-10-27 | 2022-01-28 | 武汉大学 | Power grid static voltage stability margin probability prediction method considering new energy uncertainty |
CN116861321A (en) * | 2023-06-21 | 2023-10-10 | 山东大学 | Wind farm stability judging method and system based on stable domain quantification |
CN116896093A (en) * | 2023-07-13 | 2023-10-17 | 南京理工大学 | Online analysis and optimization method for grid-connected oscillation stability of wind farm |
-
2023
- 2023-11-21 CN CN202311552495.4A patent/CN117277422B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110224392A (en) * | 2019-05-10 | 2019-09-10 | 广西电网有限责任公司电力科学研究院 | It is a kind of for analyzing the Unscented transform method of the probability of voltage stabilization containing wind power system |
CN113644689A (en) * | 2021-08-16 | 2021-11-12 | 清华大学 | Method and device for constructing wind power system operation stable domain, electronic equipment and readable storage medium thereof |
CN113725910A (en) * | 2021-10-08 | 2021-11-30 | 南通大学 | Stability analysis and quantitative evaluation method for wind power plant grid-connected system |
CN113991651A (en) * | 2021-10-27 | 2022-01-28 | 武汉大学 | Power grid static voltage stability margin probability prediction method considering new energy uncertainty |
CN116861321A (en) * | 2023-06-21 | 2023-10-10 | 山东大学 | Wind farm stability judging method and system based on stable domain quantification |
CN116896093A (en) * | 2023-07-13 | 2023-10-17 | 南京理工大学 | Online analysis and optimization method for grid-connected oscillation stability of wind farm |
Non-Patent Citations (1)
Title |
---|
张旸等.集成静止无功发生装置的直驱风场序阻抗网络模型与稳定性分析.中国电机工程学报.2020,(第09期),全文. * |
Also Published As
Publication number | Publication date |
---|---|
CN117277422A (en) | 2023-12-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liao et al. | Fault diagnosis of power transformers using graph convolutional network | |
CN106600059B (en) | Intelligent power grid short-term load prediction method based on improved RBF neural network | |
CN111860982A (en) | Wind power plant short-term wind power prediction method based on VMD-FCM-GRU | |
Ayodeji et al. | Causal augmented ConvNet: A temporal memory dilated convolution model for long-sequence time series prediction | |
CN108051660A (en) | A kind of transformer fault combined diagnosis method for establishing model and diagnostic method | |
US20210334658A1 (en) | Method for performing clustering on power system operation modes based on sparse autoencoder | |
Zhang et al. | A novel data-driven method based on sample reliability assessment and improved CNN for machinery fault diagnosis with non-ideal data | |
Qi et al. | A new adaptation method based on adaptability under k-nearest neighbors for case adaptation in case-based design | |
CN110321493A (en) | A kind of abnormality detection of social networks and optimization method, system and computer equipment | |
CN114609994A (en) | Fault diagnosis method and device based on multi-granularity regularization rebalance incremental learning | |
Cui et al. | Intelligent fault quantitative identification via the improved deep deterministic policy gradient (DDPG) algorithm accompanied with imbalanced sample | |
CN111950840A (en) | Intelligent operation and maintenance knowledge retrieval method and system for metrological verification device | |
CN112561303B (en) | Power system dynamic analysis method based on integrated learning and power grid topology change | |
Vo et al. | Harnessing attention mechanisms in a comprehensive deep learning approach for induction motor fault diagnosis using raw electrical signals | |
CN114239949A (en) | Website access amount prediction method and system based on two-stage attention mechanism | |
CN117277422B (en) | Method, system, terminal and medium for evaluating stability of direct-drive wind farm | |
CN116316699A (en) | Large power grid frequency security situation prediction method, device and storage medium | |
CN113111588B (en) | NO of gas turbine X Emission concentration prediction method and device | |
CN112419098B (en) | Power grid safety and stability simulation sample screening and expanding method based on safety information entropy | |
CN115270983A (en) | Switch cabinet fault prediction method based on AdaBoost-RBF algorithm | |
Luo et al. | A Novel Generative Adversarial Networks via Music Theory Knowledge for Early Fault Intelligent Diagnosis of Motor Bearings | |
Bin et al. | Simple and effective fault diagnosis method of power lithium-ion battery based on GWA-DBN | |
Qi et al. | Analysis and Prediction of Energy Consumption in Neural Networks Based on Machine Learning | |
Yin et al. | Graph attention-based U-net conditional generative adversarial networks for the identification of synchronous generation unit parameters | |
CN111798049B (en) | Voltage stability evaluation method based on integrated learning and multi-target planning |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |