CN111814395A - Static voltage stability evaluation method based on principal component analysis and confidence detection - Google Patents
Static voltage stability evaluation method based on principal component analysis and confidence detection Download PDFInfo
- Publication number
- CN111814395A CN111814395A CN202010616348.9A CN202010616348A CN111814395A CN 111814395 A CN111814395 A CN 111814395A CN 202010616348 A CN202010616348 A CN 202010616348A CN 111814395 A CN111814395 A CN 111814395A
- Authority
- CN
- China
- Prior art keywords
- evaluation
- confidence
- sample set
- principal component
- voltage stability
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 76
- 238000001514 detection method Methods 0.000 title claims abstract description 23
- 238000000513 principal component analysis Methods 0.000 title claims abstract description 17
- 230000003068 static effect Effects 0.000 title claims description 24
- 238000000034 method Methods 0.000 claims abstract description 43
- 238000013210 evaluation model Methods 0.000 claims abstract description 26
- 238000005259 measurement Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 8
- 230000001360 synchronised effect Effects 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 7
- 238000013144 data compression Methods 0.000 claims description 6
- 238000012935 Averaging Methods 0.000 claims description 5
- 238000012847 principal component analysis method Methods 0.000 claims description 4
- 238000007906 compression Methods 0.000 claims description 3
- 230000006835 compression Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000013097 stability assessment Methods 0.000 claims description 3
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 abstract 1
- 238000012360 testing method Methods 0.000 description 12
- 238000011160 research Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 230000006641 stabilisation Effects 0.000 description 3
- 238000011105 stabilization Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011056 performance test Methods 0.000 description 1
- 238000010206 sensitivity analysis Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000000192 social effect Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Economics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Medical Informatics (AREA)
- Educational Administration (AREA)
- Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Primary Health Care (AREA)
- Operations Research (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
A method for online voltage stability evaluation based on principal component analysis and confidence detection comprises the following specific steps: the method comprises the following steps: establishing an initial sample set based on historical operating data of the power system, and normalizing; step two: compressing and reducing dimensions of the initial sample set to obtain a high-efficiency sample set; step three: completing the updating of the evaluation model; step four: performing real-time voltage stability evaluation by using the updated evaluation model to obtain an online evaluation result; step five: and evaluating the real-time voltage stability evaluation result by using a confidence detection method to obtain a final online evaluation result. The invention aims to provide a high-precision and high-efficiency online voltage stability evaluation model of a power system, which is beneficial to system operators to better determine the running state of the power system, improves the running safety and reliability of a power grid and reduces the social and economic losses caused by serious power system accidents.
Description
Technical Field
The invention belongs to the field of voltage stability evaluation of power systems, and particularly relates to a static voltage stability evaluation method based on principal component analysis and confidence detection.
Background
The power system is a complex industrial system, the operating state of which has been the focus of attention. In recent years, with the sustainable development of renewable energy sources, the wide-area interconnection of modern power systems, the operation of new equipment, high-voltage-level power transmission and the increasing of transmission capacity. The operation burden of the power system is continuously increased, and a great challenge is brought to the safe operation of the power system. However, the static voltage stabilization is a key for safe operation of the power system, a lot of major power failure accidents worldwide are related to the static voltage stabilization, voltage collapse may cause huge economic loss and adverse social effects, and research on evaluation of the static voltage stabilization is increasingly receiving wide attention. Therefore, reliable, rapid and accurate voltage stability assessment is carried out, the safety state of the system is identified, the system is prevented, controlled and protected, and the system is maintained to operate safely and stably, and the method has important significance. In general, the static Voltage Stability evaluation is conducted by measuring the distance from a certain operating point to a Voltage breakdown point according to a Voltage Stability Margin (VSM). Aiming at the technology, various voltage stability evaluation methods are available, which mainly include two angles: mechanism research and data driving.
The method based on mechanism research comprises a singular value decomposition method, a sensitivity analysis method and a continuous power flow method. However, modern power systems are increasingly complex and increasingly large in scale. The traditional mechanism research method is difficult to use in a large system and has low calculation speed, and real-time and effective evaluation on the power system cannot be guaranteed.
The data driving-based method mainly comprises an artificial neural network, an extreme learning machine, a decision tree, a support vector machine and the like. However, with the rapid development of the wide area measurement system and the wide application of the synchronous phasor measurement unit, data collection becomes more convenient and faster, and analysis and processing become more and more complicated due to the generated huge data. Therefore, the conventional data driving method has many defects and shortcomings, such as: processing a large amount of sample data is difficult and often lost when processing data.
In summary, the current static voltage evaluation method cannot meet the requirement of the modern power system on voltage stability evaluation, and a high-adaptability and high-precision evaluation method is urgently needed.
Disclosure of Invention
In view of the above problems and deficiencies, the present invention provides an online voltage stability assessment method based on principal component analysis and confidence detection to more effectively address the problems encountered in power system safety assessment. The method for analyzing the principal components is used for completing data compression and dimensionality reduction, constructing a high-precision and high-efficiency online voltage evaluation model, evaluating the safety state of a system in operation in real time, and improving the safety and stability of the operation of the power system.
In order to achieve the above object, the method of the present invention specifically comprises the following steps:
a static voltage stability evaluation method based on principal component analysis and confidence detection comprises the following steps:
step one): based on historical operating data of the power system, establishing a Voltage Stability Margin (VSM) index, establishing an initial sample set, and performing standard normalization;
step two): aiming at the high-efficiency initial sample set, a principal component analysis method is used to realize the compression and dimension reduction of the initial sample set and obtain a high-efficiency sample set;
step three): based on the high-efficiency sample set, a static voltage stability evaluation model is constructed by combining a regression tree and integrated learning, and the high-efficiency sample set is updated to complete the updating of the evaluation model;
step four): collecting power system operation data in real time based on the synchronous phasor measurement unit and the wide area measurement system, carrying out corresponding analysis processing, and carrying out real-time voltage stability evaluation by using an updated evaluation model;
step five): and evaluating the real-time voltage stability evaluation result by using a confidence detection method to obtain a final online evaluation result.
In the first step), a continuous power flow method is used for determining a voltage stability limit for each working point according to historical operating data, continuous VSM indexes are constructed according to the active power difference value of the load between the working point and the voltage collapse point, and the steady-state operating data characteristic of each working point corresponds to a corresponding stability margin index so as to form a corresponding relation between the characteristic and the index.
In step one): determining the limit of the static voltage stability of the system by adopting a continuous power flow method to obtain a P-V curve of the system, and calculating the VSM of the power system through the P-V curve; the VSM of a power system is defined as shown in equation (1):
in the formula: Δ P is the active power margin; pmaxIs the maximum power before voltage collapse.
In the first step), the initial sample set is subjected to standard normalization to reduce the computational burden of the machine, as shown in formula (2)
In the formula:
the value of a certain operation variable after standard normalization; x is the number ofiIs the original value of the operating variable; x is the number ofi_minThe minimum value of the variable in the obtained sample; x is the number ofi_maxThe maximum value of the variable in the obtained sample; in this way, the values of all variables are varied from 0 to 1.
In the second step), when compressing and dimensionality reduction is performed on the initial sample set, the method specifically comprises the following steps:
(I): constructing a starting sample set as
Where p is the number of feature variables and n is the number of samples, which is constructed in a form expressed by a principal component W (i ═ 1, 2.., p), as shown in equation (3):
(II): solving the principal components specifically comprises the following steps:
(1) averaging of initial sample sets
A covariance matrix S of the initial sample set;
(2) solving a characteristic equation | S- λ I | ═ 0, wherein I is an identity matrix;
(3) solving a unit characteristic vector corresponding to the characteristic value;
(4) writing a principal component expression;
(5) and setting a selection rule of the main component.
The steps are completed to generate the efficient data set, the purposes of data compression and dimension reduction are achieved, and meanwhile, the main characteristics of the data are extracted.
In the third step), according to the regression requirement in the voltage stability evaluation, directly adopting a continuity index or carrying out discretization mapping on the index again; combining ensemble learning, and simultaneously constructing a series of parallel regression trees to form an ensemble learning frame and an online voltage stability integrated evaluation model; and performing off-line training on the integrated model by using the high-efficiency data set after analysis processing, taking the generated high-efficiency data set as input, and outputting the VSM corresponding to the high-efficiency data set.
In the step four), selecting corresponding characteristic data based on the power system operation data collected by the synchronous phasor measurement unit and the wide area measurement system in real time, and performing online evaluation by using the updated evaluation model to obtain an online evaluation result.
In the step five), a corresponding regression confidence decision rule is made, so that the use of an untrusted result in the ensemble learning is avoided, the problem that the overall evaluation accuracy is influenced due to a large error of the evaluation result of a single learner is solved, and the final online evaluation result is obtained.
When using the method of confidence detection, confidence criteria are formulated for a single regression tree, as shown in equation (10):
in the formula: y isiA single evaluation value given for the ith regression tree, i 1, 2., N;
is a set of unitary evaluation values y1,...,yi,...,yN]A median of (d);
the confidence decision rule of the integrated evaluation model is as follows:
for a given N univariate model evaluations, there are W confident univariate evaluations and N-W inconclusive univariate evaluations, respectively.
If N-W is larger than or equal to T (T is smaller than or equal to N, and T is a user-defined critical value), the evaluation result is not trusted;
otherwise, the evaluation result is confidence, and the corresponding confidence evaluation result TSM, as shown in equation (11):
based on the confidence decision rule, the method can avoid using an untrusted result in the ensemble learning so as to solve the problem that the accuracy of the overall evaluation is influenced by a larger error of a single learner result and obtain a final online evaluation result.
The method for compressing and reducing the dimension of the data sample set of the power system running state information comprises the following steps when compressing and reducing the dimension of an initial sample set: (I): constructing a starting sample set as
Where p is the number of feature variables and n is the number of samples, which is constructed in a form expressed by a principal component W (i ═ 1, 2.., p), as shown in equation (3):
(II): solving the principal components specifically comprises the following steps:
(1) averaging of initial sample sets
A covariance matrix S of the initial sample set;
(2) solving a characteristic equation | S- λ I | ═ 0, wherein I is an identity matrix;
(3) solving a unit characteristic vector corresponding to the characteristic value;
(4) writing a principal component expression;
(5) setting a selection rule of the main component;
the steps are completed to generate the high-efficiency data set, and the purposes of data compression and dimension reduction are achieved.
By adopting the technical method, the technical effects that:
(1) by using the principal component analysis method, an efficient sample set is generated, the purposes of compressing and reducing the dimension of an original data set can be realized, and the time consumed by off-line training is saved;
(2) the method is based on the high-efficiency sample set, combines the integrated learning and the confidence detection, reduces the calculation burden of a single model, improves the precision of the overall evaluation model, avoids an untrusted result through the confidence detection, and further improves the accuracy of the evaluation.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a flow chart of principal component analysis proposed by the present invention;
FIG. 3 is a diagram of an IEEE 30 node system topology employed by an embodiment of the present invention;
FIG. 4 is a graph comparing the performance of four different models tested in accordance with an embodiment of the present invention;
FIG. 5 is a graph comparing training time and predicted time for four different models tested in accordance with an embodiment of the present invention.
Detailed Description
As shown in fig. 1, the method proposed by the present invention specifically includes the following steps:
step one): determining a voltage stability limit by using a continuous power flow method based on historical operation data of the power system, constructing a VSM index, establishing an initial sample set, and performing standard normalization;
step two): aiming at the high-efficiency initial sample set, a principal component analysis method is used to realize the compression and dimension reduction of the initial sample set and obtain a high-efficiency sample set;
step three): based on the high-efficiency sample set, a static voltage stability evaluation model is constructed by combining a regression tree and integrated learning, and the high-efficiency sample set is updated to complete the updating of the evaluation model;
step four): collecting power system operation data in real time based on the synchronous phasor measurement unit and the wide area measurement system, carrying out corresponding analysis processing, and carrying out real-time voltage stability evaluation by using an updated evaluation model;
step five): and evaluating the real-time voltage stability evaluation result by using a confidence detection method to obtain a final evaluation result.
In step one): from a large amount of historical operating data, a voltage stability limit (i.e., a voltage collapse point) is determined for each operating point using a continuous power flow method. And constructing a continuous VSM index according to the active power difference of the load between the working point and the voltage collapse point. The steady-state operation data characteristics of each working point correspond to a corresponding stability margin index so as to form a corresponding relation between the characteristics and the indexes.
And determining the limit of the static voltage stability of the system by adopting the continuous power flow idea to obtain a P-V curve of the system, and calculating the VSM of the power system through the P-V curve. The VSM of a power system is defined as shown in equation (1):
in the formula: Δ P is the active power margin; pmaxIs the maximum power before voltage collapse.
Normalizing the initial set to reduce the machine computation burden, as shown in equation (2):
in the formula:
the value of a certain operation variable after standard normalization; x is the number ofiIs the original value of the operating variable; x is the number ofi_minThe minimum value of the variable in the obtained sample; x is the number ofi_maxThe maximum value of the variable in the obtained sample; in this way, the values of all variables are varied from 0 to 1.
In step two), the basic idea of principal component analysis is to replace a large number of relevant variables with a small number of independent variables while preserving the information of the original variables as much as possible. As shown in fig. 2, the method specifically comprises the following steps:
(I): set initial sample set to
Where p is the number of features and n is the number of samples, which is constructed in a form expressed by a principal component W (i ═ 1, 2.., p), as shown in equation (3):
equation (3) satisfies the following requirement:
(i) the sum of the squares of the coefficients of each principal component is 1, i.e. a1i 2+a2i 2+...+api 2＝1；
(ii) The principal components being independent of each other without overlapping information, i.e. Cov (W)i,Wj)＝0,i≠j,i,j＝1,2,...,p；
(iii) The variance and importance of the principal component decrease in sequence, namely Var (W)1)≥Var(W2)≥...≥Var(WP)。
(II): solving the principal components, the concrete steps are as follows:
(1) averaging of initial sample sets
The covariance matrix of the initial sample set is obtained as shown in equation (4):
(2) solving the characteristic equation | S- λ I | ═ 0, as shown in equation (5):
solving p characteristic roots to be lambda1,λ2,...,λp(λ1≥λ2≥...≥λp)；
(3) Lambda is foundkCorresponding unit feature vector alphak( k 1, 2.. times.p), solving a system of equations (S- λ)kI) 0, as shown in formula (6):
in the formula: alpha is alphakIs a unit vector, then a1k 2+a2k 2+...+apk 2The solution obtained is shown in equation (7):
(4) writing a principal component expression as shown in equation (8):
(5) the selection rule of the main components is as follows:
defining the cumulative variance contribution of m principal components as shown in equation (9):
in application, m principal components are selected and the cumulative variance contribution ratio satisfies a certain threshold, and as the threshold approaches 1, the principal components extract more features from the sample set. And replacing the initial sample set matrix X by a matrix formed by m main components for subsequent data analysis. Therefore, the purposes of data compression and dimension reduction are achieved, and the main characteristics of the data are extracted.
In step three): according to the regression requirement in the voltage stability evaluation, directly adopting a continuity index or carrying out discretization mapping on the index again; combining ensemble learning, and simultaneously constructing a series of parallel regression trees to form an ensemble learning frame and an online voltage stability integrated evaluation model; and performing off-line training on the integrated model by using the high-efficiency data set after analysis processing, taking the generated high-efficiency data set as input, and outputting the VSM corresponding to the high-efficiency data set.
In step four): and selecting corresponding characteristic data based on the power system operation data collected by the synchronous phasor measurement unit and the wide area measurement system in real time, and finishing real-time voltage stability evaluation by using the trained corresponding voltage stability evaluation model to obtain an online evaluation result.
In step five): confidence criteria are formulated for a single regression tree, as shown in equation (10):
in the formula: y isiA single evaluation value given for the ith regression tree, i 1, 2., N;
is a set of unitary evaluation values y1,...,yi,...,yN]The median of (3).
The confidence decision rule of the integrated evaluation model is as follows:
for a given N univariate model evaluations, there are W confident univariate evaluations and N-W inconclusive univariate evaluations, respectively.
If N-W is larger than or equal to T (T is smaller than or equal to N, and T is a user-defined critical value), the evaluation result is not trusted;
otherwise, the evaluation result is confidence, and the corresponding confidence evaluation result TSM, as shown in equation (11):
based on the confidence decision rule, the method can avoid using an untrusted result in the ensemble learning so as to solve the problem that the accuracy of the overall evaluation is influenced by a larger error of a single learner result and obtain a final online evaluation result.
Example (b): the embodiment used in the present invention is based on an IEEE 30 node system as shown in fig. 3, which contains 30 nodes, 6 generators and 37 transmission lines. All the steps of the method are adopted in the test, the computer provided with an Intel Core i7 processor and an 8GB memory is used for testing, a 10-time cross validation method is adopted in the test, and all the tests are repeated for 10 times until the mean value and the standard deviation of the precision tend to be stable. Based on historical operating data and a series of simulations, a total of 4400 samples were generated for testing and evaluation.
Using residual squared error R2And Root Mean Square Error (RMSE) index, the performance of the model, R2The definition of RMSE is shown as formula (12) and formula (13):
in the formula: s is a sample set, n is the number of samples, xiIs an optimized input feature quantity, yiIs the corresponding VSM value, d (x)i) Is an evaluation value of the time of the measurement,
is yiIs measured.
In general, R2The larger the size, the better the model performance; and the smaller the RMSE, the smaller the error of the representative model, i.e., the better the performance.
In practical applications, the processing time of PMU data should be less than 0.033 seconds. In order to verify the rapid evaluation and analysis capability of the model of the invention on the online voltage stability evaluation model, as shown in table 1, the data processing speed of the IEEE 30 node test system is tested, and the result shows that the processing speed of the model of the invention can meet the requirement of online application.
TABLE 1
| Test system | Off-line training time (seconds) | Predicted time (seconds) |
| IEEE 30 node | 39.45(3960 samples) | 1.93(440 samples) |
In order to verify the robustness of the model adapting to the topological change of the power system, the topological relation of the test system is changed, a new sample is generated for testing the model, and finally the performance of the model is predicted. The test result shows that the model has good robustness for adapting to the topological change, and the aim of the invention is achieved.
TABLE 2
As shown in fig. 4, performance test results for four different models are given; as shown in FIG. 5, four are givenThe time required for testing the different models. As can be seen from FIGS. 4 and 5, the integrated regression tree model R of the present invention2Max, RMSE min, and training and prediction times are both shortest. The integrated regression tree evaluation model has the best performance compared with the other three models.
The test results show that the voltage stability evaluation model provided by the invention has good performance and can be applied to an actual power system.
Claims (10)
1. A static voltage stability evaluation method based on principal component analysis and confidence detection is characterized by comprising the following steps:
step one): based on historical operating data of the power system, establishing a Voltage Stability Margin (VSM) index, establishing an initial sample set, and performing standard normalization;
step two): aiming at the high-efficiency initial sample set, a principal component analysis method is used to realize the compression and dimension reduction of the initial sample set and obtain a high-efficiency sample set;
step three): based on the high-efficiency sample set, a static voltage stability evaluation model is constructed by combining a regression tree and integrated learning, and the high-efficiency sample set is updated to complete the updating of the evaluation model;
step four): collecting power system operation data in real time based on the synchronous phasor measurement unit and the wide area measurement system, carrying out corresponding analysis processing, and carrying out real-time voltage stability evaluation by using an updated evaluation model;
step five): and evaluating the real-time voltage stability evaluation result by using a confidence detection method to obtain a final online evaluation result.
2. The method for evaluating the stability of the static voltage based on the principal component analysis and the confidence detection as claimed in claim 1, wherein: in the first step), a continuous power flow method is used for determining a voltage stability limit for each working point according to historical operating data, continuous VSM indexes are constructed according to the active power difference value of the load between the working point and the voltage collapse point, and the steady-state operating data characteristic of each working point corresponds to a corresponding stability margin index so as to form a corresponding relation between the characteristic and the index.
3. The static voltage stability assessment method based on principal component analysis and confidence detection according to claim 2, characterized in that in step one): determining the limit of the static voltage stability of the system by adopting a continuous power flow method to obtain a P-V curve of the system, and calculating the VSM of the power system through the P-V curve; the VSM of a power system is defined as shown in equation (1):
in the formula: Δ P is the active power margin; pmaxIs the maximum power before voltage collapse.
4. The method for evaluating the stability of the static voltage based on the principal component analysis and the confidence detection as claimed in claim 1, wherein: in the first step), the initial sample set is subjected to standard normalization to reduce the computational burden of the machine, as shown in formula (2)
In the formula:
the value of a certain operation variable after standard normalization; x is the number ofiIs the original value of the operating variable; x is the number ofi_minThe minimum value of the variable in the obtained sample; x is the number ofi_maxThe maximum value of the variable in the obtained sample; in this way, the values of all variables are varied from 0 to 1.
5. The method for evaluating the stability of the static voltage based on the principal component analysis and the confidence detection as claimed in claim 1, wherein: in the second step), when compressing and dimensionality reduction is performed on the initial sample set, the method specifically comprises the following steps:
(I): constructing a starting sample set as
Where p is the number of feature variables and n is the number of samples, which is constructed in a form expressed by a principal component W (i ═ 1, 2.., p), as shown in equation (3):
(II): solving the principal components specifically comprises the following steps:
(1) averaging of initial sample sets
A covariance matrix S of the initial sample set;
(2) solving a characteristic equation | S- λ I | ═ 0, wherein I is an identity matrix;
(3) solving a unit characteristic vector corresponding to the characteristic value;
(4) writing a principal component expression;
(5) setting a selection rule of the main component;
the steps are completed to generate the efficient data set, the purposes of data compression and dimension reduction are achieved, and meanwhile, the main characteristics of the data are extracted.
6. The method for evaluating the stability of the static voltage based on the principal component analysis and the confidence detection as claimed in claim 1, wherein: in the third step), according to the regression requirement in the voltage stability evaluation, directly adopting a continuity index or carrying out discretization mapping on the index again; combining ensemble learning, and simultaneously constructing a series of parallel regression trees to form an ensemble learning frame and an online voltage stability integrated evaluation model; and performing off-line training on the integrated model by using the high-efficiency data set after analysis processing, taking the generated high-efficiency data set as input, and outputting the VSM corresponding to the high-efficiency data set.
7. The method for evaluating the stability of the static voltage based on the principal component analysis and the confidence detection as claimed in claim 1, wherein: in the step four), selecting corresponding characteristic data based on the power system operation data collected by the synchronous phasor measurement unit and the wide area measurement system in real time, and performing online evaluation by using the updated evaluation model to obtain an online evaluation result.
8. The method for evaluating the stability of a static voltage based on principal component analysis and confidence detection according to one of claims 1 to 7, wherein: in the step five), a corresponding regression confidence decision rule is made, so that the use of an untrusted result in the ensemble learning is avoided, the problem that the overall evaluation accuracy is influenced due to a large error of the evaluation result of a single learner is solved, and the final online evaluation result is obtained.
9. The method for evaluating the stability of static voltage based on principal component analysis and confidence detection according to one of claims 1 to 7, wherein when the method for confidence detection is used, a confidence criterion is established for a single regression tree, as shown in formula (10):
in the formula: y isiA single evaluation value given for the ith regression tree, i 1, 2., N;
is a set of unitary evaluation values y1,...,yi,...,yN]A median of (d);
the confidence decision rule of the integrated evaluation model is as follows:
corresponding to given N single model evaluation values, wherein W single evaluation results with confidence and N-W single evaluation results with no confidence are respectively obtained;
if N-W is larger than or equal to T (T is smaller than or equal to N, and T is a user-defined critical value), the evaluation result is not trusted;
otherwise, the evaluation result is confidence, and the corresponding confidence evaluation result TSM, as shown in equation (11):
based on the confidence decision rule, the method can avoid using an untrusted result in the ensemble learning so as to solve the problem that the accuracy of the overall evaluation is influenced by a larger error of a single learner result and obtain a final online evaluation result.
10. The method for compressing and reducing the dimension of the data sample set of the power system running state information is characterized by comprising the following steps when compressing and reducing the dimension of an initial sample set: (I): constructing a starting sample set as
Where p is the number of feature variables and n is the number of samples, which is constructed in a form expressed by a principal component W (i ═ 1, 2.., p), as shown in equation (3):
(II): solving the principal components specifically comprises the following steps:
(1) averaging of initial sample sets
A covariance matrix S of the initial sample set;
(2) solving a characteristic equation | S- λ I | ═ 0, wherein I is an identity matrix;
(3) solving a unit characteristic vector corresponding to the characteristic value;
(4) writing a principal component expression;
(5) setting a selection rule of the main component;
the steps are completed to generate the high-efficiency data set, and the purposes of data compression and dimension reduction are achieved.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010616348.9A CN111814395A (en) | 2020-06-30 | 2020-06-30 | Static voltage stability evaluation method based on principal component analysis and confidence detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010616348.9A CN111814395A (en) | 2020-06-30 | 2020-06-30 | Static voltage stability evaluation method based on principal component analysis and confidence detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111814395A true CN111814395A (en) | 2020-10-23 |
Family
ID=72855260
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010616348.9A Pending CN111814395A (en) | 2020-06-30 | 2020-06-30 | Static voltage stability evaluation method based on principal component analysis and confidence detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111814395A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106709598A (en) * | 2016-12-15 | 2017-05-24 | 全球能源互联网研究院 | One-class sample-based voltage stability prediction judgment method |
| CN110264116A (en) * | 2019-07-31 | 2019-09-20 | 三峡大学 | A kind of Electrical Power System Dynamic safety evaluation method explored based on relationship with regression tree |
-
2020
- 2020-06-30 CN CN202010616348.9A patent/CN111814395A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106709598A (en) * | 2016-12-15 | 2017-05-24 | 全球能源互联网研究院 | One-class sample-based voltage stability prediction judgment method |
| CN110264116A (en) * | 2019-07-31 | 2019-09-20 | 三峡大学 | A kind of Electrical Power System Dynamic safety evaluation method explored based on relationship with regression tree |
Non-Patent Citations (3)
| Title |
|---|
| 刘颂凯: ""基于数据与机理的电力系统安全评估研究"", 《中国博士学位论文全文数据库工程科技Ⅱ辑》, pages 042 - 80 * |
| 周宝玉: ""基于改进主成分分析的电网分区方案评价模型的研究"", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》, pages 042 - 116 * |
| 赵万明等: ""基于主成分分析和支持向量机的静态电压稳定评估"", 《华东电力》, vol. 36, no. 7, pages 1 - 2 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhu et al. | Time series shapelet classification based online short-term voltage stability assessment | |
| CN112699913A (en) | Transformer area household variable relation abnormity diagnosis method and device | |
| CN111523778A (en) | Power grid operation safety assessment method based on particle swarm algorithm and gradient lifting tree | |
| CN111401749A (en) | Dynamic safety assessment method based on random forest and extreme learning regression | |
| CN110417011A (en) | A kind of online dynamic secure estimation method based on mutual information Yu iteration random forest | |
| CN105139289A (en) | Power system transient state voltage stability evaluating method based on misclassification cost classified-learning | |
| CN110363334B (en) | Grid line loss prediction method of photovoltaic grid connection based on gray neural network model | |
| CN109378835A (en) | Based on the large-scale electrical power system Transient Stability Evaluation system that mutual information redundancy is optimal | |
| CN111814284A (en) | On-line voltage stability evaluation method based on correlation detection and improved random forest | |
| CN112684295A (en) | Power distribution network fault line selection method and system under high permeability based on similarity separation degree | |
| CN106874676B (en) | State evaluation method for electric energy metering device | |
| CN111401792A (en) | Dynamic safety assessment method based on extreme gradient lifting decision tree | |
| CN113162037B (en) | Power system transient voltage stability self-adaptive evaluation method and system | |
| CN111585277B (en) | Power system dynamic security assessment method based on hybrid integration model | |
| Fan et al. | Real-time static voltage stability assessment in large-scale power systems based on maximum-relevance minimum-redundancy ensemble approach | |
| CN109684749B (en) | Photovoltaic power station equivalent modeling method considering operating characteristics | |
| Chen et al. | Data-driven distribution network topology identification considering correlated generation power of distributed energy resource | |
| CN105701265A (en) | Double-fed wind generator modeling method and apparatus | |
| CN111814395A (en) | Static voltage stability evaluation method based on principal component analysis and confidence detection | |
| CN113177717B (en) | Quick evaluation method for toughness of power transmission system based on influence increment sensitivity | |
| CN111797919A (en) | Dynamic security assessment method based on principal component analysis and convolutional neural network | |
| CN106296461B (en) | Power grid parameter identification method based on particle swarm optimization algorithm and local state estimation | |
| Zhang et al. | Identification of critical nodes in power system based on graph theory and pagerank algorithm | |
| Chen et al. | Estimating reactive margin for determining transfer limits | |
| Ji et al. | Probabilistic optimal power flow considering the dependence of multiple wind farms using pair diffusive kernel copula |
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 |