CN111797919A - Dynamic security assessment method based on principal component analysis and convolutional neural network - Google Patents
Dynamic security assessment method based on principal component analysis and convolutional neural network Download PDFInfo
- Publication number
- CN111797919A CN111797919A CN202010616366.7A CN202010616366A CN111797919A CN 111797919 A CN111797919 A CN 111797919A CN 202010616366 A CN202010616366 A CN 202010616366A CN 111797919 A CN111797919 A CN 111797919A
- Authority
- CN
- China
- Prior art keywords
- power system
- sample set
- principal component
- dynamic
- component analysis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000013527 convolutional neural network Methods 0.000 title claims abstract description 28
- 238000000513 principal component analysis Methods 0.000 title claims abstract description 18
- 238000011156 evaluation Methods 0.000 claims abstract description 12
- 238000013210 evaluation model Methods 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims description 20
- 238000004088 simulation Methods 0.000 claims description 15
- 239000011159 matrix material Substances 0.000 claims description 8
- 230000001052 transient effect Effects 0.000 claims description 7
- 238000013144 data compression Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 6
- 238000012847 principal component analysis method Methods 0.000 claims description 6
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000005206 flow analysis Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000012423 maintenance Methods 0.000 abstract description 2
- 230000002265 prevention Effects 0.000 abstract 1
- 238000012360 testing method Methods 0.000 description 11
- 238000004364 calculation method Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000011056 performance test Methods 0.000 description 1
- 238000005381 potential energy Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Images
Classifications
-
- 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
- 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
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
-
- 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
-
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Strategic Management (AREA)
- Evolutionary Computation (AREA)
- Marketing (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Power Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
A dynamic security assessment method based on principal component analysis and a convolutional neural network specifically comprises the following steps: the method comprises the following steps: acquiring a system operation data sample, and constructing a dynamic safety index to form a corresponding initial sample set; step two: generating an efficient sample set; step three: updating the efficient sample set to complete the updating of the evaluation model; step four: and based on the real-time operation data of the power system, finishing the evaluation of the real-time dynamic safety state of the power system by using the continuously updated dynamic safety evaluation model to obtain a final online dynamic safety evaluation result. The online dynamic safety model of the power system can provide rapid, efficient and accurate prediction evaluation for the power system, is beneficial to system maintenance and safety measure prevention of power personnel, improves the safety and stability of operation of the power system, and improves the reliability of power supply.
Description
Technical Field
The invention relates to the field of dynamic security assessment of power systems, in particular to a dynamic security assessment method based on principal component analysis and a convolutional neural network.
Background
With the changing recombination of the power system, the large-scale development of the wide-area interconnection of the system and the increasing penetration of renewable energy sources (especially wind power generation), the prediction of the operation state of the power system becomes difficult, and when the power system forms the wide-area interconnection, the range affected by a large disturbance accident is enlarged, and the risk of a large power failure accident is increased. The scale of the power system is continuously enlarged, so that the operation environment of the power system is more complicated and changeable, and the potential emergency accident further increases the dynamic unsafe risk of the power grid. The problem of safe operation of power systems is therefore facing a great challenge. How to quickly and accurately evaluate the safety state of a power system becomes one of the key issues that are currently focused on by power personnel. However, due to the huge scale and various operating conditions of the actual power system, the current online dynamic safety assessment method is increasingly difficult to meet the actual requirements. Therefore, the research on the dynamic safety assessment method with high adaptability and high precision has great significance for the construction and development of modern power systems.
At present, research on dynamic safety assessment of a power system mainly comprises two aspects, namely a method based on mechanism analysis and a method based on data driving. The method based on the mechanism analysis comprises a direct method and a time domain simulation, wherein the methods comprise a potential energy boundary method, an extended equal area method, a leading unbalanced method, a transient energy method, an extended equal area method and the like. The time domain simulation method is based on a large-scale nonlinear differential equation, and dynamic safety indexes of the system are constructed by solving the differential equation. With the rapid development of wide area measurement systems and the wide application of synchrophasor measurement units in power systems, some data-driven methods have been applied. Such as Decision Trees (DTs), Artificial Neural Networks (ANNs), Extreme Learning Machines (ELMs), Support Vector Machines (SVMs), and the like. However, the existing dynamic security assessment method for the power system still has some disadvantages such as:
(1) the traditional mechanism analysis method mainly depends on off-line calculation, wherein a time domain simulation method depends on the accuracy of modeling, and has huge calculation amount and long calculation time; due to the continuous enlargement of the scale of the modern power system, the method is difficult to analyze large-scale samples, and is difficult to meet the requirements of real-time dynamic safety evaluation on the calculation speed, cannot provide stability margin information and other defects;
(2) when the traditional data driving method is applied to dynamic security assessment of a power system, various influence factors possibly existing in actual power grid operation, such as efficiency problems of a training sample set, are not considered, assessment results are not evaluated, and visualization cannot be provided for dynamic security information. Meanwhile, the training time is too long, and the method is difficult to be applied to large-scale data. Some unexpected situations often occur in an actually operating power system, and the traditional dynamic safety evaluation model is difficult to evaluate the unexpected situations.
As described above, the conventional method has difficulty in meeting the requirements of dynamic safety assessment of modern power systems, and a real-time assessment method capable of meeting high adaptability and high precision is urgently needed.
Disclosure of Invention
The invention aims to provide a dynamic safety assessment method which is beneficial to improving assessment speed and prediction precision, and is beneficial to system operators to take preventive control measures in time, so that the operation stability and the power supply reliability of a power system are improved.
The purpose of the invention is realized as follows:
a dynamic security assessment method based on principal component analysis and convolutional neural network comprises the following steps:
step one): based on historical operating data of the power system and simulation of system faults, obtaining a system operating data sample, and constructing a dynamic safety index to form a corresponding initial sample set;
step two): aiming at the initial sample set, a principal component analysis method is used for achieving the purposes of data compression and dimension reduction, and an efficient sample set is generated;
step three): inputting a high-efficiency sample set, training a Convolutional Neural Network (CNN) to obtain a dynamic security assessment model of the power system, and updating the high-efficiency sample set to complete the updating of the assessment model;
step four): and based on the real-time operation data of the power system, finishing the evaluation of the real-time dynamic safety state of the power system by using the continuously updated dynamic safety evaluation model to obtain a real-time dynamic safety evaluation result.
In the first step), power flow analysis and time domain simulation are carried out based on historical operating data and an expected accident set of the power system, and therefore an initial sample set is obtained.
When Time domain simulation is carried out, obtaining the Critical Clearing Time (CCT) of each fault position in each running state, determining the CCT corresponding to each fault position, and when the CCT is greater than ACT, establishing a transient state safety index TSM, as shown in formula (1):
in the formula: CCT (China telecom computing) coreiThe limit cutting time of a certain position of the power system under the accident i is set; ACTiThe actual cutting time of the fault point under the accident i is taken as the actual cutting time; TSMiA transient safety margin for the location; the definition of TSM is shown in formula (2):
in step one), the initial sample set is normalized according to the formula (3):
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 the principal component analysis method is used, the method specifically includes the following steps:
(I): constructing the initial sample set obtained in the step one) into Where p is the number of feature variables and n is the number of samples, which is constructed in the form expressed by the principal component W (i ═ 1, 2.., p), as shown in equation (4):
(II): solving the principal components specifically comprises the following steps:
(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 a high-efficiency sample set, so that the purposes of data compression and dimension reduction are achieved, meanwhile, the purpose of extraction is achieved, and the high-efficiency sample set is generated;
and in the third step), obtaining a high-efficiency sample set based on principal component analysis, training the CNN by combining the TSM corresponding to each feature, and obtaining a mapping relation between the key features and the corresponding TSM, so as to construct a CNN-based power system dynamic security model.
In the third step), in the operation of the power system, when an unexpected factor is met, a model updating step is performed so as to continuously update the dynamic security assessment model of the power system, and after an updated sample set is obtained, the offline training model is used for continuing training so as to obtain the updated dynamic security assessment model.
And in the step four), acquiring the operation variables of the power system in real time by using the synchronous phasor measurement unit and the wide area monitoring system, predicting the dynamic safety state of the power system by using the updated dynamic safety assessment model based on real-time data, and obtaining an online dynamic safety assessment result.
By adopting the technical scheme, the technical effects that can be achieved are as follows:
(1) by using the principal component analysis method, the original data set can be compressed and dimension reduced, and the time consumed by off-line training is saved;
(2) according to the method, the dynamic security assessment model is constructed based on the efficient sample set and the CNN, so that the calculation burden is reduced, and the accuracy of the overall assessment model is further improved;
(3) the dynamic security assessment model constructed by the invention also considers some influence factors based on the possible situations in the actual power system operation, so as to update the model, and can provide better robustness for the model.
Drawings
The invention is further illustrated by the following examples in conjunction with the accompanying drawings:
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a principal component analysis flow chart proposed by the present invention;
FIG. 3 is an online dynamic security assessment model proposed by the present invention;
FIG. 4 is a graph comparing the performance of 5 different models in accordance with example 5 of the present invention;
FIG. 5 is a graph comparing training time and prediction time for 5 different models in accordance with the present invention.
Detailed Description
A dynamic security assessment method based on principal component analysis and convolutional neural network is disclosed, as shown in FIG. 1, which specifically comprises the following steps:
step one): based on historical operating data of the power system and simulation of a series of faults of the system, acquiring a system operating data sample, and constructing a dynamic safety index to form a corresponding initial sample set;
step two): aiming at the initial sample set, a principal component analysis method is used for achieving the purposes of data compression and dimension reduction, and an efficient sample set is generated;
step three): inputting a high-efficiency sample set, training CNN to obtain a dynamic security assessment model of the power system, and updating the high-efficiency sample set to complete the updating of the assessment model;
step four): and based on the real-time operation data of the power system, finishing the evaluation of the real-time dynamic safety state of the power system by using the continuously updated dynamic safety evaluation model to obtain a real-time dynamic safety evaluation result.
In step one): based on historical operating data and an expected accident set of the power system, detailed power flow analysis and time domain simulation are carried out, and therefore an initial sample set is obtained.
And performing time domain simulation by using PSS/E software to obtain the CCT of each fault position in each running state. And determining the CCT corresponding to each fault position through a series of simulations. Generally, when the CCT is greater than ACT, the operating state of the system is judged to be safe. Therefore, the constructed transient safety index, i.e., TSM, is shown in equation (1):
in the formula: CCT (China telecom computing) coreiThe limit cutting time of a certain position of the power system under the accident i is set; ACTiThe actual cutting time of the fault point under the accident i is taken as the actual cutting time; TSMiIs the transient safety margin for that location. The definition of TSM is shown in formula (2):
in the simulation, the stability of the system is judged by checking whether the maximum rotor angle deviation of any two generators exceeds 360 degrees.
To reduce the computational burden on the machine, the initial sample set is normalized by the standard, as shown in equation (3):
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 ofiMin is the minimum value of the variable in the acquired 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): principal component analysis has the basic idea of replacing a large number of relevant variables with a small set of independent variables while preserving as much information as possible of the original variables. As shown in fig. 2, the method specifically comprises the following steps:
(I): setting the initial sample set obtained in the step one) as Where p is the number of features and n is the number of samples, which is constructed in the form expressed by the principal component W (i ═ 1, 2.. multidot., p), as shown in equation (4):
equation (4) 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 setsAnd (3) solving a covariance matrix of the initial sample set, wherein the covariance matrix is shown in a formula (5):
(2) solving the characteristic equation | S- λ I | ═ 0, as shown in equation (6):
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), by solving a system of equations (S- λ)kI) Obtained as 0, as shown in formula (7)
In the formula: alpha is alphakIs a unit vector, then a1k 2+a2k 2+...+apk 2The solution obtained is shown in equation (8):
(4) writing a principal component expression as shown in equation (9):
(5) the selection rule of the main components is as follows:
defining the cumulative variance contribution ratio β of m principal components as shown in equation (10):
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): and obtaining a high-efficiency sample set based on principal component analysis, training the CNN by combining the TSM corresponding to each feature, and obtaining a mapping relation between the key feature and the corresponding TSM so as to construct a CNN-based power system dynamic security model. In the actual operation of the power system, some unexpected factors are often involved, including: emergency, power grid maintenance plan, economic dispatch, wave crest/wave trough change, load characteristic change and generator/load power distribution change; therefore, the offline training cannot cover all possible operating states, and a model updating step is required, so that the dynamic safety assessment model of the power system is continuously updated. And obtaining an updated sample set according to some unexpected factors, and continuing training by using the offline training model to obtain an updated dynamic security assessment model.
In step four): the flow of the online evaluation model provided by the invention is shown in fig. 3, the synchronous phasor measurement unit and the wide area monitoring system are used for acquiring the operation variables of the power system in real time, the dynamic safety state of the power system is predicted by using the updated dynamic safety evaluation model based on real-time data, and the final online evaluation result is obtained.
Example (b):
the tests were performed in the present invention using an IEEE 39 node system comprising 39 nodes, 10 generators and 46 transmission lines. This test was performed on a computer equipped with an Intel Core i7 processor and 8GB memory. The system adopts the dynamic security assessment scheme based on the invention, and the performance of the dynamic security assessment model is checked. The test adopts a 10-time cross validation method, and each test is repeated for 10 times until the average value and the standard deviation of the precision tend to be stable. Based on historical operating data and a seriesColumn simulation, resulting in 4800 samples for training and testing. Using residual squared error R2And Root Mean Square Error (RMSE) index, the performance of the model, R2The definition of RMSE is shown as formula (11) and formula (12):
in the formula: s is a sample set, n is the number of samples, xiIs an optimized input feature quantity, yiIs the corresponding TSM value, d (x)i) Is an evaluation value of the time of the measurement,is yiIs measured.
Training and testing used 5 different models, and 4 other models included: logistic Regression (LR), SVM, Regression Tree (RT), and ANN. As shown in fig. 4, performance test results for 5 different models are given; as shown in fig. 5, the time consumed by 5 different model tests is given.
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. As can be seen from FIG. 4, the CNN model R2At maximum, RMSE is minimal and it can be seen from fig. 5 that both the CNN model training time and the prediction time are minimal. The CNN evaluation model showed the best performance compared with the other 4 models.
In order to verify the influence of the topology change on the system operation and the robustness of the model of the invention for adapting to the topology change of the power system when the power system actually operates, the test changes some topology relations of the IEEE 39 node test system. The new sample generated after the change is used for testing, as shown in table 1, different network topologies and test results are given.
TABLE 1
The result shows that the model has better robustness for adapting to the topological change and can be applied to the actual operation of the electric power system.
Claims (7)
1. A dynamic security assessment method based on principal component analysis and convolutional neural network is characterized by comprising the following steps:
step one): based on historical operating data of the power system and simulation of system faults, obtaining a system operating data sample, and constructing a dynamic safety index to form a corresponding initial sample set;
step two): aiming at the initial sample set, a principal component analysis method is used for achieving the purposes of data compression and dimension reduction, and an efficient sample set is generated;
step three): inputting a high-efficiency sample set, training a Convolutional Neural Network (CNN) to obtain a dynamic security assessment model of the power system, and updating the high-efficiency sample set to complete the updating of the assessment model;
step four): and based on the real-time operation data of the power system, finishing the evaluation of the real-time dynamic safety state of the power system by using the continuously updated dynamic safety evaluation model to obtain a real-time dynamic safety evaluation result.
2. The dynamic security assessment method based on principal component analysis and convolutional neural network as claimed in claim 1, wherein: in the first step), carrying out power flow analysis and time domain simulation based on historical operating data and an expected accident set of the power system so as to obtain an initial sample set;
when Time domain simulation is carried out, obtaining the Critical Clearing Time (CCT) of each fault position in each running state, determining the CCT corresponding to each fault position, and when the CCT is greater than ACT, establishing a transient state safety index TSM, as shown in formula (1):
in the formula: CCT (China telecom computing) coreiThe limit cutting time of a certain position of the power system under the accident i is set; ACTiThe actual cutting time of the fault point under the accident i is taken as the actual cutting time; TSMiA transient safety margin for the location; the definition of TSM is shown in formula (2):
3. the dynamic security assessment method based on principal component analysis and convolutional neural network as claimed in claim 1, wherein: in step one), the initial sample set is normalized according to the formula (3):
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.
4. The dynamic security assessment method based on principal component analysis and convolutional neural network of claim 1, wherein: in the second step), when the principal component analysis method is used, the method specifically includes the following steps:
(I): constructing the initial sample set obtained in the step one) into Where p is the number of feature variables and n is the number of samples, which is constructed in the form expressed by the principal component W (i ═ 1, 2.., p), as shown in equation (4):
(II): solving the principal components specifically comprises the following steps:
(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 sample set, so that the purposes of data compression and dimension reduction are achieved, meanwhile, the purpose of extraction is achieved, and the high-efficiency sample set is generated.
5. The dynamic security assessment method based on principal component analysis and convolutional neural network as claimed in claim 1, wherein: and in the third step), obtaining a high-efficiency sample set based on principal component analysis, training the CNN by combining the TSM corresponding to each feature, and obtaining a mapping relation between the key features and the corresponding TSM, so as to construct a CNN-based power system dynamic security model.
6. The dynamic security assessment method based on principal component analysis and convolutional neural network as claimed in claim 1 or 5, wherein: in the third step), in the operation of the power system, when an unexpected factor is met, a model updating step is performed so as to continuously update the dynamic security assessment model of the power system, and after an updated sample set is obtained, the offline training model is used for continuing training so as to obtain the updated dynamic security assessment model.
7. The dynamic security assessment method based on principal component analysis and convolutional neural network as claimed in claim 1 or 7, wherein: and in the step four), acquiring the operation variables of the power system in real time by using the synchronous phasor measurement unit and the wide area monitoring system, predicting the dynamic safety state of the power system by using the updated dynamic safety assessment model based on real-time data, and obtaining an online dynamic safety assessment result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010616366.7A CN111797919A (en) | 2020-06-30 | 2020-06-30 | Dynamic security assessment method based on principal component analysis and convolutional neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010616366.7A CN111797919A (en) | 2020-06-30 | 2020-06-30 | Dynamic security assessment method based on principal component analysis and convolutional neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111797919A true CN111797919A (en) | 2020-10-20 |
Family
ID=72809773
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010616366.7A Pending CN111797919A (en) | 2020-06-30 | 2020-06-30 | Dynamic security assessment method based on principal component analysis and convolutional neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111797919A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112396344A (en) * | 2020-11-30 | 2021-02-23 | 天津大学 | Chemical process reliability online evaluation method based on product quality |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109033702A (en) * | 2018-08-23 | 2018-12-18 | 国网内蒙古东部电力有限公司电力科学研究院 | A kind of Transient Voltage Stability in Electric Power System appraisal procedure based on convolutional neural networks CNN |
CN109886604A (en) * | 2019-03-13 | 2019-06-14 | 华北电力大学 | A kind of transient stability evaluation in power system method based on one-dimensional convolutional neural networks |
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 |
CN110879917A (en) * | 2019-11-08 | 2020-03-13 | 北京交通大学 | Electric power system transient stability self-adaptive evaluation method based on transfer learning |
CN111046341A (en) * | 2019-12-12 | 2020-04-21 | 重庆地质矿产研究院 | Unconventional natural gas fracturing effect evaluation and capacity prediction method based on principal component analysis |
CN111242082A (en) * | 2020-01-21 | 2020-06-05 | 扬州大学 | Face super-resolution reconstruction identification method based on fractional order orthogonal partial least square |
CN111241587A (en) * | 2020-01-21 | 2020-06-05 | 恒瑞通(福建)信息技术有限公司 | Data desensitization method and device |
CN111275624A (en) * | 2020-01-21 | 2020-06-12 | 扬州大学 | Face image super-resolution reconstruction recognition method based on multi-set canonical correlation analysis |
-
2020
- 2020-06-30 CN CN202010616366.7A patent/CN111797919A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109033702A (en) * | 2018-08-23 | 2018-12-18 | 国网内蒙古东部电力有限公司电力科学研究院 | A kind of Transient Voltage Stability in Electric Power System appraisal procedure based on convolutional neural networks CNN |
CN109886604A (en) * | 2019-03-13 | 2019-06-14 | 华北电力大学 | A kind of transient stability evaluation in power system method based on one-dimensional convolutional neural networks |
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 |
CN110879917A (en) * | 2019-11-08 | 2020-03-13 | 北京交通大学 | Electric power system transient stability self-adaptive evaluation method based on transfer learning |
CN111046341A (en) * | 2019-12-12 | 2020-04-21 | 重庆地质矿产研究院 | Unconventional natural gas fracturing effect evaluation and capacity prediction method based on principal component analysis |
CN111242082A (en) * | 2020-01-21 | 2020-06-05 | 扬州大学 | Face super-resolution reconstruction identification method based on fractional order orthogonal partial least square |
CN111241587A (en) * | 2020-01-21 | 2020-06-05 | 恒瑞通(福建)信息技术有限公司 | Data desensitization method and device |
CN111275624A (en) * | 2020-01-21 | 2020-06-12 | 扬州大学 | Face image super-resolution reconstruction recognition method based on multi-set canonical correlation analysis |
Non-Patent Citations (1)
Title |
---|
张连刚 著: "《省域流通产业竞争力评价 体系构建、实证分析与提升路径》", 30 June 2014, 厦门:厦门大学出版社, pages: 93 - 94 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112396344A (en) * | 2020-11-30 | 2021-02-23 | 天津大学 | Chemical process reliability online evaluation method based on product quality |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110311376B (en) | Dynamic safety assessment comprehensive model and space-time visualization method for power system | |
CN111652479B (en) | Data driving method for dynamic security assessment of power system | |
CN106786560A (en) | A kind of power system stability characteristic automatic extraction method and device | |
CN110417011A (en) | A kind of online dynamic secure estimation method based on mutual information Yu iteration random forest | |
CN104392056A (en) | Direct-driven wind turbine generator modeling method and device | |
CN111401476A (en) | Transient state safety evaluation method based on boundary region importance sampling and kernel vector machine | |
CN111401792A (en) | Dynamic safety assessment method based on extreme gradient lifting decision tree | |
CN109378835A (en) | Based on the large-scale electrical power system Transient Stability Evaluation system that mutual information redundancy is optimal | |
CN114997566A (en) | Power grid blocking risk assessment method and system considering node connectivity loss | |
Ren et al. | A super-resolution perception-based incremental learning approach for power system voltage stability assessment with incomplete PMU measurements | |
CN113128005B (en) | Power grid node clustering and frequency safety analysis method and system based on dynamic characteristics | |
Ma et al. | Impact analysis of wind generation on voltage stability and system load margin | |
CN105701265A (en) | Double-fed wind generator modeling method and apparatus | |
CN111797919A (en) | Dynamic security assessment method based on principal component analysis and convolutional neural network | |
CN106410862A (en) | Wind power plant single machine equivalent method based on active recovery slope correction | |
CN111475915B (en) | Successive fault online evaluation method based on fault probability and time domain simulation quasi-steady state | |
He et al. | Power system frequency situation prediction method based on transfer learning | |
Zhu et al. | Efficient identification of critical load model parameters affecting power system voltage stability | |
He et al. | A method for transient stability assessment based on pattern recognition | |
Li et al. | Frequency prediction after disturbance of grid-connected wind power systems based on WOA and Attention-LSTM | |
Song et al. | Risk-based dynamic security assessment under typhoon weather for power transmission system | |
Khalel et al. | Dynamic security assessment for the power system in the presence of wind turbines | |
Voloshin et al. | Development of an algorithm for generating the equivalent power system according to PMU | |
Ye et al. | Switching Convolution of Node Graph and Line Graph-driven Method for Fast Static Security Analysis | |
Cheng et al. | Reliability evaluation of wind integrated composite power system based on third-order polynomial normal transformation |
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 |