CN109711435A - A kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm - Google Patents
A kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm Download PDFInfo
- Publication number
- CN109711435A CN109711435A CN201811468216.5A CN201811468216A CN109711435A CN 109711435 A CN109711435 A CN 109711435A CN 201811468216 A CN201811468216 A CN 201811468216A CN 109711435 A CN109711435 A CN 109711435A
- Authority
- CN
- China
- Prior art keywords
- voltage stability
- genetic algorithm
- support vector
- sample data
- vector machines
- 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
- 238000012706 support-vector machine Methods 0.000 title claims abstract description 53
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 35
- 230000002068 genetic effect Effects 0.000 title claims abstract description 35
- 238000012544 monitoring process Methods 0.000 title claims abstract description 30
- 238000012549 training Methods 0.000 claims abstract description 25
- 238000012360 testing method Methods 0.000 claims abstract description 20
- 238000005259 measurement Methods 0.000 claims abstract description 17
- 210000000349 chromosome Anatomy 0.000 claims description 18
- 238000002790 cross-validation Methods 0.000 claims description 7
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 108090000623 proteins and genes Proteins 0.000 claims description 6
- 238000001514 detection method Methods 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 3
- 230000008774 maternal effect Effects 0.000 claims description 3
- 230000035772 mutation Effects 0.000 claims description 3
- 238000011056 performance test Methods 0.000 claims description 3
- 238000005215 recombination Methods 0.000 claims description 3
- 230000006798 recombination Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000013480 data collection Methods 0.000 claims 1
- 238000004458 analytical method Methods 0.000 abstract description 6
- 238000013528 artificial neural network Methods 0.000 abstract description 6
- 238000010801 machine learning Methods 0.000 abstract description 5
- 230000007547 defect Effects 0.000 abstract description 3
- 239000011159 matrix material Substances 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 6
- 238000010276 construction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006641 stabilisation Effects 0.000 description 3
- 238000011105 stabilization Methods 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 3
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 2
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Abstract
A kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm, including passing through network system phasor measurement units (Phasor Measurement Unit, PMU) device obtains initial data, and is calculated using these data by conventional Load Flow to determine Network Voltage Stability margin index;The sample data of support vector machines is generated in conjunction with the PMU primary data obtained and Network Voltage Stability margin index, and sample data is divided into training sample data and test sample data, the present invention be directed to the singularity that traditional maximum defect of Voltage Stability Analysis method is the Jacobian matrix at maximum load points, and it is computationally intensive, application on site ability is poor;And new machine learning techniques such as ANNs is limited to training time amount and learning parameter value and proposes when handling nonlinear regression problem.
Description
Technical field
The invention belongs to field of power systems, and in particular to a kind of support vector machines on-Line Voltage based on genetic algorithm is steady
Qualitative monitoring method.
Background technique
With the rapid development of electric power network technique, the power grid connection of today has constituted the big system of a height interconnection.
Large-scale power system is made to have more complexity in monitoring and operation in this way.Because the interference of certain a part of system may influence
Entire electric system.Electric system observability is electric system real-time monitoring, protection and the necessary condition of control, it influences whole
A wide area monitoring and protecting and control system (Wide Area Monitoring Protection and Control Systems,
Performing effectively WAMPC).In recent years, collapse of voltage is the main reason for global many electric system have a power failure.Traditional voltage is steady
Setting analysis method is depended on using conventional Load flow calculation such as Gauss-Saden that (Gauss-Seidel) or newton-pressgang
Static Analysis Methods such as gloomy (Newton-Raphson).The major defect of these technical methods is the Ya Ke at maximum load points
Than the singularity of matrix.In addition, such as artificial neural network (Artificial Neural Network, ANN), fuzzy logic,
The machine learning techniques such as pattern-recognition, support vector machines (Support Vector Machine, SVM) are already used to carry out electricity
Force system analysis.For example apply ANN model in on-line voltage security assessment, use radial basis function (Radial
Basis Function, RBF) network estimates power system voltage stabilization grade under emergency case, Hashemi becomes with small echo
It brings and extracts voltage waveform characteristic to estimate voltage stability margin etc..
Summary of the invention
It is the surprise of the Jacobian matrix at maximum load points for traditional maximum defect of Voltage Stability Analysis method
The opposite sex, and it is computationally intensive, application on site ability is poor;And new machine learning techniques such as ANNs is asked in processing nonlinear regression
When topic, it is limited to training time amount and learning parameter value, the invention proposes a kind of support vector machines based on genetic algorithm to exist
The method of line voltage STABILITY MONITORING.
The purpose of invention is achieved in that
A kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm, comprising the following steps:
Step 1: it is obtained by network system phasor measurement units (Phasor Measurement Unit, PMU) device
Initial data, and calculated using these data by conventional Load Flow to determine Network Voltage Stability margin index;
Step 2: the sample of support vector machines is generated in conjunction with the PMU primary data obtained and Network Voltage Stability margin index
Notebook data, and sample data is divided into training sample data and test sample data;
Step 3: supporting vector machine model of the building based on genetic algorithm, and use genetic algorithm (Genetic
Algorithm, GA) find optimal Network Voltage Stability margin index parameter;
Step 4: support vector machines (Support is constructed by the optimal Network Voltage Stability margin index parameter searched out
Vector Machine, SVM) model, and combined training sample is trained SVM model;
Step 5: after training, performance test sample data tests the SVM model after training;
Step 6: after test passes through, the support vector machines (Genetic with gained SVM model construction based on genetic algorithm
Algorithm based Support Vector Machine, GA-SVM) model online Network Voltage Stability monitoring model.
In step 1, acquisition and PMU uncertainties model including voltage stability margin index.
The data of PMU measurement mainly include voltage magnitude and phase angle.
In step 2, present operating point is calculated in conventional Load Flow voltage magnitude, voltage phase angle, wattful power are carried out
Rate, reactive power, the new data being then calculated using these constitute sample data required for VSM, and by sample number
According to being divided into training sample data and test sample data by a certain percentage.
In step 3, specifically includes the following steps:
(1) parameter of GA-SVM is respectively included into regularization parameter C, RBF kernel function bandwidth σ2And insensitive loss letter
Number pipe radius ε carries out coding and generates chromosome x, then chromosome x indicates are as follows: X={ x1,x2,x3, wherein x1、x2And x3Table respectively
Show GA-SVM parameter C, σ2And ε.(may I ask and mark red mean herein?)
(2) the SVM parameter selection optimized with Cross-Validation technique, in K- folding cross validation, by data set
Be randomly divided into the subset of K parts of equivalent, and use wherein (K-1) part subset data SVM regression model, SVM are established as training set
The performance of parameter is tested by k-th sub collection, is constantly repeated the above process, so that each subset is carried out as test subset
Primary test.
(3) initial population has 20 genomes being randomly generated at based between convergence time and population diversity
Balance selects initial population size, and calculates the adaptive value for the chromosome that each is randomly generated.
(4) new population is created to replace existing population with modes such as selection, intersection and mutation, then pass through roulette
Mode by adaptive value preferably new chromosome put into recombination pond in.The gene of male parent and maternal chromosome swaps, and obtains
Take new offspring;
(5) (3) step is repeated to (4) step, this process is repeated until that gene dosage reaches limit value.
In step 4, specifically includes the following steps: with optimal GA-SVM parameter C, σ obtained in step 32And ε,
GA-SVM model is constructed, and is trained using training sample data, for the sample data of different network systems, is obtained not
Same GA-SVM model of fit.
In step 5, the fitness of the GA-SVM model of test sample data detection building is utilized.
In step 6, the online Network Voltage Stability monitoring model based on GA-SVM model passes through actual electric network sample number
According to inspection, and complete prediction to Network Voltage Stability nargin.
By adopting the above technical scheme, following technical effect can be brought:
1) electric system that the present invention relies on is that have the big regional power grid system of various forms of electric power energy accesses, is adopted
Sample data have also fully considered the uncertain of PMUs acquisition data from all PMUs units of entire big regional power grid system
Property to GA-SVM model online Network Voltage Stability monitoring influence, establish PMU uncertainty models.According to ieee standard
C37-118-1 specifications of surveys guarantees the synchro measure error of PMU less than 1% using the synchronous vector standard of TVE.
2) SVM machine learning techniques are used in method provided by the invention, support vector machines (SVM) is a kind of new function
Powerful machine learning techniques, its VC dimension based on Statistical Learning Theory is theoretical and structural risk minimization principle, SVM are established
Optimum network structure can be very good to find equalization point in experience error and VC confidence interval.The present invention is well by SVM technology
It applies in the on-Line Voltage monitoring system of electric system, there is good timeliness, and calculate the time to obtain further
Optimization.
3) in method provided by the invention, advantage is to be realized using GA genetic algorithm to the optimal of SVM modeling parameters
Change, the BP neural network (Multilayer perceived compared to grid-search (Grid Search, GS) method and multilayer maincenter
Perceptron-back Propagation Neural Network, MLP-BPNN), GA method is returning operation of receiver spy
Linearity curve (Regression Receiver Operating Characteristic, RROC) and excessively RROC song domain curve
(Area Over the RROC-AOC) has more superior performance and shorter calculating time.
Detailed description of the invention
Present invention will be further explained below with reference to the attached drawings and examples:
Fig. 1 is the online Network Voltage Stability monitoring model flow chart in the present invention based on GA-SVM;
Fig. 2 is actual P-V curve graph in the specific embodiment of the invention.
Specific embodiment
A kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm, comprising the following steps:
Step 1: it is obtained by network system phasor measurement units (Phasor Measurement Unit, PMU) device
Initial data, and calculated using these data by conventional Load Flow to determine Network Voltage Stability margin index;
Step 2: the sample of support vector machines is generated in conjunction with the PMU primary data obtained and Network Voltage Stability margin index
Notebook data, and sample data is divided into training sample data and test sample data;
Step 3: supporting vector machine model of the building based on genetic algorithm, and use genetic algorithm (Genetic
Algorithm, GA) find optimal Network Voltage Stability margin index parameter;
Step 4: support vector machines (Support is constructed by the optimal Network Voltage Stability margin index parameter searched out
Vector Machine, SVM) model, and combined training sample is trained SVM model;
Step 5: after training, performance test sample data tests the SVM model after training;
Step 6: after test passes through, the support vector machines (Genetic with gained SVM model construction based on genetic algorithm
Algorithm based Support Vector Machine, GA-SVM) model online Network Voltage Stability monitoring model.
In step 1, including two voltage stability margin index, PMU uncertainties model aspects, it is described as follows respectively:
(1) main target of voltage stability margin index, Network Voltage Stability analysis is determining network system work at present
Whether point is stable, if meets each operation standard.And voltage power curve (as shown in Figure 2) can be used to obtain the stabilization of power grids
Nargin.Assuming that the active power that present operating point passes to load is Pc, and maximum active power is PM, then voltage stability margin
(VSM) it can indicate are as follows:
VSMi=PM, i-PC, i, i=1,2 ..., l (1)
Wherein: l indicates that total load branch of the network system, i indicate wherein i-th branch.
Then Network Voltage Stability margin index can convert are as follows:
VSMIndexChange between 0-1, as shown in Figure 2.When collapse of voltage point is shifted in operating point, defined by formula 2
VSMIndexClose to 0.In order to determine collapse of voltage point, PV curve is got by continuous tide technology (CPF), because of continuous tide
Technology may include PV curve work cusp (i.e. collapse of voltage point).
(2) PMU uncertainties model, the data of PMU measurement mainly include voltage magnitude and phase angle, are used to VSMIndexInto
Row prediction.Although the measurement accuracy of PMU is very high, the possibility of measurement error is constantly present in measurement process, according to IEEE
Standard C37-118-1 specifications of surveys, these uncertainties are in synchronized phase measurement, it is necessary to consider vector overall error (Total
Vector Error), vector overall error (TVE) is a very important standard in synchronized phase measurement, and value is in stabilization
It should be less than 1% when state.Here vector overall error TVE can be indicated are as follows:
Wherein: xr *And xi *Respectively indicate the real and imaginary parts of measured value, xrAnd xiIndicate the real and imaginary parts of ideal value.
By actual voltage magnitude error delta ViWith vector error Δ θiVector overall error formula TVE is imported, in this way selection Δ
ViWith Δ θiSo that TVE≤1%.
In step 2 shown in Fig. 1, the sample data, the data obtained first by PMUs in step I shown in Fig. 1,
Carry out present operating point is calculated in conventional Load Flow voltage magnitude, voltage phase angle, active power and reactive power.Then sharp
The new data being calculated with these constitute sample data required for VSM, and sample data are divided into instruction by a certain percentage
Practice sample data and test sample data.
In step 3 shown in Fig. 1, the GA-SVM model construction and GA-SVM parameter optimization divide following several steps complete
At:
(1) parameter of GA-SVM is respectively included into regularization parameter C, RBF kernel function bandwidth σ2And insensitive loss letter
Number pipe radius ε carries out coding and generates chromosome x, then chromosome x indicates are as follows:
X={ x1,x2,x3, wherein x1、x2And x3Respectively indicate GA-SVM parameter C, σ2And ε.
(2) over-fitting or poor fitting of GA-SVM model in order to prevent, optimizes with Cross-Validation technique
Data set is randomly divided into the subset of K parts of equivalent in K- folding cross validation by SVM parameter selection, and with wherein (K-1) one's share of expenses for a joint undertaking
Collection data establish SVM regression model as training set.The performance of SVM parameter is tested by k-th sub collection.It constantly repeats above-mentioned
Process, so that each subset is once tested as test subset.Cross validation, fitting are rolled over using 5- in the present invention
Function can indicate are as follows:
Minf=MAPEcross_val (4)
Wherein: m is sample data training set, AiActual value and PiIt is predicted value, MAPEcross_valSolution it is smaller, it is more organic
It can survive in the next generation.
(3) initial population has 20 genomes being randomly generated at based between convergence time and population diversity
Balance selects initial population size, and the adaptive value for the chromosome that each is randomly generated is calculated by formula (4) and (5).
(4) new population is created to replace existing population with modes such as selection, intersection and mutation, then pass through roulette
Mode by adaptive value preferably new chromosome put into recombination pond in.The gene of male parent and maternal chromosome swaps, and obtains
New offspring is taken to be desirably to obtain better solution, a possibility that each group chromosome creates new chromosome, is set as 0.8,
The probability for changing chromosome coding by being mutated is set as 0.05.
For (5) (3) steps to (4) step, this process is repeated until that gene dosage reaches limit value.
In order to completely examine the performance of GA-SVM model, MAPE, NMSE and WIA above-mentioned as measurement index, MAPE's
Be worth it is smaller, indicate predicted value and actual value it is closer;WIA characterizes recurrence degree, it changes between 0-1, indicates pre- closer to 1
Measured value is more accurate.
In step 4 shown in Fig. 1, with optimal GA-SVM parameter C, σ obtained in step 3 shown in Fig. 12And ε, building
GA-SVM model, and be trained using training sample data, so that GA-SVM model has better fitness.For difference
Network system sample data, available different GA-SVM model of fit is as shown in table 1.
Table 1
In step 5 shown in Fig. 1, using the fitness of the GA-SVM model of test sample data detection building, situation is examined
As shown in table 2.
Table 2
In step 6 shown in Fig. 1, the online Network Voltage Stability monitoring model based on GA-SVM model passes through actual electric network
The inspection of sample data, and the prediction to Network Voltage Stability nargin is completed, it can be seen by the comparison of the table in table 3 and table 4
The constructed online Network Voltage Stability monitoring model based on GA-SVM model is better than other existing voltage monitoring models out.
Table 3
Table 4
Claims (8)
1. a kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm, which is characterized in that including following
Step:
Step 1: initial data are obtained by network system phasor measurement units device, and pass through routine using these data
Load flow calculation is to determine Network Voltage Stability margin index;
Step 2: the primary data and Network Voltage Stability margin index obtained in conjunction with phasor measurement units generates support vector machines
Sample data, and sample data is divided into training sample data and test sample data;
Step 3: supporting vector machine model of the building based on genetic algorithm, and optimal network voltage is found with genetic algorithm
Stability margin index parameter;
Step 4: supporting vector machine model is constructed by the optimal Network Voltage Stability margin index parameter searched out, and is combined
Training sample is trained supporting vector machine model;
Step 5: after training, performance test sample data tests the supporting vector machine model after training;
Step 6: after test passes through, the supporting vector machine model based on genetic algorithm is constructed with gained supporting vector machine model
Online Network Voltage Stability monitoring model.
2. a kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm according to claim 1,
It is characterized by: in step 1, acquisition and PMU uncertainties model including voltage stability margin index.
3. a kind of support vector machines on-Line Voltage STABILITY MONITORING based on genetic algorithm according to claim 1 or 2
Method, it is characterised in that: the data of PMU measurement mainly include voltage magnitude and phase angle.
4. a kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm according to claim 1,
It is characterized by: in step 2, carries out conventional Load Flow and the voltage magnitude of present operating point, voltage phase angle, active is calculated
Power, reactive power, the new data being then calculated using these constitute sample number required for voltage stability margin
According to, and sample data is divided into training sample data and test sample data by a certain percentage.
5. a kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm according to claim 1,
It is characterized in that, in step 3, specifically includes the following steps:
(1) parameter of support vector machines is respectively included into regularization parameter C, radial basis function kernel function bandwidth σ2And it is insensitive
Loss function pipe radius ε carries out coding and generates chromosome x, then chromosome x indicates are as follows: X={ x1,x2,x3, wherein x1、x2And x3
Respectively indicate support vector machines parameter C, σ based on genetic algorithm2And ε;
(2) the support vector machines parameter selection optimized with Cross-Validation technique, in K- folding cross validation, by data
Collection is randomly divided into the subset of K part equivalent, and use wherein (K-1) part subset data SVM regression model is established as training set, branch
The performance for holding vector machine parameter is tested by k-th sub collection, is constantly repeated the above process, so that each subset is as test
Subset is once tested;
(3) initial population has 20 genomes being randomly generated at based on the balance between convergence time and population diversity
Property selection initial population size, and calculate the adaptive value for the chromosome that each is randomly generated;
(4) new population is created to replace existing population, then the side for passing through roulette with modes such as selection, intersection and mutation
Formula by adaptive value, preferably put into recombination pond by new chromosome.The gene of male parent and maternal chromosome swaps, and obtains new
Offspring;
(5) (3) step is repeated to (4) step, this process is repeated until that gene dosage reaches limit value.
6. a kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm according to claim 1,
It is characterized in that, in step 4, specifically includes the following steps: with optimal based on genetic algorithm obtained in step 3
Support vector machines parameter C, σ2And ε, the supporting vector machine model based on genetic algorithm of building, and using training sample data into
Row training obtains the different support vector machines fitting moulds based on genetic algorithm for the sample data of different network systems
Type.
7. a kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm according to claim 1,
It is characterized in that, utilizing the supporting vector machine model based on genetic algorithm of test sample data detection building in step 5
Fitness.
8. a kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm according to claim 1,
It is characterized in that, in step 6, the online Network Voltage Stability monitoring model of the supporting vector machine model based on genetic algorithm
By the inspection of actual electric network sample data, and complete the prediction to Network Voltage Stability nargin.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811468216.5A CN109711435A (en) | 2018-12-03 | 2018-12-03 | A kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811468216.5A CN109711435A (en) | 2018-12-03 | 2018-12-03 | A kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109711435A true CN109711435A (en) | 2019-05-03 |
Family
ID=66253897
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811468216.5A Pending CN109711435A (en) | 2018-12-03 | 2018-12-03 | A kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109711435A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111398679A (en) * | 2020-03-09 | 2020-07-10 | 华北电力大学 | Sub-synchronous oscillation identification and alarm method based on PMU (phasor measurement Unit) |
CN111461892A (en) * | 2020-03-31 | 2020-07-28 | 支付宝(杭州)信息技术有限公司 | Method and device for selecting derived variables of risk identification model |
CN111814394A (en) * | 2020-06-30 | 2020-10-23 | 三峡大学 | Power system safety assessment method based on correlation and redundancy detection |
CN115017735A (en) * | 2022-07-05 | 2022-09-06 | 重庆大学 | Multi-agent probability voltage stability calculation method for high-dimensional system |
CN116910637A (en) * | 2023-04-03 | 2023-10-20 | 山东科技大学 | Improved IGA-RBF neural network-based short-term load prediction method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104537382A (en) * | 2015-01-12 | 2015-04-22 | 杭州电子科技大学 | Electromyographic signal gait recognition method for optimizing support vector machine based on genetic algorithm |
CN106503867A (en) * | 2016-11-14 | 2017-03-15 | 吉林大学 | A kind of genetic algorithm least square wind power forecasting method |
CN108805355A (en) * | 2018-06-09 | 2018-11-13 | 贵州电网有限责任公司 | A kind of ground net corrosion prediction technique based on least square method supporting vector machine |
-
2018
- 2018-12-03 CN CN201811468216.5A patent/CN109711435A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104537382A (en) * | 2015-01-12 | 2015-04-22 | 杭州电子科技大学 | Electromyographic signal gait recognition method for optimizing support vector machine based on genetic algorithm |
CN106503867A (en) * | 2016-11-14 | 2017-03-15 | 吉林大学 | A kind of genetic algorithm least square wind power forecasting method |
CN108805355A (en) * | 2018-06-09 | 2018-11-13 | 贵州电网有限责任公司 | A kind of ground net corrosion prediction technique based on least square method supporting vector machine |
Non-Patent Citations (2)
Title |
---|
K.S.SAJAN等: "Genetic algorithm based support vector machine for on-line voltage stability monitoring", 《ELECTRICAL POWER AND ENERGY SYSTEMS》 * |
李大虎等: "基于PMU和混合支持向量机网络的电力系统暂态稳定性分析", 《电网技术》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111398679A (en) * | 2020-03-09 | 2020-07-10 | 华北电力大学 | Sub-synchronous oscillation identification and alarm method based on PMU (phasor measurement Unit) |
CN111398679B (en) * | 2020-03-09 | 2021-05-25 | 华北电力大学 | Sub-synchronous oscillation identification and alarm method based on PMU (phasor measurement Unit) |
CN111461892A (en) * | 2020-03-31 | 2020-07-28 | 支付宝(杭州)信息技术有限公司 | Method and device for selecting derived variables of risk identification model |
CN111461892B (en) * | 2020-03-31 | 2021-07-06 | 支付宝(杭州)信息技术有限公司 | Method and device for selecting derived variables of risk identification model |
CN111814394A (en) * | 2020-06-30 | 2020-10-23 | 三峡大学 | Power system safety assessment method based on correlation and redundancy detection |
CN111814394B (en) * | 2020-06-30 | 2023-08-25 | 三峡大学 | Power system safety assessment method based on correlation and redundancy detection |
CN115017735A (en) * | 2022-07-05 | 2022-09-06 | 重庆大学 | Multi-agent probability voltage stability calculation method for high-dimensional system |
CN115017735B (en) * | 2022-07-05 | 2023-08-15 | 重庆大学 | Multi-agent probability voltage stability calculation method for high-dimensional system |
CN116910637A (en) * | 2023-04-03 | 2023-10-20 | 山东科技大学 | Improved IGA-RBF neural network-based short-term load prediction method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109711435A (en) | A kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm | |
Li et al. | Real-time faulted line localization and PMU placement in power systems through convolutional neural networks | |
Mosavi et al. | A learning framework for size and type independent transient stability prediction of power system using twin convolutional support vector machine | |
CN108197744B (en) | Method and system for determining photovoltaic power generation power | |
CN102269972B (en) | Method and device for compensating pipeline pressure missing data based on genetic neural network | |
CN110417011A (en) | A kind of online dynamic secure estimation method based on mutual information Yu iteration random forest | |
CN110542819B (en) | Transformer fault type diagnosis method based on semi-supervised DBNC | |
WO2022021726A1 (en) | Pmu-based power system state estimation performance evaluation method | |
CN109088407B (en) | Power distribution network state estimation method based on deep belief network pseudo-measurement modeling | |
CN111523778A (en) | Power grid operation safety assessment method based on particle swarm algorithm and gradient lifting tree | |
CN109710661A (en) | Based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis | |
CN103246762A (en) | Method of comprehensive evaluation for simulation credibility of electric propulsion system | |
CN115310378A (en) | Power grid toughness evaluation and differentiation planning method under extreme typhoon disaster | |
CN109378835A (en) | Based on the large-scale electrical power system Transient Stability Evaluation system that mutual information redundancy is optimal | |
Liu et al. | Wind power short-term forecasting based on LSTM neural network with dragonfly algorithm | |
Sajan et al. | Genetic algorithm based artificial neural network model for voltage stability monitoring | |
Alharbi et al. | Short-term wind speed and temperature forecasting model based on gated recurrent unit neural networks | |
Xia et al. | Outlier reconstruction based distribution system state estimation using equivalent model of long short-term memory and Metropolis-Hastings sampling | |
CN110110944A (en) | One kind predicting wind electricity digestion demand quantity algorithm based on modified wavelet neural network | |
CN111814394B (en) | Power system safety assessment method based on correlation and redundancy detection | |
Theocharides et al. | Impact of data quality on day-ahead photovoltaic power production forecasting | |
Moradzadeh et al. | Image processing-based data integrity attack detection in dynamic line rating forecasting applications | |
CN111628531A (en) | Data driving method for static voltage stability evaluation of power system | |
CN111798049B (en) | Voltage stability evaluation method based on integrated learning and multi-target planning | |
CN116702629B (en) | Power system transient stability evaluation method with migration capability |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190503 |
|
RJ01 | Rejection of invention patent application after publication |