CN110348540A - Electrical power system transient angle stability Contingency screening method and device based on cluster - Google Patents
Electrical power system transient angle stability Contingency screening method and device based on cluster Download PDFInfo
- Publication number
- CN110348540A CN110348540A CN201910670195.3A CN201910670195A CN110348540A CN 110348540 A CN110348540 A CN 110348540A CN 201910670195 A CN201910670195 A CN 201910670195A CN 110348540 A CN110348540 A CN 110348540A
- Authority
- CN
- China
- Prior art keywords
- cluster
- rotor angle
- generator rotor
- transient
- training sample
- 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.)
- Granted
Links
- 230000001052 transient effect Effects 0.000 title claims abstract description 156
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000012216 screening Methods 0.000 title claims abstract description 33
- 238000004458 analytical method Methods 0.000 claims abstract description 20
- 238000013097 stability assessment Methods 0.000 claims abstract description 10
- 238000012549 training Methods 0.000 claims description 82
- 238000005070 sampling Methods 0.000 claims description 23
- 238000010606 normalization Methods 0.000 claims description 14
- 230000011218 segmentation Effects 0.000 claims description 11
- 230000000153 supplemental effect Effects 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 7
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 4
- 230000014509 gene expression Effects 0.000 claims description 4
- 239000012141 concentrate Substances 0.000 claims description 3
- 230000006641 stabilisation Effects 0.000 claims 1
- 238000011105 stabilization Methods 0.000 claims 1
- 238000011156 evaluation Methods 0.000 abstract description 3
- 238000004590 computer program Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005086 pumping Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Economics (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Probability & Statistics with Applications (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The electrical power system transient angle stability Contingency screening method and device based on cluster that the invention discloses a kind of, for each failure, based on safety on line analysis using the result data of Transient Stability Evaluation function storage, sampled according to transient state generator rotor angle nargin range to historical data；By Transient angle stability is all consistent with stable mode and transient state generator rotor angle nargin similar in sample clustering at several history run mode clusters, the optimal number of clusters of each failure is determined with silhouette coefficient method.Classify to each failure of the specified method of operation, then estimates its transient state generator rotor angle nargin；It finally obtains and the catastrophe failure collection that the transient rotor angle stability failure of given threshold forms is greater than by transient state generator rotor angle unstability fault set and transient state generator rotor angle estimation nargin.Using the present invention can quickly to on-line analysis applicating expedition transient rotor angle stability assessment fault set screen, effectively reduce transient rotor angle stability assessment number of faults, thus improve transient safe and stable analysis calculating speed.
Description
Technical field
The present invention relates to power system security stability analysis technical field more particularly to a kind of electric system based on cluster
Transient rotor angle stability Contingency screening method and device.
Background technique
As the on-line security and stability analysis application function of power grid real-time running data is widely applied, online transient state peace
Full stability analysis has become the urgent need and important reference indicator of bulk power grid management and running, and calculating cycle generally requires
It is completed within 5 minutes.With the construction of alternating current-direct current mixing bulk power grid, electric system scale increases year by year, needs to carry out online
For the forecast failure number of transient safe and stable analysis also with increasing, the assessment time of single failure also can be elongated.For thousands of
Generator, the bulk power grid of tens of thousands of a calculate nodes complete up to ten thousand if do not screened to forecast failure within 5 minutes
The transient safe and stable of forecast failure is analyzed, then needs to configure a large amount of computing resource.Common solution is dispatching of power netwoks
Operations staff filters out a small amount of forecast failure with off-line analysis experience and carries out online transient safe and stable analysis, reliability
It is closely related with the experience of dispatching of power netwoks operations staff.Therefore, it is badly in need of proposing that the online transient safe and stable analysis of electric system is pre-
Think Contingency screening method, reduces the setting of forecast failure collection and the degree of dependence of dispatching of power netwoks operations staff experience.
Summary of the invention
The electrical power system transient angle stability Contingency screening method that the purpose of the present invention is to provide a kind of based on cluster and
Device, can the failure quickly to on-line analysis applicating expedition screen, effectively reduce the number of faults of Transient Stability Evaluation
Amount, to improve the calculating speed of transient safe and stable analysis.
Realization that the present invention adopts the following technical solutions, specifically, including the following steps:
Electrical power system transient angle stability Contingency screening method based on cluster, comprising:
(1) training sample set is obtained for each failure；The training sample set is temporary from application is analyzed based on safety on line
In the historical data of state angle stability assessment storage, acquisition is sampled according to transient state generator rotor angle nargin range；
(2) training sample set is clustered according to cluster principle；The cluster principle are as follows: by with benchmark mode
Transient angle stability is consistent with stable mode, and training sample similar in transient state generator rotor angle nargin is clustered into cluster；
(3) the optimal number of clusters of cluster is determined according to the cluster result of every wheel；
(4) based on the optimal number of clusters of the cluster, the transient state generator rotor angle nargin of the failure to be investigated of the specified method of operation is calculated
Estimated value；
(5) Contingency screening is carried out according to the transient state generator rotor angle nargin estimated value of the failure to be investigated of the specified method of operation.
It is above-mentioned to be sampled according to transient state generator rotor angle nargin range, comprising:
By the value range [- 100,100] of transient state generator rotor angle nargin, be divided into two nargin sections [- 100,0) and [0,100]；
Sampling gear d is set separately in two nargin sections1And d2, obtain [- 100 ,-d1), [- d1, 0), [0, d2), [d2,
100] four sampling gear segmentations；
Calculate the sample size sampled needed for each sampling gear segmentation:
Wherein, YiIndicate the sample size sampled needed for i-th of sampling gear segmentation, DiIndicate i-th of sampling gear segmentation
Historical data quantity, N indicate historical data total amount；
Y is randomly selected from the historical data that each sampling gear is segmentediA historical data is as training sample, all pumpings
Training sample set of the training sample of sample gear segmentation as this failure.
Benchmark mode above-mentioned, which is chosen, includes:
Gathering is randomly choosed when initial clustering and is combined into sky, and first training sample is classified as first newly-generated cluster, simultaneously
It is also the benchmark mode as first cluster；
Training sample is clustered according to cluster principle, if existing cluster can not be included into, which individually becomes
New cluster, first is added benchmark mode of the training sample as the cluster of the cluster；
In cluster process, using K mean algorithm calculate cluster virtual center point, by with the Euclidean distance of virtual center point most
Sample in the close cluster benchmark mode brand new as this.
It is above-mentioned that the training sample set is clustered according to cluster principle, comprising:
11) training sample to be clustered is consistent with the Transient angle stability of benchmark mode, comprising:
Wherein, MbiFor the transient state generator rotor angle nargin of the benchmark mode of i-th of cluster；MjFor the transient state of training sample j to be clustered
Generator rotor angle nargin, wherein j ∈ [1, J], J indicate the sample size not clustered；
Indicate that the two Transient angle stability is consistent when W is 1, both expressions Transient angle stability is different when being 0；
12) training sample to be clustered is consistent with the stable mode of benchmark mode need to meet following three conditions:
A) the neck pre-group unit G of training sample j to be clusteredadv-jWith the neck pre-group unit of the benchmark mode of i-th of cluster
Gadv-iIn unit number, group name, to stop state completely the same for unit throwing；
B) lag group's unit G of training sample j to be clusteredlft-jWith lag group's unit of the benchmark mode of i-th of cluster
Glft-iIn unit number, group name, to stop state completely the same for unit throwing；
C) the critical circuits L of training sample j to be clusteredline-jWith the critical circuits L of the benchmark mode of i-th of clusterline-i
Route number, line name, that throwing stops state is completely the same；
13) Rule of judgment similar in the transient state generator rotor angle nargin of training sample to be clustered and benchmark mode are as follows:
|Mbi-Mj|≤Mmax
Wherein, MjIndicate the transient state generator rotor angle nargin of the training sample j of cluster, MmaxFor the threshold value of setting.
The termination condition of cluster above-mentioned are as follows: the quantity and benchmark mode of the cluster of cluster no longer change or reach specified change
It is terminated after generation number.
The optimal number of clusters above-mentioned that cluster is determined according to the cluster result of every wheel, comprising:
31) for the cluster result after every wheel iteration, Euclidean distance method is respectively adopted and carries out in cluster between dissmilarity degree and cluster
Dissimilar degree calculates, and calculates as follows:
Average Euclidean distance a of the calculating training sample i to same remaining training sample of clusteri, dissmilarity degree referred to as in cluster；
Calculate training sample i to cluster CjIn all training samples average Euclidean distance bi,j, then take bi,jMinimum value bi,
Dissmilarity degree referred to as between cluster；
32) according to dissmilarity degree between dissmilarity degree and cluster in the cluster of training sample i, the silhouette coefficient of training sample i is calculated:
It is known as the silhouette coefficient of epicycle cluster with the silhouette coefficient mean value of training samples all in cluster:
Wherein, M is indicated with training sample sums all in cluster；
33) according to the silhouette coefficient of every wheel cluster, the corresponding number of clusters of maximum value for choosing silhouette coefficient in all rounds is made
For the optimal number of clusters of cluster.
Optimal number of clusters above-mentioned based on the cluster, the transient state generator rotor angle for calculating the failure to be investigated of the specified method of operation are abundant
Spend estimated value, comprising:
41) the key feature amount and supplemental characteristic amount of each failure to be investigated are extracted from the specified method of operation, and to pass
Key characteristic quantity is normalized；
42) Euclidean distance between the benchmark mode of the specified consistent cluster of method of operation supplemental characteristic amount is calculated, institute is taken
It states distance in the optimal number of clusters of cluster and is not more than η dminX cluster, transient state generator rotor angle nargin is denoted as stable set not less than 0 cluster
Close Ca, the cluster by transient state generator rotor angle nargin less than 0 is denoted as unstability set Cb；Wherein, dminFor nearest Euclidean distance, η is not less than 1
Floating number, X is integer greater than 0；
43) selection refers to cluster: preferentially selecting the unstability set CbThe nearest cluster of middle Euclidean distance, which is used as, refers to cluster, if
Unstability set CbFor sky, then from stable set CaThe middle cluster for selecting Euclidean distance nearest is as with reference to cluster；
44) the benchmark mode with reference to cluster is taken, transient state generator rotor angle nargin estimated value is calculated:
Wherein, M indicates the transient state generator rotor angle nargin estimated value of failure to be investigated, MbFor with reference to cluster transient state generator rotor angle nargin,
Expression leads the participation factor of stablizing of j-th of unit in pre-group to include the flat of training sample set in the cluster in the benchmark mode with reference to cluster
Mean value,Indicating to refer to the stable participation factor that j-th of unit in group is lagged in the benchmark mode of cluster in the cluster includes sample set
Average value, P1b,jIt indicates to lead the value after the active power normalization of j-th of unit of pre-group, P ' in the benchmark mode with reference to cluster2b,j
Indicate to lag the value after the active power normalization of j-th of unit of group, P ' in the benchmark mode with reference to clusterjIndicate specified operation side
Value after leading the active power normalization of j-th of unit in pre-group in formula, P "jIt indicates to lag jth in group in the specified method of operation
Value after the active power normalization of a unit, GadvAnd GlftRespectively neck pre-group and lag group's unit set.
It is above-mentioned to extract stablizing for training sample and participate in factor absolute value not less than the neck pre-group unit of given threshold and stagnant
Group's unit is as Key generating unit afterwards, and takes the active power of Key generating unit as key feature amount；
The throwing for extracting Key generating unit stops state and the throwing of alternating current circuit stops state as supplemental characteristic amount；
Key feature amount is normalized, comprising:
Data prediction is carried out to key feature amount using z-score average variance method, is then returned according to the following formula
One changes:
Wherein, pkFor certain unit k-th of training sample active power,And σxIt is the unit in the equal of all samples
Value and standard deviation, p'kFor the value obtained after normalization.
The transient state generator rotor angle nargin estimated value of failure to be investigated above-mentioned according to the specified method of operation carries out failure sieve
Choosing, comprising:
51) the transient state generator rotor angle nargin estimated value of failure to be investigated indicates transient state generator rotor angle unstability less than 0, is otherwise transient state generator rotor angle
Stablize；All failures to be investigated of transient state generator rotor angle unstability are added to catastrophe failure to concentrate；
52) transient state generator rotor angle nargin threshold value is set as Mmar, Mmar∈ [0,100] takes transient state generator rotor angle nargin estimated value to be not more than
MmarFailure to be investigated be added to catastrophe failure concentration.
Electrical power system transient angle stability Contingency screening device based on cluster, comprising: training sample set obtains module, gathers
Generic module, optimum cluster determining module, transient state generator rotor angle nargin estimation block and Contingency screening module,
The training sample set obtains module and is used to obtain training sample set for each failure；The training sample set from
Based on safety on line analysis using transient rotor angle stability assessment storage historical data in, according to transient state generator rotor angle nargin range into
Line sampling obtains；
The cluster module is used to cluster the training sample set according to cluster principle；The cluster principle are as follows:
Will be consistent with the Transient angle stability and stable mode of benchmark mode, and training sample similar in transient state generator rotor angle nargin is poly-
Class is at cluster；
The optimum cluster determining module is used to determine the optimal number of clusters of cluster according to the cluster result of every wheel；
The transient state generator rotor angle nargin estimation block is used for the optimal number of clusters based on the cluster, calculates the specified method of operation
The transient state generator rotor angle nargin estimated value of failure to be investigated；
The Contingency screening module is used to be estimated according to the transient state generator rotor angle nargin of the failure to be investigated of the specified method of operation
Calculation value carries out Contingency screening.
The beneficial effects obtained by the present invention are as follows are as follows:
The present invention can the failure quickly to on-line analysis applicating expedition screen, effectively reduce the event of Transient Stability Evaluation
Hinder quantity, to improve the calculating speed of transient safe and stable analysis.
While not influencing to judge failure stability, the setting and dispatching of power netwoks for reducing forecast failure collection are run the present invention
The degree of dependence of personnel's experience also meets the timeliness requirement of scale grid line transient safe and stable analysis.
Detailed description of the invention
Fig. 1 is that the present invention is based on the flow charts of the electrical power system transient angle stability Contingency screening method of cluster.
Specific embodiment
The invention will be further described with reference to the accompanying drawing and by specific embodiment, and following embodiment is descriptive
, it is not restrictive, this does not limit the scope of protection of the present invention.
Referring in Fig. 1, the present invention provides a kind of electrical power system transient angle stability Contingency screening method based on cluster, packet
Include the following contents:
Step 1) is gone through based on safety on line analysis using the result of transient rotor angle stability assessment storage for each failure
History data are sampled historical data according to transient state generator rotor angle nargin range, obtain training sample set.It is specific as follows:
11) value range of transient state generator rotor angle nargin known to is [- 100,100], and separation 0 is divided into two nargin sections
[- 100,0) and [0,100], transient state generator rotor angle unstability and transient rotor angle stability are respectively indicated, in above-mentioned two nargin section respectively
Setting sampling gear d1And d2, wherein d1,d2∈[1,100].Set d1=d2=50, then gear of sampling segmentation according to [-
100, -50), [- 50,0), [and 0,50), [50,100] are divided, and the sample size Y of each sampling gear is calculatedi, specifically press
Formula (1) calculates:
Wherein: YiIndicate the sample size of i-th of sampling gear, DiIndicate i-th of sampling gear historical data quantity, N table
Show historical data total amount.
12) after the sample size for obtaining each sampling gear, Y is randomly selected from the historical data of each sampling geari
A historical data is as training sample.
Training sample set of the historical data of all sampling gears as this failure is finally obtained, wherein each sample data
Result information all comprising transient rotor angle stability assessment: transient state generator rotor angle nargin, neck pre-group and lag group's unit lead pre-group and stagnant
Stablizing for group's unit participates in the factor afterwards, leads pre-group and lags the active power of group's unit and throwing stops state and alternating current circuit
Throwing stops state.
Step 2), it is based on transient rotor angle stability assessment as a result, extracting each sample data stablizes ginseng for each failure
Neck pre-group and lag group's unit with factor absolute value not less than given threshold take the active of Key generating unit as Key generating unit
Power is as key feature amount and makees normalized, and the throwing for extracting Key generating unit stops state and the throwing of alternating current circuit stops state
As supplemental characteristic amount.
Key feature amount is extracted and normalization, specifically includes:
21) setting neck pre-group and lagging the participation factor max-thresholds of stablizing of group's unit is respectively λadv,λadv∈ (0,1] and
λlft,λlft∈ (0,1], take the unit stablized and participated in neck pre-group of the factor absolute value not less than respective threshold value and lag group to constitute
Key generating unit, then the active power of Key generating unit is taken to constitute key feature amount；
22) data prediction is carried out to key feature amount using z-score average variance method, specifically uses formula (2)
It is normalized, eliminates the difference of the order of magnitude between different characteristic amount.
Wherein: pkFor certain unit k-th of sample active power,And σxFor the unit all samples mean value and
Standard deviation, p'kParameter of the key feature amount as calculating distance for obtained value after normalization, after normalizing.
Step 3) respectively clusters the training sample set of each failure, randomly chooses not belong to for the first time when cluster and appoint
Other are not belonging to transient state generator rotor angle any cluster and with benchmark mode by benchmark mode of one sample of what cluster as new cluster
Stability is consistent with stable mode and transient state generator rotor angle nargin similar in sample be polymerized to cluster.Then K mean algorithm (K- is used
Means algorithm) calculate cluster virtual center point, using the base brand new as this of the sample in the cluster nearest apart from virtual center point
Quasi- mode, then continue to be clustered according to the above method in a manner of new benchmark, until the quantity and benchmark mode of cluster no longer become
Until changing or reaching after given number of iterations, optimal number of clusters is finally determined with silhouette coefficient method.
It is specific as follows:
31) Transient angle stability of sample is unanimously identified according to formula (3):
In formula: MbiFor the transient state generator rotor angle nargin of the benchmark mode of i-th of cluster；MjFor the transient state of other sample j not clustered
Generator rotor angle nargin, wherein j ∈ [1, J], J indicate the sample size not clustered.
Indicate that the two Transient angle stability is consistent when W is 1, both expressions Transient angle stability is different when being 0.
32) sample j to be clustered is consistent with the stable mode of benchmark mode of i-th of cluster need to meet following three conditions:
A) the neck pre-group unit G of sample j to be clusteredadv-jWith the neck pre-group unit G of the benchmark mode of i-th of clusteradv-iIn
Unit number, group name, to stop state completely the same for unit throwing；
B) lag group's unit G of sample j to be clusteredlft-jWith lag group's unit G of the benchmark mode of i-th of clusterlft-iIn
Unit number, group name, to stop state completely the same for unit throwing；
C) the critical circuits L of sample j to be clusteredline-jWith the critical circuits L of the benchmark mode of i-th of clusterline-iLine
Road number, line name, that throwing stops state is completely the same.
33) specific as shown in formula (4) with Rule of judgment similar in the transient state generator rotor angle nargin of benchmark mode:
|Mbi-Mj|≤Mmax (4)
Wherein: MjIndicate the transient state generator rotor angle nargin of sample j, MbiFor the transient state generator rotor angle nargin of the benchmark mode of i-th of cluster,
MmaxFor the threshold value of setting, value range be (0,15].
34) the distance between sample uses Euclidean distance calculation method, specific as shown in formula (5):
Wherein: disi,jIndicate the distance between sample i and j, p 'i,k, p'j,kSample i and j Key generating unit k is respectively indicated to return
Value after one change, K indicate Key generating unit number.
Optimal number of clusters is determined with silhouette coefficient method, specific as follows:
35) for the cluster result after every wheel iteration, Euclidean distance method is respectively adopted and carries out in cluster between dissmilarity degree and cluster
Dissimilar degree calculates.
Average Euclidean distance a of the calculating sample i to same other samples of clusteri, dissmilarity degree referred to as in cluster.
Calculate sample i to other clusters CjIn all samples average Euclidean distance bi,j, then take bi,jMinimum value bi, referred to as
For sample i and cluster CjDissimilar degree, be called dissmilarity degree between doing cluster.
36) according to dissmilarity degree between dissmilarity degree and cluster in the cluster of sample i, the profile system of sample i is calculated using formula (6)
Number si。
It is referred to as the silhouette coefficient SC of cluster result with the mean value of the silhouette coefficient of samples all in cluster, specifically uses formula
(7) it is calculated.
M is indicated with total sample numbers all in cluster.
For silhouette coefficient SC value range between [- 1,1], the value is bigger, indicates that cluster is more reasonable.
37) the silhouette coefficient SC clustered according to each roundz, take the corresponding number of clusters of the maximum value of silhouette coefficient in all rounds
Optimal number of clusters as cluster.
Step 4), for each failure of the specified method of operation to be investigated, extraction step 2) described in Key generating unit
Key feature amount and supplemental characteristic amount calculate between the benchmark mode of the specified consistent cluster of method of operation supplemental characteristic amount
Euclidean distance, each of select several clusters for being closer based on stability conservatism, then estimate the specified method of operation therefore
The transient state generator rotor angle nargin of barrier.
Transient state generator rotor angle nargin estimation process is as follows:
41) the key feature amount and supplemental characteristic amount of each failure to be investigated are extracted from specified running mode data, it is right
Key feature amount is normalized by formula (1).
42) Euclidean distance between the benchmark mode of the specified consistent cluster of method of operation supplemental characteristic amount is calculated, if most
Nearly Euclidean distance is dmin(failure is between the benchmark mode key feature amount of key feature amount and cluster under the specified method of operation
Euclidean distance), then take Euclidean distance no more than η dminX cluster, cluster here refer to step 3 cluster after the completion of cluster.
Wherein, η is the floating number not less than 1, and X is the integer greater than 0.Cluster by transient state generator rotor angle nargin not less than 0 is denoted as stable set
Close Ca, cluster of the transient state generator rotor angle nargin less than 0 be denoted as unstability set Cb。
43) cluster is referred to based on the selection of stability conservative estimation principle, i.e., preferentially unstability set C described in selection step 42)b
The nearest cluster of middle Euclidean distance, if set CbFor sky, then from stable set CaThe nearest cluster of middle selection Euclidean distance.
44) the benchmark mode with reference to cluster is taken, if its transient state generator rotor angle nargin is Mb, neck pre-group and lag group's unit set are distinguished
For GadvAnd Glft, estimate the transient state generator rotor angle nargin of the failure of the specified method of operation to be investigated, specifically calculated using formula (8).
Wherein, M indicates the estimated value of the transient state generator rotor angle nargin of failure to be investigated,It indicates to lead the in pre-group in benchmark mode
The stablizing of j unit participates in the average value that the factor includes sample set in the cluster,It indicates to lag in group j-th in benchmark mode
Stablizing for unit participates in the factor in average value of the cluster comprising sample set, P '1b,jIt indicates to lead j-th of machine of pre-group in benchmark mode
The active power of group takes according to the value after formula (1) normalization, P '2b,jIt indicates to lag j-th of unit of group in benchmark mode
Active power takes according to the value after formula (1) normalization, P 'jJ-th of unit is indicated to lead in pre-group in the specified method of operation
Active power, P "jIt indicates the active power for lagging j-th of unit in group in the specified method of operation, is normalized according to formula (1)
Value afterwards.
Step 5) judges the stability of failure according to the transient state generator rotor angle nargin estimated value of failure to be investigated, finally obtains packet
What the unstability failure of generator rotor angle containing transient state and transient state generator rotor angle nargin estimated value were formed no more than the transient rotor angle stability failure of given threshold
Catastrophe failure collection.
Detailed process is as follows:
51) judge Transient angle stability of the failure to be investigated under real time operation mode: the transient state function of failure to be investigated
Angle nargin estimated value indicates transient state generator rotor angle unstability less than 0, is otherwise transient rotor angle stability.
52) all failures to be investigated of transient state generator rotor angle unstability catastrophe failure is added to concentrate.
53) it sets transient state generator rotor angle nargin threshold value and is set as Mmar, Mmar∈ [0,100] takes transient state generator rotor angle nargin estimated value to be not more than
MmarTransient stability failure be added to catastrophe failure concentration.
The present invention also provides a kind of electrical power system transient angle stability Contingency screening device based on cluster, comprising: training
Sample set obtains module, cluster module, optimum cluster determining module, transient state generator rotor angle nargin estimation block and Contingency screening module,
The training sample set obtains module and is used to obtain training sample set for each failure；The training sample set from
Based on safety on line analysis using transient rotor angle stability assessment storage historical data in, according to transient state generator rotor angle nargin range into
Line sampling obtains；
The cluster module is used to cluster the training sample set according to cluster principle；The cluster principle are as follows:
Will be consistent with the Transient angle stability and stable mode of benchmark mode, and training sample similar in transient state generator rotor angle nargin is poly-
Class is at cluster；
The optimum cluster determining module is used to determine the optimal number of clusters of cluster according to the cluster result of every wheel；
The transient state generator rotor angle nargin estimation block is used for the optimal number of clusters based on the cluster, calculates the specified method of operation
The transient state generator rotor angle nargin estimated value of failure to be investigated；
The Contingency screening module is used to be estimated according to the transient state generator rotor angle nargin of the failure to be investigated of the specified method of operation
Calculation value carries out Contingency screening.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that can be realized by computer program instructions each in flowchart and/or the block diagram
The combination of process and/or box in process and/or box and flowchart and/or the block diagram.It can provide these computers
Processor of the program instruction to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices
To generate a machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute
For realizing the function of being specified in one or more flows of the flowchart and/or one or more blocks of the block diagram
Device.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that instruction stored in the computer readable memory generation includes
The manufacture of command device, the command device are realized in one box of one or more flows of the flowchart and/or block diagram
Or the function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer
Or the instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or box
The step of function of being specified in figure one box or multiple boxes.
Finally it should be noted that: the above embodiments are merely illustrative of the technical scheme of the present invention and are not intended to be limiting thereof, to the greatest extent
Invention is explained in detail referring to above-described embodiment for pipe, it should be understood by those ordinary skilled in the art that: still
Can with modifications or equivalent substitutions are made to specific embodiments of the invention, and without departing from spirit and scope of the invention appoint
What modification or equivalent replacement, should all cover within the scope of the claims of the present invention.
Claims (10)
1. the electrical power system transient angle stability Contingency screening method based on cluster characterized by comprising
(1) training sample set is obtained for each failure；The training sample set applies transient state function from based on safety on line analysis
In the historical data of angle Stability Assessment storage, acquisition is sampled according to transient state generator rotor angle nargin range；
(2) training sample set is clustered according to cluster principle；The cluster principle are as follows: by the transient state with benchmark mode
Power-angle stability is consistent with stable mode, and training sample similar in transient state generator rotor angle nargin is clustered into cluster；
(3) the optimal number of clusters of cluster is determined according to the cluster result of every wheel；
(4) based on the optimal number of clusters of the cluster, the transient state generator rotor angle nargin estimation of the failure to be investigated of the specified method of operation is calculated
Value；
(5) Contingency screening is carried out according to the transient state generator rotor angle nargin estimated value of the failure to be investigated of the specified method of operation.
2. the electrical power system transient angle stability Contingency screening method according to claim 1 based on cluster, feature exist
In described to be sampled according to transient state generator rotor angle nargin range, comprising:
By the value range [- 100,100] of transient state generator rotor angle nargin, be divided into two nargin sections [- 100,0) and [0,100]；
Sampling gear d is set separately in two nargin sections1And d2, obtain [- 100 ,-d1), [- d1, 0), [0, d2), [d2,100]
Four sampling gear segmentations；
Calculate the sample size sampled needed for each sampling gear segmentation:
Wherein, YiIndicate the sample size sampled needed for i-th of sampling gear segmentation, DiIndicate going through for i-th of sampling gear segmentation
History data bulk, N indicate historical data total amount；
Y is randomly selected from the historical data that each sampling gear is segmentediA historical data is as training sample, all sampling shelves
Training sample set of the training sample of position segmentation as this failure.
3. the electrical power system transient angle stability Contingency screening method according to claim 1 based on cluster, feature exist
In the benchmark mode, which is chosen, includes:
Gathering is randomly choosed when initial clustering and is combined into sky, and first training sample is classified as first newly-generated cluster, is also simultaneously
Benchmark mode as first cluster；
Training sample is clustered according to cluster principle, if existing cluster can not be included into, which individually becomes new
Cluster, first is added benchmark mode of the training sample as the cluster of the cluster；
In cluster process, the virtual center point of cluster is calculated using K mean algorithm, it will be nearest with the Euclidean distance of virtual center point
Sample in the cluster benchmark mode brand new as this.
4. the electrical power system transient angle stability Contingency screening method according to claim 1 based on cluster, feature exist
In described to be clustered to the training sample set according to cluster principle, comprising:
11) training sample to be clustered is consistent with the Transient angle stability of benchmark mode, comprising:
Wherein, MbiFor the transient state generator rotor angle nargin of the benchmark mode of i-th of cluster；MjFor the transient state generator rotor angle of training sample j to be clustered
Nargin, wherein j ∈ [1, J], J indicate the sample size not clustered；
Indicate that the two Transient angle stability is consistent when W is 1, both expressions Transient angle stability is different when being 0；
12) training sample to be clustered is consistent with the stable mode of benchmark mode need to meet following three conditions:
A) the neck pre-group unit G of training sample j to be clusteredadv-jWith the neck pre-group unit G of the benchmark mode of i-th of clusteradv-iIn
Unit number, group name, to stop state completely the same for unit throwing；
B) lag group's unit G of training sample j to be clusteredlft-jWith lag group's unit G of the benchmark mode of i-th of clusterlft-iIn
Unit number, group name, to stop state completely the same for unit throwing；
C) the critical circuits L of training sample j to be clusteredline-jWith the critical circuits L of the benchmark mode of i-th of clusterline-iLine
Road number, line name, that throwing stops state is completely the same；
13) Rule of judgment similar in the transient state generator rotor angle nargin of training sample to be clustered and benchmark mode are as follows:
|Mbi-Mj|≤Mmax
Wherein, MjIndicate the transient state generator rotor angle nargin of the training sample j of cluster, MmaxFor the threshold value of setting.
5. the electrical power system transient angle stability Contingency screening method according to claim 1 based on cluster, feature exist
In the termination condition of the cluster are as follows: the quantity and benchmark mode of the cluster of cluster no longer change or reach given number of iterations
After terminate.
6. the electrical power system transient angle stability Contingency screening method according to claim 1 based on cluster, feature exist
In the optimal number of clusters for determining cluster according to the cluster result of every wheel, comprising:
31) for the cluster result after every wheel iteration, not phase is respectively adopted in Euclidean distance method progress cluster between dissmilarity degree and cluster
It calculates, calculates as follows like degree:
Average Euclidean distance a of the calculating training sample i to same remaining training sample of clusteri, dissmilarity degree referred to as in cluster；
Calculate training sample i to cluster CjIn all training samples average Euclidean distance bi,j, then take bi,jMinimum value bi, referred to as
The dissmilarity degree between cluster；
32) according to dissmilarity degree between dissmilarity degree and cluster in the cluster of training sample i, the silhouette coefficient of training sample i is calculated:
It is known as the silhouette coefficient of epicycle cluster with the silhouette coefficient mean value of training samples all in cluster:
Wherein, M is indicated with training sample sums all in cluster；
33) according to the silhouette coefficient of every wheel cluster, the corresponding number of clusters of maximum value of silhouette coefficient in all rounds is chosen as poly-
The optimal number of clusters of class.
7. the electrical power system transient angle stability Contingency screening method according to claim 1 based on cluster, feature exist
In the optimal number of clusters based on the cluster calculates the transient state generator rotor angle nargin estimation of the failure to be investigated of the specified method of operation
Value, comprising:
41) the key feature amount and supplemental characteristic amount of each failure to be investigated are extracted from the specified method of operation, and to crucial special
Sign amount is normalized；
42) Euclidean distance between the benchmark mode of the specified consistent cluster of method of operation supplemental characteristic amount is calculated, is taken described poly-
Distance is not more than η d in the optimal number of clusters of classminX cluster, transient state generator rotor angle nargin is denoted as stable set C not less than 0 clustera,
Cluster by transient state generator rotor angle nargin less than 0 is denoted as unstability set Cb；Wherein, dminFor nearest Euclidean distance, η is the floating-point not less than 1
Number, X are the integer greater than 0；
43) selection refers to cluster: preferentially selecting the unstability set CbThe nearest cluster of middle Euclidean distance, which is used as, refers to cluster, if unstability
Set CbFor sky, then from stable set CaThe middle cluster for selecting Euclidean distance nearest is as with reference to cluster；
44) the benchmark mode with reference to cluster is taken, transient state generator rotor angle nargin estimated value is calculated:
Wherein, M indicates the transient state generator rotor angle nargin estimated value of failure to be investigated, MbFor with reference to cluster transient state generator rotor angle nargin,It indicates
Leading the stabilization of j-th of unit in pre-group to participate in the factor in the cluster in benchmark mode with reference to cluster includes being averaged for training sample set
Value,Indicating to refer to the stable participation factor that j-th of unit in group is lagged in the benchmark mode of cluster in the cluster includes sample set
Average value, P '1b,jIt indicates to lead the value after the active power normalization of j-th of unit of pre-group, P ' in the benchmark mode with reference to cluster2b,j
Indicate to lag the value after the active power normalization of j-th of unit of group, P ' in the benchmark mode with reference to clusterjIndicate specified operation side
Value after leading the active power normalization of j-th of unit in pre-group in formula, P "jIt indicates to lag jth in group in the specified method of operation
Value after the active power normalization of a unit, GadvAnd GlftRespectively neck pre-group and lag group's unit set.
8. the electrical power system transient angle stability Contingency screening method according to claim 7 based on cluster, feature exist
In,
The neck pre-group unit for participating in factor absolute value not less than given threshold of stablizing for extracting training sample is made with lag group's unit
For Key generating unit, and take the active power of Key generating unit as key feature amount；
The throwing for extracting Key generating unit stops state and the throwing of alternating current circuit stops state as supplemental characteristic amount；
Key feature amount is normalized, comprising:
Data prediction is carried out to key feature amount using z-score average variance method, is then normalized according to the following formula:
Wherein, pkFor certain unit k-th of training sample active power,And σxFor the unit all samples mean value and
Standard deviation, p'kFor the value obtained after normalization.
9. the electrical power system transient angle stability Contingency screening method according to claim 7 based on cluster, feature exist
In the transient state generator rotor angle nargin estimated value of the failure to be investigated according to the specified method of operation carries out Contingency screening, comprising:
51) the transient state generator rotor angle nargin estimated value of failure to be investigated indicates transient state generator rotor angle unstability less than 0, otherwise steady for transient state generator rotor angle
It is fixed；All failures to be investigated of transient state generator rotor angle unstability are added to catastrophe failure to concentrate；
52) transient state generator rotor angle nargin threshold value is set as Mmar, Mmar∈ [0,100] takes transient state generator rotor angle nargin estimated value no more than Mmar's
Failure to be investigated is added to catastrophe failure concentration.
10. the electrical power system transient angle stability Contingency screening device based on cluster characterized by comprising training sample set
Module, cluster module, optimum cluster determining module, transient state generator rotor angle nargin estimation block and Contingency screening module are obtained,
The training sample set obtains module and is used to obtain training sample set for each failure；The training sample set is from being based on
In historical data of the safety on line analysis using transient rotor angle stability assessment storage, it is sampled according to transient state generator rotor angle nargin range
It obtains；
The cluster module is used to cluster the training sample set according to cluster principle；The cluster principle are as follows: will be with
The Transient angle stability of benchmark mode is consistent with stable mode, and training sample similar in transient state generator rotor angle nargin is clustered into one
Cluster；
The optimum cluster determining module is used to determine the optimal number of clusters of cluster according to the cluster result of every wheel；
The transient state generator rotor angle nargin estimation block is used for the optimal number of clusters based on the cluster, calculates needing checking for the specified method of operation
Examine the transient state generator rotor angle nargin estimated value of failure；
The Contingency screening module is used for the transient state generator rotor angle nargin estimated value of the failure to be investigated according to the specified method of operation
Carry out Contingency screening.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910670195.3A CN110348540B (en) | 2019-07-24 | 2019-07-24 | Clustering-based method and device for screening transient power angle stability faults of power system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910670195.3A CN110348540B (en) | 2019-07-24 | 2019-07-24 | Clustering-based method and device for screening transient power angle stability faults of power system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110348540A true CN110348540A (en) | 2019-10-18 |
| CN110348540B CN110348540B (en) | 2021-06-01 |
Family
ID=68180102
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910670195.3A Active CN110348540B (en) | 2019-07-24 | 2019-07-24 | Clustering-based method and device for screening transient power angle stability faults of power system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110348540B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111244937A (en) * | 2020-01-09 | 2020-06-05 | 清华大学 | Method for screening serious faults of transient voltage stability of power system |
| CN112287284A (en) * | 2020-10-28 | 2021-01-29 | 山东电力研究院 | Transient stability fault screening method and system considering N-m fault time interval |
| CN116093952A (en) * | 2023-03-06 | 2023-05-09 | 国网浙江省电力有限公司温州供电公司 | Transient voltage stability monitoring bus analysis method |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5638297A (en) * | 1994-07-15 | 1997-06-10 | British Columbia Hydro And Power Authority | Method of on-line transient stability assessment of electrical power systems |
| CN101794998A (en) * | 2010-02-05 | 2010-08-04 | 湖南大学 | Online transient stability analysis method based on concise expression form of electromagnetic power of single generator in multi-machine power system |
| CN102074955A (en) * | 2011-01-20 | 2011-05-25 | 中国电力科学研究院 | Method based on knowledge discovery technology for stability assessment and control of electric system |
| CN107590604A (en) * | 2017-09-13 | 2018-01-16 | 国网福建省电力有限公司 | The people having the same aspiration and interest unit grouping method and system of a kind of combination S-transformation and 2DPCA |
| CN108681973A (en) * | 2018-05-14 | 2018-10-19 | 广州供电局有限公司 | Sorting technique, device, computer equipment and the storage medium of power consumer |
| CN108876163A (en) * | 2018-06-27 | 2018-11-23 | 国电南瑞科技股份有限公司 | The transient rotor angle stability fast evaluation method of comprehensive causality analysis and machine learning |
| CN109861206A (en) * | 2018-12-29 | 2019-06-07 | 国电南瑞科技股份有限公司 | A kind of transient rotor angle stability Contingency screening system and method based on support vector machines |
| JP2019091124A (en) * | 2017-11-10 | 2019-06-13 | 株式会社東芝 | Reliability evaluation system, reliability evaluation method and program |
| CN109902352A (en) * | 2019-01-24 | 2019-06-18 | 南瑞集团有限公司 | A kind of transient rotor angle stability analysis of key Characteristic Extraction method and system |
-
2019
- 2019-07-24 CN CN201910670195.3A patent/CN110348540B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5638297A (en) * | 1994-07-15 | 1997-06-10 | British Columbia Hydro And Power Authority | Method of on-line transient stability assessment of electrical power systems |
| CN101794998A (en) * | 2010-02-05 | 2010-08-04 | 湖南大学 | Online transient stability analysis method based on concise expression form of electromagnetic power of single generator in multi-machine power system |
| CN102074955A (en) * | 2011-01-20 | 2011-05-25 | 中国电力科学研究院 | Method based on knowledge discovery technology for stability assessment and control of electric system |
| CN107590604A (en) * | 2017-09-13 | 2018-01-16 | 国网福建省电力有限公司 | The people having the same aspiration and interest unit grouping method and system of a kind of combination S-transformation and 2DPCA |
| JP2019091124A (en) * | 2017-11-10 | 2019-06-13 | 株式会社東芝 | Reliability evaluation system, reliability evaluation method and program |
| CN108681973A (en) * | 2018-05-14 | 2018-10-19 | 广州供电局有限公司 | Sorting technique, device, computer equipment and the storage medium of power consumer |
| CN108876163A (en) * | 2018-06-27 | 2018-11-23 | 国电南瑞科技股份有限公司 | The transient rotor angle stability fast evaluation method of comprehensive causality analysis and machine learning |
| CN109861206A (en) * | 2018-12-29 | 2019-06-07 | 国电南瑞科技股份有限公司 | A kind of transient rotor angle stability Contingency screening system and method based on support vector machines |
| CN109902352A (en) * | 2019-01-24 | 2019-06-18 | 南瑞集团有限公司 | A kind of transient rotor angle stability analysis of key Characteristic Extraction method and system |
Non-Patent Citations (3)
| Title |
|---|
| B. DORA ARUL SELVI ET AL: "Investigation of power system transient stability using Clustering Based Support Vector Machines and preventive control by rescheduling generators", 《2007 IET-UK INTERNATIONAL CONFERENCE ON INFORMATION AND COMMUNICATION TECHNOLOGY IN ELECTRICAL SCIENCES (ICTES 2007)》 * |
| FANG SHI ET AL: "Association Rules Analysis between Power System Operating States and Transient Stability Margin", 《2017 IEEE INNOVATIVE SMART GRID TECHNOLOGIES - ASIA (ISGT-ASIA)》 * |
| 王成山等: "基于聚类分析的电力系统暂态稳定故障筛选", 《电网技术》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111244937A (en) * | 2020-01-09 | 2020-06-05 | 清华大学 | Method for screening serious faults of transient voltage stability of power system |
| CN112287284A (en) * | 2020-10-28 | 2021-01-29 | 山东电力研究院 | Transient stability fault screening method and system considering N-m fault time interval |
| CN116093952A (en) * | 2023-03-06 | 2023-05-09 | 国网浙江省电力有限公司温州供电公司 | Transient voltage stability monitoring bus analysis method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110348540B (en) | 2021-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110417011B (en) | Online dynamic security assessment method based on mutual information and iterative random forest | |
| CN106505557B (en) | Remote measurement error identification method and device | |
| CN110348540A (en) | Electrical power system transient angle stability Contingency screening method and device based on cluster | |
| CN107340492A (en) | Electric power meter failure analysis methods with scene anticipation are excavated based on big data | |
| CN109501834A (en) | A kind of point machine failure prediction method and device | |
| CN109102157A (en) | A kind of bank's work order worksheet processing method and system based on deep learning | |
| CN113922412B (en) | New energy multi-station short-circuit ratio panoramic evaluation method, system, storage medium and computing equipment | |
| CN109378835A (en) | Based on the large-scale electrical power system Transient Stability Evaluation system that mutual information redundancy is optimal | |
| CN107729939A (en) | A kind of CIM extended method and device towards newly-increased power network resources | |
| CN110348683A (en) | The main genetic analysis method, apparatus equipment of electrical energy power quality disturbance event and storage medium | |
| CN113625103A (en) | Line selection method for single-phase earth fault of small current grounding system | |
| CN103902798B (en) | Data preprocessing method | |
| CN103885867A (en) | Online evaluation method of performance of analog circuit | |
| CN114385403A (en) | Distributed cooperative fault diagnosis method based on double-layer knowledge graph framework | |
| CN117034149A (en) | Fault processing strategy determining method and device, electronic equipment and storage medium | |
| Hogan et al. | Towards effective clustering techniques for the analysis of electric power grids | |
| CN109902352A (en) | A kind of transient rotor angle stability analysis of key Characteristic Extraction method and system | |
| CN106571969B (en) | A kind of cloud service usability evaluation method and system | |
| CN109670526A (en) | A kind of interference source type discrimination method, device, equipment and the storage medium of power grid | |
| CN105842535B (en) | A kind of main syndrome screening technique of harmonic wave based on similar features fusion | |
| CN114781495A (en) | Intelligent ammeter fault classification method based on sample global rebalancing | |
| Zhou et al. | A multi-stage multi-criteria data analytics approach to rank commercial service airports | |
| CN109492913B (en) | Modular risk prediction method and device for power distribution terminal and storable medium | |
| Matijašević et al. | Voltage-based machine learning algorithm for distribution of end-users consumption among the phases | |
| CN106681967A (en) | Meter changing operation method based on data backtracking mechanism |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |