CN110674864B - Wind power abnormal data identification method comprising synchronous phasor measurement device - Google Patents
Wind power abnormal data identification method comprising synchronous phasor measurement device Download PDFInfo
- Publication number
- CN110674864B CN110674864B CN201910892258.XA CN201910892258A CN110674864B CN 110674864 B CN110674864 B CN 110674864B CN 201910892258 A CN201910892258 A CN 201910892258A CN 110674864 B CN110674864 B CN 110674864B
- Authority
- CN
- China
- Prior art keywords
- data
- wind power
- sample
- abnormal data
- abnormal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 63
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000005259 measurement Methods 0.000 title claims abstract description 17
- 230000001360 synchronised effect Effects 0.000 title claims abstract description 9
- 239000011159 matrix material Substances 0.000 claims description 23
- 238000004422 calculation algorithm Methods 0.000 claims description 15
- 238000012937 correction Methods 0.000 claims description 10
- 238000005070 sampling Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000004088 simulation Methods 0.000 claims description 4
- 238000012795 verification Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 description 3
- 238000007619 statistical method Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 2
- 241001513371 Knautia arvensis Species 0.000 description 1
- 230000002547 anomalous effect Effects 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012545 processing 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
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Abstract
The invention relates to a wind power abnormal data identification method comprising a synchronous phasor measurement device. Compared with the prior art, the invention has the advantages of excellent identification speed, precision and the like.
Description
Technical Field
The invention relates to the technical field of power big data analysis and preprocessing, in particular to a wind power abnormal data identification method with a synchronous phasor measurement device.
Background
The environment and energy crisis are highly valued in the whole society, and the distributed renewable energy source is rapidly developed for improving the energy consumption structure. The capacity of the total assembly machine of the wind turbine generator system rises year by year, and the influence of the randomness of the output of the total assembly machine on the power grid is increasingly remarkable. In order to reduce the influence of the randomness of wind power generation on the stability of a power system, an important method is wind power big data measurement and excavation so as to improve the predictability of operation. However, in the wind power data obtained by measurement, a large amount of abnormal data generally exist due to the reasons of manual regulation, shutdown, communication faults and the like, and serious interference is brought to the analysis and the use of the data. Therefore, abnormal data identification and correction are necessary to be carried out on wind power data so as to provide guarantee for subsequent research.
There are many studies on abnormal data identification at present, but most of them are identification methods proposed based on SCADA sampling systems. With the gradual application of PMU measurement devices, the high-precision and high-frequency sampling (with a typical sampling interval of 20 ms) characteristics also put new demands on the identification method, while the traditional statistical-based method is not suitable due to the huge data volume. Meanwhile, research in the related field of wind power operation abnormal data identification is relatively lacking, and influence of wind speed change on wind power data identification is not considered. When the wind speed is lower than the cut-in speed or higher than the cut-out speed, the abnormal data identification is greatly disturbed, and the influence of the wind speed is required to be considered when the abnormal wind power data is identified.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a wind power abnormal data identification method with a synchronous phasor measurement device.
The aim of the invention can be achieved by the following technical scheme:
a wind power abnormal data identification method comprising a synchronous phasor measurement device considers the influence of wind speed on abnormal data identification, adopts an improved FCM algorithm to cluster fan operation data, adopts a statistical method to calculate a clustering result to obtain a feasible region matrix, and then sequentially carries out abnormal data identification.
Preferably, the method specifically comprises the steps of:
step 1, classifying wind power operation big data according to the wind speed;
step 2, obtaining the number of clusters and an initial cluster center by using a subtractive mean clustering algorithm;
step 3, carrying out fuzzy mean clustering based on the number of the clusters obtained in the step 2 and the initial cluster center;
step 4, calculating a feasible domain matrix of each cluster by using a statistical rule;
step 5, distinguishing abnormal data by using the feasible domain matrix of each cluster obtained in the step 4;
and 6, correcting the abnormal data identified in the step 5 by adopting a vertical comparison method.
Preferably, the step 1 includes: the PMU measuring device is installed at the sampling node, and meanwhile wind speed is not suddenly changed in a continuous period of time, so that wind power big data are classified according to the wind speed, and wind power data lower than cut-in wind speed and higher than cut-out wind speed are paid attention to.
Preferably, in the step 2, in view of the fact that the initial cluster number and the cluster center thereof cannot be effectively determined by the traditional FCM algorithm, the wind power data are considered to have the characteristics of large data volume and high dimension, and the cluster number and the cluster center do not need to be determined before clustering by adopting the subtractive clustering algorithm and are easy to converge.
Preferably, in the step 3, the FCM algorithm is adopted to calculate the weights of all the classes, and the class with the largest weight is selected as the class.
Preferably, the specific process of the step 4 is as follows:
calculating a sample mean value:
in which x is i Represents the wind power data of a certain type at the ith moment,and the average value of the wind power data is represented, and n is the sampling number.
Randomly adding 1 abnormal data into the raw sample data, respectively calculating the raw sample standard deviation and the sample standard deviation after the abnormal data are added, and calculating the ratio a of the front standard deviation and the rear standard deviation 1 :
Delta in 2 、For standard deviation and mean after addition of anomaly data, delta 1 U is the standard deviation and mean of the original sample.
Calculating a feasible region range:
K 1 、K 2 、K 3 the meaning is as follows:
preferably, the step 6 includes:
considering that a high-precision PMU measuring device is installed, the real value of abnormal data at a certain moment is relatively close to the measured values at the front and rear moments, so the invention adopts a vertical comparison method to correct the abnormal data
In which x is ij Sample data representing a unit at the j-th moment of the i-th day,representing the unitSample data at the j-th time of day i-1,>sample data representing the ith-1 th day, jth-1 th moment, deltax of the unit i,j-1 The difference between the sample data at the j-th time on the i-1 day and the sample data at the j-th time on the i-th day is shown.
Preferably, the method is based on the proposed abnormal data identification and correction model, adopts simulation examples and builds a simulation model for verification analysis.
Compared with the prior art, the invention has the following advantages:
firstly, primarily classifying wind power data according to the wind speed, then performing cluster analysis on the primarily classified wind power data through an improved fuzzy mean algorithm, then calculating a feasible region matrix of each classification cluster according to a statistical rule, judging whether the wind power data is abnormal data according to the feasible region matrix, finally, considering high sampling frequency and high precision of a PMU measuring device, and correcting the abnormal wind power data by adopting a vertical comparison method. The wind power abnormal data identification method with the PMU measuring device has excellent identification speed and precision, plays a supporting role in wind power data processing and mining, and can provide auxiliary decision for the operation of a fan.
Drawings
FIG. 1 is a wind turbine generator wind speed-power scatter plot;
FIG. 2 is a flow chart of the method of the present invention;
FIG. 3 is a graph showing the improved fuzzy mean two-dimensional clustering result of the present invention;
FIG. 4 is a graph showing the improved fuzzy mean three-dimensional clustering result of the present invention;
FIG. 5 is a line graph of presence of anomaly data;
fig. 6 is an effect diagram after correction of abnormal data points.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
According to the wind power abnormal data identification method comprising the PMU device, the influence of wind speed on wind power abnormal data identification is calculated, clustering analysis is carried out on wind power big data through improving an FCM algorithm, a statistical method is adopted to calculate a clustering result to obtain a feasible region matrix, whether measured data are abnormal is further judged sequentially, and finally the characteristic of high frequency and high precision of the PMU measuring device is utilized to selectively correct the wind power abnormal data through a vertical comparison method.
The method comprises the following main steps:
step 1, classifying wind power measurement data according to the wind speed;
step 2, obtaining the cluster number and an initial cluster center by using a subtractive mean clustering algorithm;
step 3, carrying out fuzzy mean clustering based on the cluster number obtained in the step 2 and the initial cluster center;
step 4, calculating a feasible domain matrix of each cluster based on a statistical rule;
step 5, distinguishing the abnormal data by using the feasible domain matrix of each cluster obtained in the step 4;
and 6, correcting wind power abnormal data by adopting a vertical comparison method.
Specifically, the first step includes: installing a PMU measuring device at a sampling node, wherein the PMU has the characteristics of high sampling frequency and high precision compared with other measuring devices such as SCADA; meanwhile, the wind speed is considered not to be suddenly changed in continuous time, so wind power measurement data are classified according to the wind speed and are mainly classified into three types of wind speed lower than cut-in wind speed and wind speed higher than cut-out wind speed and between the two types.
The fan output is directly related to the wind speed, and if the wind power data is directly clustered without considering the wind speed, erroneous judgment on abnormal data can be caused. For example, when the wind speed is lower than the cut-in speed or higher than the cut-out speed, the fan output is 0, and if the wind speed is not classified, the wind speed is judged to be abnormal data.
Wherein P represents the output power of the wind turbine generator; c (C) p Representing wind energy utilization rate; ρ represents the air density; a represents the effective area swept by the fan blade; v represents the measured wind speed; v in 、v out 、v n The cut-in wind speed, the cut-out wind speed and the rated wind speed are indicated, respectively.
Step two, obtaining the cluster number and an initial cluster center by using a subtractive mean value clustering algorithm, which concretely comprises the following sub-steps:
1) Given sample set x= { X 1 ,x 2 ,x 3 ,...,x n Assuming that the number of clusters is k (arbitrary value), initializing k=0, each sample point x i The density index of (2) is:
wherein r is a For the radius of the field (a preset parameter), calculating the density index of each sample point one by one, taking the sample point with the maximum density index as the first clustering center, and making k=1.
2) On the basis of 1), the density index of each sample point is corrected according to the following
In which x is ck For the kth selected cluster center, D ck Is the corresponding index density; r is (r) b Is a predetermined positive parameter. Selecting the data point with the highest density indexAs a new cluster center, the corresponding density index is +.>And let k=k+1.
3) Judging whether or not the following is true
If so, the operation is ended, and k is the cluster number and x is obtained ck Is a clustering center; if not, return to 2). Parameter r in subtractive clustering a 、r b Defined as the radius of the field, data points outside the radius have very little effect on the density index of the point. The following formula is the calculation method of the radius of the field:
the cluster center and the cluster number obtained after subtraction clustering are used as initial values of the step three FCM clustering, so that the speed and the accuracy of the clustering are improved.
Step three, fuzzy mean clustering, specifically comprising: 1) Establishing a fuzzy membership matrix; 2) Clustering according to membership matrix weight. The constraint conditions of the membership matrix are as follows:
v in i For the ith cluster center, V is the set of cluster centers. w is a weighted index, the value range is [1, + ], the value of w can influence the fuzzy degree of clustering, and x j For the first sample element, u ij Representing the membership of the jth sample in the ith class.
In the FCM clustering algorithm, membership matrix elements and initial cluster centers are generated by random assignment. To achieve the objectMinimum value min { J of function w (U, V) } multiple iterations are required.
In the above, x j Is the actual value, v i For the cluster center of the numerical value, uij w The number belongs to the membership degree of the cluster. To obtain the optimal solution for the above objective function, the membership matrix element may be optimized by the lagrangian multiplier method as follows,
the membership matrix element is solved by the Lagrange multiplier method, and constraint conditions are as follows:
in the above formula, w is a weight coefficient, alpha is Lagrangian multiplier, and u ij 、x j 、v i As described in formula (13). The following formula can be obtained:
thereby:
updating the elements of the membership matrix element by an iterative formula of the membership matrix element:
similarly, the iterative formula for the cluster center can be found as follows:
the update formula of the cluster center obtained by calculation is as follows:
in the formulae (15) to (20), u ij 、x j 、v i Are membership, actual numerical value and clustering center respectively. Repeating the calculation steps until the cluster center is converged within a certain value range epsilon.
Step four, calculating a feasible domain matrix of each cluster according to a statistical rule, wherein the feasible domain matrix comprises 1) calculating a sample mean value and a standard deviation; 2) A sample feasible region range is calculated. Specifically, for a sample set, x= { X 1 ,x 2 ,...,x n Wind power data of which sample mean is represented by u and standard deviation is delta 1 . After adding some abnormal data, the sample mean becomesStandard deviation becomes delta 2 . The relationship between the sample mean values before and after the addition of the abnormal data is as follows:
for a new sample set x= { X 1 ,x 2 ,...,x n ,x n+1 The relationship between the standard deviations is as follows:
after finishing, the method can obtain:
the above formula is taken into the previous formula calculation to obtain:
regarding the above as x n+1 Solving the unitary quadratic equation of (2) to obtain:
k in the formula 1 、K 2 、K 3 The meaning is as follows:
the range threshold of the normal data can be obtained through the solution, and the upper and lower boundaries of the range threshold are respectively as follows:and->For data exceeding the upper and lower boundaries, the measurement standard deviation is significantly larger than the normal measurement standard deviation, and can be determined as abnormal data.
And fifthly, judging abnormal data in the feasible domain matrix of each cluster obtained in the step four by utilizing the feasible domain matrix. The method specifically comprises the following steps: 1) Judging each wind power data in sequence by using the feasible domain matrix obtained in the step four, wherein the abnormal data is the data exceeding the threshold value boundary; 2) And collecting all abnormal data to obtain a wind power abnormal data set.
Step six, correcting abnormal data by using a vertical comparison method, wherein a correction formula is as follows:
in which x is ij Sample data representing a unit at the j-th moment of the i-th day,sample data representing the ith-1 th day and jth time of the unit, < + >>Sample data representing the ith-1 th day, jth-1 th moment, deltax of the unit i,j-1 The difference between the sample data at the j-th time on the i-1 day and the sample data at the j-th time on the i-th day is shown.
The following description of the embodiments of the invention is further presented in conjunction with the drawings and the actual examples.
Step one, classifying wind power data according to the wind speed. And considering the influence of wind speed on wind power abnormal data identification, classifying wind power data into three types. As shown in fig. 1, the data on the coordinate axis is the wind-discarding data, that is, the wind-electricity data when the wind speed is lower than the cut-in speed or higher than the cut-out speed; the data deviating from the power curve are outlier isolated data; the data at the edges of the power curve are cluster bias data.
And step two, obtaining the cluster number and an initial cluster center by using a subtractive mean value clustering algorithm. The invention utilizes the subsubust function in the MATLAB clustering tool box to develop the subtraction mean value clustering. The subtractive clustering method assumes that each data point is a potential cluster center and calculates a likelihood measure that each data point is defined as a cluster center based on the density of surrounding data points. The algorithm specifically performs the following operations: the data point with the highest potential is selected as the first cluster center and all data points near the first cluster center (determined by radius) are deleted to determine the next data cluster and its center location. This process is iterated until all the data is within the radius of the cluster center. The subtraction clustering operation is carried out by taking active, reactive and wind speed three-dimensional data of 10 months in 2018 of a No. 2 unit of the east China sea bridge wind power plant as a sample, and the result is as follows:
table 1 Donghai wind farm No. 2 unit 18 year 10 month three-dimensional subtractive clustering results
And thirdly, taking the clustering center and the clustering number obtained in the second step as initial values of FCM clustering, and carrying out clustering operation by means of a Matlab clustering tool box. The operation flow is shown in fig. 2, and the clustering result is shown in fig. 3 and 4.
Step four, based on the wind power data clustering obtained in the step three, the feasible region range of the wind power data of each cluster is calculated respectively, and the result is shown in the following table
Table 2 results of calculation of the feasible region ranges for each cluster
And fifthly, judging the abnormal data one by one according to the feasible threshold value range of various data to obtain an abnormal data set. For example, if a certain active sampling value in cluster 1 is 300kW and is not within the feasible region [447.9, 667.5], determining that the data is abnormal; otherwise, if 500kW is within the feasible region, judging that the data is normal.
And step six, correcting the abnormal data by adopting a vertical comparison method for the data in the abnormal data set. To examine the correction effect, a number of abnormal data were artificially added, and the abnormal data correction deviation rate was calculated, and the results are shown in table 3.
In the above formula, p is the deviation rate of actual value and true value, and x i Is the actual value, x j Is a true value.
TABLE 3 error Rate of corrected anomalous data
After the correction is completed, the abnormal data is circularly identified until the identification accuracy reaches a stable value.
B=N 1 /N 2 (29)
In the above, N is the identification accuracy 1 To successfully identify the total number of abnormal data, N 2 Is the total number of abnormal data samples. The calculation result of the identification accuracy is shown in the following table:
TABLE 4 total number of successful identifications before and after correction
It should be noted that the foregoing examples are provided merely for the purpose of illustrating embodiments of the present invention and are not to be construed as limiting the scope of the invention or as limiting the structure of the invention in any way. It should be noted that variations and modifications can be made by those skilled in the art without departing from the spirit of the invention, which falls within the scope of the invention.
According to the method, the influence of wind speed on wind power abnormal data identification is considered, wind power data is initially classified based on wind speed, improved fuzzy C-means clustering analysis is carried out, and then statistical methods are adopted to calculate each cluster.
While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions of equivalents may be made and equivalents will be apparent to those skilled in the art without departing from the scope of the invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (4)
1. A wind power abnormal data identification method containing a synchronous phasor measurement device is characterized in that,
the method specifically comprises the following steps:
step 1, classifying wind power operation big data according to the wind speed;
step 2, obtaining the number of clusters and an initial cluster center by using a subtractive mean clustering algorithm;
step 3, carrying out fuzzy mean clustering based on the number of the clusters obtained in the step 2 and the initial cluster center;
step 4, calculating a feasible domain matrix of each cluster by using a statistical rule;
step 5, distinguishing abnormal data by using the feasible domain matrix of each cluster obtained in the step 4;
step 6, correcting the abnormal data identified in the step 5 by adopting a vertical comparison method;
step 2, obtaining the cluster number and an initial cluster center by using a subtractive mean value clustering algorithm, which specifically comprises the following sub-steps:
a) Given sample set x= { X 1 ,x 2 ,x 3 ,…,x n Setting the number of clusters as k, initializing k=0, and then each sample point x i The density index of (2) is:
wherein r is a Calculating density indexes of all sample points one by one for the radius of the field, taking the sample point with the largest density index as a first clustering center, and enabling k=1;
b) Correcting the density index of each sample point according to the following
In which x is ck For the kth selected cluster center, D ck Is the corresponding index density; r is (r) b For the radius of the field, a predetermined positive parameter is selected as the data point with the highest density indexAs a means ofA new cluster center, which corresponds to the density index ofAnd let k=k+1;
c) Judging whether or not the following is true
If so, the operation is ended, and k is the cluster number and x is obtained ck Is a clustering center; if not, returning to the step B); calculating the radius r of the field according to the formula (4) or (5) a 、r b :
The specific process of the step 4 is as follows:
calculating a sample mean value:
in which x is i Represents the wind power data of a certain type at the ith moment,representing the average value of the wind power data, wherein n is the sampling number;
randomly adding 1 abnormal data into the raw sample data, respectively calculating the raw sample standard deviation and the sample standard deviation after the abnormal data are added, and calculating the ratio a of the front standard deviation and the rear standard deviation 1 :
Delta in 2 、For standard deviation and mean after addition of anomaly data, delta 1 U is the standard deviation and the mean of the original sample;
calculating a feasible region range:
K 1 、K 2 、K 3 the meaning is as follows:
the step 6 comprises the following steps:
correction of abnormal data by vertical comparison
In which x is ij Sample data representing a unit at the j-th moment of the i-th day,sample data representing the ith-1 th day and jth time of the unit, < + >>Sample data of the ith-1 day and the jth-1 moment of the unit are represented, and Deltax i,j-1 The difference between the sample data at the j-th time on the i-1 day and the sample data at the j-th time on the i-th day is shown.
2. The method for identifying abnormal wind power data including a synchrophasor measurement device according to claim 1, wherein the step 1 comprises: wind power big data is classified according to wind speed, and wind power data lower than cut-in wind speed and higher than cut-out wind speed are paid attention to.
3. The method for identifying abnormal wind power data with synchronous phasor measurement device according to claim 1, wherein the step 3 uses FCM algorithm to calculate the weights of the belonging classes, and selects the class with the largest weight as the belonging class.
4. The method for identifying wind power abnormal data comprising the synchronous phasor measurement device according to claim 1, wherein the method is based on the proposed abnormal data identification and correction model, adopts simulation calculation examples and builds a simulation model for verification analysis.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910892258.XA CN110674864B (en) | 2019-09-20 | 2019-09-20 | Wind power abnormal data identification method comprising synchronous phasor measurement device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910892258.XA CN110674864B (en) | 2019-09-20 | 2019-09-20 | Wind power abnormal data identification method comprising synchronous phasor measurement device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110674864A CN110674864A (en) | 2020-01-10 |
CN110674864B true CN110674864B (en) | 2024-03-15 |
Family
ID=69078464
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910892258.XA Active CN110674864B (en) | 2019-09-20 | 2019-09-20 | Wind power abnormal data identification method comprising synchronous phasor measurement device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110674864B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115600119B (en) * | 2022-12-13 | 2023-06-16 | 青岛左岸数据科技有限公司 | Data processing method and system suitable for wind power generation |
CN116881746B (en) * | 2023-09-08 | 2023-11-14 | 国网江苏省电力有限公司常州供电分公司 | Identification method and identification device for abnormal data in electric power system |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105550943A (en) * | 2016-01-18 | 2016-05-04 | 重庆大学 | Method for identifying abnormity of state parameters of wind turbine generator based on fuzzy comprehensive evaluation |
CN106055918A (en) * | 2016-07-26 | 2016-10-26 | 天津大学 | Power system load data identification and recovery method |
CN106548270A (en) * | 2016-09-30 | 2017-03-29 | 许昌许继软件技术有限公司 | A kind of photovoltaic plant power anomalous data identification method and device |
CN107067100A (en) * | 2017-01-25 | 2017-08-18 | 国网冀北电力有限公司 | Wind power anomalous data identification method and device for identifying |
CN107392304A (en) * | 2017-08-04 | 2017-11-24 | 中国电力科学研究院 | A kind of Wind turbines disorder data recognition method and device |
CN107808209A (en) * | 2017-09-11 | 2018-03-16 | 重庆大学 | Abnormal data of wind power plant discrimination method based on weighting kNN distances |
CN109086793A (en) * | 2018-06-27 | 2018-12-25 | 东北大学 | A kind of abnormality recognition method of wind-driven generator |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9379951B2 (en) * | 2014-01-10 | 2016-06-28 | Instep Software, Llc | Method and apparatus for detection of anomalies in integrated parameter systems |
-
2019
- 2019-09-20 CN CN201910892258.XA patent/CN110674864B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105550943A (en) * | 2016-01-18 | 2016-05-04 | 重庆大学 | Method for identifying abnormity of state parameters of wind turbine generator based on fuzzy comprehensive evaluation |
CN106055918A (en) * | 2016-07-26 | 2016-10-26 | 天津大学 | Power system load data identification and recovery method |
CN106548270A (en) * | 2016-09-30 | 2017-03-29 | 许昌许继软件技术有限公司 | A kind of photovoltaic plant power anomalous data identification method and device |
CN107067100A (en) * | 2017-01-25 | 2017-08-18 | 国网冀北电力有限公司 | Wind power anomalous data identification method and device for identifying |
CN107392304A (en) * | 2017-08-04 | 2017-11-24 | 中国电力科学研究院 | A kind of Wind turbines disorder data recognition method and device |
CN107808209A (en) * | 2017-09-11 | 2018-03-16 | 重庆大学 | Abnormal data of wind power plant discrimination method based on weighting kNN distances |
CN109086793A (en) * | 2018-06-27 | 2018-12-25 | 东北大学 | A kind of abnormality recognition method of wind-driven generator |
Non-Patent Citations (2)
Title |
---|
The anomalous data identification study of reactive power optimization system based on big data;Sheng Wanxing 等;2016 International Conference on Probabilistic Methods Applied to Power Systems (PMAPS);全文 * |
基于用电大数据的用电异常状态辨识方法;张秋雁 等;电力大数据;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN110674864A (en) | 2020-01-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103887815B (en) | Based on wind energy turbine set parameter identification and the Dynamic Equivalence of service data | |
CN109751206B (en) | Fan blade icing fault prediction method and device and storage medium | |
CN109086928A (en) | Photovoltaic plant realtime power prediction technique based on SAGA-FCM-LSSVM model | |
CN106570790B (en) | Wind power plant output data restoration method considering wind speed data segmentation characteristics | |
CN103942736B (en) | A kind of wind power plant multimachine equivalent modeling method | |
CN110503153B (en) | Photovoltaic system fault diagnosis method based on differential evolution algorithm and support vector machine | |
CN105389634A (en) | Combined short-term wind power prediction system and method | |
CN110674864B (en) | Wind power abnormal data identification method comprising synchronous phasor measurement device | |
CN103489046A (en) | Method for predicting wind power plant short-term power | |
CN105825002B (en) | A kind of wind power plant dynamic equivalent modeling method based on dynamic Gray Association Analysis | |
CN105719029A (en) | Combined wind power prediction method based on wind speed fluctuation characteristic extraction | |
CN112186761B (en) | Wind power scene generation method and system based on probability distribution | |
CN106548256A (en) | A kind of method and system of wind energy turbine set space-time dynamic correlation modeling | |
CN109167387A (en) | Wind field wind power forecasting method | |
CN105701562B (en) | Training method, applicable method for predicting generated power and respective system | |
WO2020097979A1 (en) | Wind farm control parameter optimization method and system | |
CN103996079A (en) | Wind power weighting predication method based on conditional probability | |
CN112288157A (en) | Wind power plant power prediction method based on fuzzy clustering and deep reinforcement learning | |
CN113991711B (en) | Capacity configuration method for energy storage system of photovoltaic power station | |
CN115204444A (en) | Photovoltaic power prediction method based on improved cluster analysis and fusion integration algorithm | |
WO2024041409A1 (en) | Method and apparatus for determining representative wind generating set, and control method and apparatus | |
CN108460228B (en) | Wind power plant equivalence method based on multi-objective optimization algorithm | |
CN109193791B (en) | Wind power convergence tendency state-based quantification method based on improved shape value | |
Li et al. | Wind power forecasting based on time series and neural network | |
Bao et al. | Iterative modeling of wind turbine power curve based on least‐square B‐spline approximation |
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 |