CN112183621A - Transformer state abnormity detection method driven by power quality monitoring data - Google Patents
Transformer state abnormity detection method driven by power quality monitoring data Download PDFInfo
- Publication number
- CN112183621A CN112183621A CN202011033652.7A CN202011033652A CN112183621A CN 112183621 A CN112183621 A CN 112183621A CN 202011033652 A CN202011033652 A CN 202011033652A CN 112183621 A CN112183621 A CN 112183621A
- Authority
- CN
- China
- Prior art keywords
- cluster
- quantity data
- electrical quantity
- characteristic attribute
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 20
- 238000001514 detection method Methods 0.000 title abstract description 13
- 230000002159 abnormal effect Effects 0.000 claims description 12
- 238000000034 method Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 description 10
- HSFWRNGVRCDJHI-UHFFFAOYSA-N alpha-acetylene Natural products C#C HSFWRNGVRCDJHI-UHFFFAOYSA-N 0.000 description 3
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000002547 anomalous effect Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 125000002534 ethynyl group Chemical group [H]C#C* 0.000 description 1
- 238000012706 support-vector machine 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
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Probability & Statistics with Applications (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Evolutionary Computation (AREA)
- Pure & Applied Mathematics (AREA)
- Algebra (AREA)
- Operations Research (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
According to the transformer state abnormity detection method driven by the power quality monitoring data, the transformer electrical quantity monitoring data in the existing detection system is utilized, the clustering algorithm is adopted, the complex correlation among the electrical quantities is simplified, the big data with obvious deviation can be well processed, and the transformer state abnormity detection is effectively realized.
Description
Technical Field
The invention relates to the technical field of power detection, in particular to a method for detecting abnormal states of a transformer driven by power quality monitoring data.
Background
The power transformer is used as a key junction device of a power system, and the safe and stable operation of the power transformer is a necessary foundation for ensuring the normal supply of high-quality power and the normal operation of social life. The traditional planned maintenance mode is troublesome and laborious, the power transformer state maintenance is a novel maintenance mode which is based on the evaluation equipment state and on the basis of the prediction state development trend, and the maintenance mode can effectively avoid the problems of insufficient maintenance, excessive maintenance and the like while the reliability of the equipment is improved. The key to apply the condition maintenance mode is to realize the condition estimation of the power transformer. Besides, the abnormal state of the transformer can be recognized in advance in the operation process, the transformer is timely subjected to state maintenance, the reliability of equipment can be improved, the possibility of accidents is reduced, and the problems of insufficient maintenance, excessive maintenance and the like are effectively avoided.
The state evaluation method of the power transformer mainly comprises an analytic hierarchy process, a Bayesian network, a support vector machine, an artificial neural network and the like. With the continuous development of the state monitoring technology, the identification of the transformer state by using the online monitoring parameters is more and more widely applied. However, the existing transformer state anomaly detection algorithm mainly uses non-electrical quantity data such as top-layer oil temperature, load, methane (CH4) volume fraction, ambient temperature, acetylene (C2H2) volume fraction and the like, and mass transformer electrical quantity monitoring data accumulated in the existing monitoring system cannot be utilized, so that the state identification algorithm is difficult to directly associate with systems such as harmonic monitoring and scheduling, and much inconvenience is brought to the state detection of the transformer.
Disclosure of Invention
In view of the above, it is desirable to provide a method for detecting abnormal transformer state driven by power quality monitoring data, which detects the transformer state by using the transformer electrical quantity monitoring data in the existing detection system.
A method for detecting abnormal state of a transformer driven by power quality monitoring data comprises the following steps:
step S001, detecting the electrical quantity data of the transformer in a normal state, accumulating the electrical quantity data of the transformer in the normal state to establish a database, and carrying out standardization processing on the characteristic attribute of the electrical quantity data in the electrical quantity database to obtain a characteristic attribute value of the electrical quantity data;
step S002, dividing the characteristic attribute values of the electrical quantity data after the standardization processing into a plurality of clustering clusters through a clustering algorithm;
step S003, taking the median in each cluster as a new cluster center, clustering the characteristic attribute values again, and evaluating the cluster clusters through a clustering objective function until an optimal cluster and an optimal cluster center are obtained, wherein the optimal cluster is taken as a normal data cluster and each normal data cluster center;
step S004, calculating the distance between the characteristic attribute value in each normal data cluster and the center of each normal data cluster, and determining the farthest distance between the characteristic attribute value and the center of each normal data cluster as the distance threshold of each cluster;
step S005, inputting the real-time detected electric quantity data, calculating the distance between the electric quantity data and each cluster center, and finding out the cluster center with the shortest distance to the electric quantity data, wherein the cluster where the cluster center is located is the cluster to which the electric quantity data belongs;
step S006 is to compare the shortest distance between the real-time detected electrical quantity data and the center of the cluster with the distance threshold of the cluster to which the electrical quantity data belongs, and if the distance between the real-time detected electrical quantity data and the cluster to which the electrical quantity data belongs is greater than the distance threshold, the electrical quantity data is abnormal data, otherwise, the electrical quantity data is normal data.
Preferably, in step S001, "normalizing the characteristic attributes of all the electrical quantity data" includes the specific steps of:
step S101, calculating an average value of the electrical quantity data:
in the formula: x is the electric quantity data, and n is the number of the electric quantity data;
step S102, calculating an average value of absolute deviations of the electrical quantity data:
step S103, normalizing the characteristic attributes of all the electrical quantity data to obtain a characteristic attribute value of the electrical quantity data:
preferably, in step S002, "the electrical quantity data after the normalization processing is divided into a plurality of clusters by a clustering algorithm" includes:
step S201, dividing all the feature attribute values into j cluster clusters according to the types of the detected parameters, where the cluster clusters are: w1，W2，…，Wj；
Step S202, j values are searched in the characteristic attribute values to serve as initial clustering centers of the clustering clusters, and the initial clustering centers are c in sequence1，c2，…，cjWherein
In the above formula, njIs a cluster WjThe number of characteristic attribute values in (1);
step S203, comparing each characteristic attribute value with each initial cluster center, calculating the distance between the characteristic attribute value and each initial cluster center, and distributing the characteristic attribute value closest to the cluster center to the cluster where the initial cluster center is located until all the characteristic attribute values are distributed.
Preferably, in step S003, the clustering objective function is:
in the formula: dij(xj,cj) Representing a characteristic attribute value xjWith new cluster center cjThe Euclidean distance between; the objective function J represents the sum of the distances between each characteristic attribute value and the cluster center;
the smaller the J value is, the more compact the characteristic attribute value in each cluster is, the optimal cluster center is obtained by continuously optimizing the J value, when the minimum value obtained by the clustering objective function meets the precision requirement or the calculated amount reaches the upper limit of the iteration times, the cluster center corresponding to the clustering objective function is the optimal cluster center, and the characteristic attribute value x in the clustering objective functionjThe formed cluster is the optimal cluster.
Has the advantages that: according to the transformer state abnormity detection method driven by the power quality monitoring data, the transformer electrical quantity monitoring data in the existing detection system is utilized, the clustering algorithm is adopted, the complex correlation among the electrical quantities is simplified, the big data with obvious deviation can be well processed, and the transformer state abnormity detection is effectively realized.
Detailed Description
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following description will be given with reference to the embodiments.
A method for detecting abnormal state of a transformer driven by power quality monitoring data comprises the following steps:
step S001, detecting the electrical quantity data of the transformer in a normal state, accumulating the electrical quantity data of the transformer in the normal state to establish a database, and carrying out standardization processing on the characteristic attribute of the electrical quantity data in the electrical quantity database to obtain a characteristic attribute value of the electrical quantity data;
step S002, dividing the characteristic attribute values of the electrical quantity data after the standardization processing into a plurality of clustering clusters through a clustering algorithm;
step S003, taking the median in each cluster as a new cluster center, clustering the characteristic attribute values again, and evaluating the cluster clusters through a clustering objective function until an optimal cluster and an optimal cluster center are obtained, wherein the optimal cluster is taken as a normal data cluster and each normal data cluster center;
step S004, calculating the distance between the characteristic attribute value in each normal data cluster and the center of each normal data cluster, and determining the farthest distance between the characteristic attribute value and the center of each normal data cluster as the distance threshold of each cluster;
step S005, inputting the real-time detected electric quantity data, calculating the distance between the electric quantity data and each cluster center, and finding out the cluster center with the shortest distance to the electric quantity data, wherein the cluster where the cluster center is located is the cluster to which the electric quantity data belongs;
step S006 is to compare the shortest distance between the real-time detected electrical quantity data and the center of the cluster with the distance threshold of the cluster to which the electrical quantity data belongs, and if the distance between the real-time detected electrical quantity data and the cluster to which the electrical quantity data belongs is greater than the distance threshold, the electrical quantity data is abnormal data, otherwise, the electrical quantity data is normal data.
Further, in step S001, "standardizing the characteristic attributes of all the electrical quantity data" includes the specific steps of:
step S101, calculating an average value of the electrical quantity data:
in the formula: x is the electric quantity data, and n is the number of the electric quantity data;
step S102, calculating an average value of absolute deviations of the electrical quantity data:
step S103, normalizing the characteristic attributes of all the electrical quantity data to obtain a characteristic attribute value of the electrical quantity data:
further, in step S002, "the electrical quantity data after the normalization processing is divided into a plurality of clusters by the clustering algorithm" includes:
step S201, dividing all the feature attribute values into j cluster clusters according to the types of the detected parameters, where the cluster clusters are: w1，W2，…，Wj；
Step S202, j values are searched in the characteristic attribute values to serve as initial clustering centers of the clustering clusters, and the initial clustering centers are c in sequence1，c2，…，cjWherein
In the above formula, njIs a cluster WjThe number of characteristic attribute values in (1);
step S203, comparing each characteristic attribute value with each initial cluster center, calculating the distance between the characteristic attribute value and each initial cluster center, and distributing the characteristic attribute value closest to the cluster center to the cluster where the initial cluster center is located until all the characteristic attribute values are distributed.
Further, in step S003, the clustering objective function is:
in the formula: dij(xj,cj) Representing a characteristic attribute value xjWith new cluster center cjThe Euclidean distance between; the objective function J represents the sum of the distances between each characteristic attribute value and the cluster center;
the smaller the J value is, the more compact the characteristic attribute value in each cluster is, the optimal cluster center is obtained by continuously optimizing the J value, when the minimum value obtained by the clustering objective function meets the precision requirement or the calculated amount reaches the upper limit of the iteration times, the cluster center corresponding to the clustering objective function is the optimal cluster center, and the characteristic attribute value x in the clustering objective functionjThe formed cluster is the optimal cluster.
The following will be described in conjunction with specific embodiments:
detecting the electrical quantity data of the transformer in the normal state, accumulating the electrical quantity data of the transformer in the normal state to establish a database, standardizing the characteristic attribute of the electrical quantity data in the electrical quantity database to obtain the characteristic attribute value of the electrical quantity data, for example, obtaining the characteristic attribute value of 120 integers from 1 to 120, dividing the characteristic attribute value into a plurality of cluster clusters according to the characteristic attribute of the electrical quantity, for example, 12 clusters, the 12 clusters including a primary side AB phase voltage, a primary side BC phase voltage, a primary side CA phase voltage, a secondary side A phase voltage, a secondary side B phase voltage, a secondary side C phase voltage, an A phase active power, a B phase active power, a C phase active power, an A phase reactive power, a B phase reactive power, a C phase reactive power, and finding 12 characteristic attribute values capable of representing 12 electrical quantity characteristic attributes in the database as initial clustering centers of the 12 clustering clusters. Such as 2, 13, 24, 35, 46, 57, 68, 79, 81, 93, 105, 116. The distance from each feature attribute value to each initial cluster center is calculated, for example, 5 to the initial cluster center is calculated. The distance between the characteristic attribute value 5 and the initial clustering center 2 is shortest, and then the characteristic attribute value 5 is attributed to the clustering cluster where the initial clustering center 2 is located until all the characteristic attribute values 1, 2, 3, 4, 5, 6 and 7 are found to be used as one clustering cluster, and other characteristic attribute values are classified similarly and are not repeated. And then taking the median 4 of the cluster characteristic attribute values 1, 2, 3, 4, 5, 6 and 7 as a new cluster center, calculating the distance between the 120 integers from 1 to 120 and the new cluster center including the characteristic attribute value 4, and determining the new cluster, for example, taking 4 as the cluster center, taking the new cluster as 1, 2, 3, 4, 5, 6, 7 and 8, and re-determining other new cluster in the same way, without repeating the description. And evaluating the clustering clusters through the clustering objective function until the optimal clustering clusters and the optimal clustering cluster centers are obtained, wherein the optimal clustering clusters are used as the normal data clustering clusters and the centers of each normal data clustering cluster. For example, the 120 integers of 1 to 120 finally obtain the optimal cluster clusters of 1 to 9, 11 to 19, 111 to 119, and the optimal cluster centers of 5, 15, 115, 10, 20 and other integers are excluded by the clustering objective function. And the optimal clustering cluster is a normal data clustering cluster, the optimal clustering center is a normal data clustering center, and the distance threshold is 4. Then, the distance between the real-time data of the online detection and each cluster center is calculated, for example, the real-time data is 13.2, the distance between the real-time data and the cluster center 15 is the nearest distance, and the distance is less than 4, then the data is normal data; if the real-time data is 20.3, which is closest to the cluster center 25 and is greater than 4, then the data is anomalous.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.
Claims (4)
1. A method for detecting abnormal state of a transformer driven by power quality monitoring data is characterized by comprising the following steps: the method comprises the following steps:
step S001, detecting the electrical quantity data of the transformer in a normal state, accumulating the electrical quantity data of the transformer in the normal state to establish a database, and carrying out standardization processing on the characteristic attribute of the electrical quantity data in the electrical quantity database to obtain a characteristic attribute value of the electrical quantity data;
step S002, dividing the characteristic attribute values of the electrical quantity data after the standardization processing into a plurality of clustering clusters through a clustering algorithm;
step S003, taking the median in each cluster as a new cluster center, clustering the characteristic attribute values again, and evaluating the cluster clusters through a clustering objective function until an optimal cluster and an optimal cluster center are obtained, wherein the optimal cluster is taken as a normal data cluster and each normal data cluster center;
step S004, calculating the distance between the characteristic attribute value in each normal data cluster and the center of each normal data cluster, and determining the farthest distance between the characteristic attribute value and the center of each normal data cluster as the distance threshold of each cluster;
step S005, inputting the real-time detected electric quantity data, calculating the distance between the electric quantity data and each cluster center, and finding out the cluster center with the shortest distance to the electric quantity data, wherein the cluster where the cluster center is located is the cluster to which the electric quantity data belongs;
step S006 is to compare the shortest distance between the real-time detected electrical quantity data and the center of the cluster with the distance threshold of the cluster to which the electrical quantity data belongs, and if the distance between the real-time detected electrical quantity data and the cluster to which the electrical quantity data belongs is greater than the distance threshold, the electrical quantity data is abnormal data, otherwise, the electrical quantity data is normal data.
2. The method for detecting abnormal state of a transformer driven by power quality monitoring data according to claim 1, wherein: in step S001, "standardizing the characteristic attributes of all the electrical quantity data" includes the specific steps of:
step S101, calculating an average value of the electrical quantity data:
in the formula: x is the electric quantity data, and n is the number of the electric quantity data;
step S102, calculating an average value of absolute deviations of the electrical quantity data:
step S103, normalizing the characteristic attributes of all the electrical quantity data to obtain a characteristic attribute value of the electrical quantity data:
3. the method for detecting abnormal state of a transformer driven by power quality monitoring data according to claim 1, wherein: in step S002, "the electrical quantity data after the normalization processing is divided into a plurality of clusters by the clustering algorithm" includes:
step S201, dividing all the feature attribute values into j cluster clusters according to the types of the detected parameters, where the cluster clusters are: w1，W2，…，Wj；
Step S202, j values are searched in the characteristic attribute values to serve as initial clustering centers of the clustering clusters, and the initial clustering centers are c in sequence1，c2，…，cjWherein
In the above formula, njIs a cluster WjThe number of characteristic attribute values in (1);
step S203, comparing each characteristic attribute value with each initial cluster center, calculating the distance between the characteristic attribute value and each initial cluster center, and distributing the characteristic attribute value closest to the cluster center to the cluster where the initial cluster center is located until all the characteristic attribute values are distributed.
4. The method for detecting abnormal state of a transformer driven by power quality monitoring data according to claim 1, wherein: in step S003, the clustering objective function is:
in the formula: dij(xj,cj) Representing a characteristic attribute value xjWith new cluster center cjThe Euclidean distance between; the objective function J represents the sum of the distances between each characteristic attribute value and the cluster center;
the smaller the J value is, the more compact the characteristic attribute value in each cluster is, the optimal cluster center is obtained by continuously optimizing the J value, when the minimum value obtained by the clustering objective function meets the precision requirement or the calculated amount reaches the upper limit of the iteration times, the cluster center corresponding to the clustering objective function is the optimal cluster center, and the characteristic attribute value x in the clustering objective functionjThe formed cluster is the optimal cluster.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011033652.7A CN112183621A (en) | 2020-09-27 | 2020-09-27 | Transformer state abnormity detection method driven by power quality monitoring data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011033652.7A CN112183621A (en) | 2020-09-27 | 2020-09-27 | Transformer state abnormity detection method driven by power quality monitoring data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112183621A true CN112183621A (en) | 2021-01-05 |
Family
ID=73943699
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011033652.7A Pending CN112183621A (en) | 2020-09-27 | 2020-09-27 | Transformer state abnormity detection method driven by power quality monitoring data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112183621A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101908065A (en) * | 2010-07-27 | 2010-12-08 | 浙江大学 | On-line attribute abnormal point detecting method for supporting dynamic update |
| CN104330675A (en) * | 2014-11-17 | 2015-02-04 | 国家电网公司 | Multivariate time series based power transformation equipment online monitoring and analysis system and method thereof |
| CN107528823A (en) * | 2017-07-03 | 2017-12-29 | 中山大学 | A kind of network anomaly detection method based on improved K Means clustering algorithms |
| CN108614895A (en) * | 2018-05-10 | 2018-10-02 | 中国移动通信集团海南有限公司 | The recognition methods of abnormal data access behavior and data processing equipment |
| CN110825548A (en) * | 2019-10-24 | 2020-02-21 | 新华三信息安全技术有限公司 | Anomaly detection method, model training method and related device |
-
2020
- 2020-09-27 CN CN202011033652.7A patent/CN112183621A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101908065A (en) * | 2010-07-27 | 2010-12-08 | 浙江大学 | On-line attribute abnormal point detecting method for supporting dynamic update |
| CN104330675A (en) * | 2014-11-17 | 2015-02-04 | 国家电网公司 | Multivariate time series based power transformation equipment online monitoring and analysis system and method thereof |
| CN107528823A (en) * | 2017-07-03 | 2017-12-29 | 中山大学 | A kind of network anomaly detection method based on improved K Means clustering algorithms |
| CN108614895A (en) * | 2018-05-10 | 2018-10-02 | 中国移动通信集团海南有限公司 | The recognition methods of abnormal data access behavior and data processing equipment |
| CN110825548A (en) * | 2019-10-24 | 2020-02-21 | 新华三信息安全技术有限公司 | Anomaly detection method, model training method and related device |
Non-Patent Citations (1)
| Title |
|---|
| 张丽梅 著: "《海洋经济评估模型与方法》", 31 October 2014 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111199016B (en) | Daily load curve clustering method for improving K-means based on DTW | |
| CN112699913A (en) | Transformer area household variable relation abnormity diagnosis method and device | |
| CN116031888B (en) | Dynamic load prediction-based power flow optimization method, system and storage medium | |
| CN114897247B (en) | Smart power grid distribution network dispatching operation management method based on big data | |
| CN113570200A (en) | Power grid operation state monitoring method and system based on multidimensional information | |
| CN115589000B (en) | Area voltage adjustment optimization method and system based on typical load classification | |
| CN117353751B (en) | Unmanned charging pile transaction data intelligent management system based on big data | |
| CN116148753A (en) | Intelligent electric energy meter operation error monitoring system | |
| CN115169719A (en) | Platform area state prediction model based on fuzzy clustering and BP neural network | |
| CN115204698A (en) | Real-time analysis method for power supply stability of low-voltage transformer area | |
| CN115640950A (en) | Method for diagnosing abnormal line loss of distribution network line in active area based on factor analysis | |
| CN107274025B (en) | System and method for realizing intelligent identification and management of power consumption mode | |
| CN116701919B (en) | Optimization monitoring method and system for gyro-type hydrogen fuel generator | |
| CN117154716B (en) | Planning method and system for accessing distributed power supply into power distribution network | |
| CN112163595B (en) | Method and device for acquiring typical electricity utilization mode of user and electronic equipment | |
| CN116743180B (en) | Intelligent storage method for energy storage power supply data | |
| CN112149052B (en) | Daily load curve clustering method based on PLR-DTW | |
| Qiu et al. | Comprehensive evaluation of power quality based on improved TOPSIS-RSR method | |
| CN112183621A (en) | Transformer state abnormity detection method driven by power quality monitoring data | |
| Rouwhorst et al. | Improving clustering-based forecasting of aggregated distribution transformer loadings with gradient boosting and feature selection | |
| Chen et al. | Interval prediction of photovoltaic power using improved NARX network and density peak clustering based on kernel mahalanobis distance | |
| CN115935212A (en) | Adjustable load clustering method and system based on longitudinal trend prediction | |
| CN114662056A (en) | Coating thickness control method and device and storage medium | |
| CN114372895A (en) | Method for analyzing power load curve based on kmeans algorithm | |
| Li et al. | Research and application of power system data anomaly identification based on time series and deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210105 |