CN109816031B - Transformer state evaluation clustering analysis method based on data imbalance measurement - Google Patents
Transformer state evaluation clustering analysis method based on data imbalance measurement Download PDFInfo
- Publication number
- CN109816031B CN109816031B CN201910093168.4A CN201910093168A CN109816031B CN 109816031 B CN109816031 B CN 109816031B CN 201910093168 A CN201910093168 A CN 201910093168A CN 109816031 B CN109816031 B CN 109816031B
- Authority
- CN
- China
- Prior art keywords
- data
- cluster
- fault
- samples
- transformer
- 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
Images
Abstract
The invention discloses a transformer state evaluation clustering analysis method based on data imbalance measurement, which comprises the following steps: according to a common fault index system of the power transformer, screening index parameters corresponding to different types of fault analysis of the power transformer from unbalanced monitoring data, and processing the index parameters by a proportional normalization method; randomly selecting two groups of data in the index parameters as initial clustering centers, and setting clustering analysis parameters; calculating Euclidean distances between various types of fault index parameters and an initial clustering center, dividing data in the unbalanced monitoring data into a lower approximate set or a boundary region of the class clusters according to the Euclidean distances, and calculating the unbalance degree between the class clusters; measuring the membership degree of the monitoring data by fusing the class cluster imbalance degrees; performing iterative computation on the cluster center according to the distribution condition of the cluster data; finally, performing state evaluation on the power transformer according to the clustering result; the method effectively improves the state evaluation precision of the power transformer.
Description
Technical Field
The invention belongs to the technical field of power transformer state evaluation, and particularly relates to a transformer state evaluation cluster analysis method based on data imbalance measurement.
Background
The power transformer is used as important power transmission and transformation equipment and is distributed at key nodes of the whole power transmission and transformation network, and safe and stable operation of the power transformer is important for guaranteeing power supply. When the failure occurs, not only can great economic loss be caused, but also social stability can be influenced. Therefore, by analyzing the monitoring data of the power transformer, the evaluation of the running state of the power transformer is of great significance for keeping the whole power system running safely and stably. The traditional state evaluation method can better diagnose the fault type of the transformer by analyzing fault data recorded by a fault recorder. However, with the wide application of the internet of things technology in the power system, mass monitoring devices generate mass monitoring data every day, the mass data include monitoring data in a normal operation state and monitoring data in an abnormal/fault state with extremely low density, and the monitoring data of the power transformer form a typical unevenly distributed data set. How to analyze the imbalance data monitoring data of the power transformer and evaluate the state of the power transformer becomes a hot point of study of recent researchers.
The clustering analysis is used as a classic unsupervised learning algorithm, data can be divided into a plurality of classes according to characteristics among the data, and the clustering analysis can be used for analyzing the unbalanced monitoring data of the power transformer. However, the traditional clustering algorithm equally divides the unbalanced monitoring data set of the power transformer, so that the monitoring data in the normal operation state is wrongly divided into fault data, and the state evaluation precision of the power transformer is influenced.
Disclosure of Invention
The invention provides a transformer state evaluation clustering analysis method based on data unbalance measurement aiming at the problem that the existing clustering algorithm has poor effect of processing an unbalance monitoring data set of a power transformer, the method screens index parameters corresponding to different types of fault analysis of the power transformer from unbalance monitoring data according to a common fault index system of the power transformer, and processes the index parameters by a proportional normalization method; finally, carrying out clustering analysis on the imbalance monitoring data by adopting a power transformer state evaluation method of imbalance measurement rough fuzzy k-means clustering, and obtaining a corresponding power transformer state evaluation result; the specific technical implementation scheme is as follows:
a transformer state evaluation cluster analysis method based on a data imbalance measurement is characterized by comprising the following steps:
screening index parameters corresponding to different types of fault analysis of the power transformer from unbalanced monitoring data according to a common fault index system of the power transformer, and processing the index parameters by a proportional normalization method;
randomly selecting two groups of data in the index parameters as initial clustering centers, and setting clustering analysis parameters of the unbalanced monitoring data according to historical data characteristics;
calculating Euclidean distances between the fault index data of each type and the initial clustering center, dividing data in the unbalance monitoring data into a lower approximate set or a boundary region of a class cluster according to the distances, and calculating the unbalance degree between the class clusters;
step four, the unbalance degrees are fused to measure the membership degrees of the unbalance monitoring data;
step five, performing iterative computation on the cluster center according to the clustering result of the data samples in the step three, counting samples of an approximation set and a boundary area under each cluster if the cluster center is not updated any more, and evaluating the state of the transformer; otherwise, returning to the third step.
Further, in the first step, the imbalance monitoring data is composed of the index parameters.
Further, the second step includes:
randomly selecting two groups of data in the index parameters, wherein one group of data is used as an initial clustering center of a normal-state cluster of the power transformer, and the other group of data is used as an initial clustering center of a fault cluster of the power transformer; and setting a distance judgment threshold and a fuzzy coefficient according to the historical data characteristics.
Further, the distance judgment threshold is used for comparing the relative sizes of the Euclidean distance between the sample and the normal cluster center and the Euclidean distance between the sample and the fault cluster center, and dividing the sample into an approximate set or a boundary region under the cluster according to the comparison result; the fuzzy coefficient is a constant coefficient 2 in the rough fuzzy K-means algorithm.
Further, the third step includes:
calculating a first Euclidean distance between the index parameters corresponding to different types of faults and the normal cluster center and a second Euclidean distance between the index parameters and the fault cluster center, judging the size of the first Euclidean distance and the second Euclidean distance, and comparing the ratio of a larger value to a smaller value;
comparing the ratio with the distance judgment threshold, and if the ratio is greater than the distance judgment threshold, dividing the index parameter into the lower approximate set of the class cluster corresponding to the smaller Euclidean distance of the first Euclidean distance and the second Euclidean distance; otherwise, dividing the image into the boundary area;
and respectively calculating the ratio of the number of the lower approximate set samples in the normal cluster and the fault cluster to the number of all the lower approximate set samples in the imbalance monitoring data to obtain the imbalance degree between the normal cluster and the fault cluster.
Further, in the fourth step:
when data in the unbalanced monitoring data belong to a lower approximate set in the normal cluster or the fault cluster, the membership value of the data is 1; when the sample data belongs to the boundary area, the membership degree of the sample data needs to be determined by a membership degree formula:
is depicted, wherein u ij Is a sample X j Membership to the ith class cluster; d ij Is a sample X j Euclidean distance to the cluster center; m is a blurring coefficient; k is the number of clusters of the cluster.
Further, the fifth step includes:
performing iterative computation on the cluster center according to the clustering result of the data samples in the step three; if the cluster center is not updated any more, counting samples of the lower approximation set and the boundary area in the corresponding cluster, and evaluating the state of the transformer: counting approximate set samples under the normal cluster, determining that the samples belong to normal data, marking the data with '1', and indicating that the power transformer corresponding to the data does not have the type fault; counting approximate set samples under the fault cluster, determining that the samples belong to fault samples, and marking the samples with 1 to indicate that the transformer has the fault; counting 'boundary region' samples, and marking '0' on the samples to indicate that the transformer is likely to have the type of fault in the future; and if the cluster center is continuously updated, returning to the step three.
Compared with the prior art, the invention has the beneficial effects that: the clustering effect of the unbalanced monitoring data set of the power transformer is good, the monitoring data are divided to be determined to belong to normal or fault clusters by introducing the concept of a rough set, and the fact that the monitoring data are determined to belong to normal or fault data is shown; the data of uncertain classification are divided into boundary regions, the monitoring data are represented as abnormal data, faults of the type can occur in the future, and the state evaluation precision of the power transformer is effectively improved.
Drawings
Fig. 1 is a block diagram illustrating a flow of a transformer state evaluation cluster analysis method based on a data imbalance metric according to an embodiment of the present invention.
Fig. 2 is an overall framework diagram illustrating the transformer state evaluation cluster analysis method based on the data imbalance metric according to the embodiment of the present invention.
FIG. 3 is a flow chart of a method for calculating membership according to an embodiment of the present invention.
FIG. 4 is a flow chart of rough fuzzy k-means clustering based on imbalance metric according to the embodiment of the present invention.
Fig. 5 is an effect diagram of performing state estimation clustering analysis on a transformer by using the method of the present invention in the embodiment of the present invention.
FIG. 6 is a diagram illustrating the effect of performing state estimation clustering analysis on a transformer by using a classical coarse fuzzy k-means algorithm according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention.
In the embodiment of the invention, a transformer state evaluation clustering analysis method based on data imbalance measurement is provided, the method realizes the state evaluation of the power transformer based on rough fuzzy k-means clustering of the imbalance measurement, referring to fig. 2, the method can carry out fault diagnosis on the power transformer with faults and can also carry out fault prediction on the power transformer without the faults by analyzing imbalance monitoring data of the power transformer; referring to fig. 1, it can be seen that the method of the present invention comprises the steps of:
step one, according to a common fault index system of the power transformer, selecting index parameters corresponding to different types of fault analysis of the power transformer from unbalanced monitoring data, and processing the index parameters by using a proportional normalization methodIndex parameters; the imbalance monitoring data in the invention is composed of index parameters, mainly data collected by power transformer monitoring equipment, such as gas H dissolved in oil 2 、CH 4 、C 2 H 4 、C 2 H 6 Etc., and voltage, current parameters of the transformer, etc.; preferably, the embodiment represents each group in the imbalance monitoring data as a 1 × 5 matrix, i.e. x ij ＝[x i1 ,x i2 ,x i3 ,x i4 ,x i5 ]1,2,3, 200, followed by the formula
Carrying out proportional normalization treatment; in other embodiments of the present invention, each group of imbalance monitoring data may also be represented as any type of matrix, and is not limited to a 1 × 5 matrix, depending on the number of the selected index systems.
Randomly selecting two groups of data in the index parameters as initial clustering centers, and setting clustering analysis parameters of the unbalanced monitoring data; the data of the initial clustering center is selected randomly, namely two groups of data in the unbalanced monitoring data are selected randomly, one group of data is used as the initial clustering center of the normal state cluster of the power transformer, and the other group of data is used as the initial clustering center of the fault cluster of the power transformer; setting a distance judgment threshold and a fuzzy coefficient according to the historical data characteristics of the actual record of the power transformer; specifically, a small amount of characteristic data is selected from state recording data of the power transformer for setting, wherein the characteristic data is recording data under the condition that the power transformer normally operates or fails to operate, so that the set distance judgment threshold value is ensured to be in accordance with the actual condition of the power transformer; the distance judgment threshold is used for comparing the relative sizes of the Euclidean distance between the sample and the normal cluster center and the Euclidean distance between the sample and the fault cluster center, and dividing the sample into an approximate set or a boundary region under the cluster according to the comparison result; the fuzzy coefficient is a constant coefficient 2 in the rough fuzzy K-means algorithm.
Preferably, the distance judgment threshold is assumed
In this embodiment, the distance determination threshold is set to
Of course, this is only a preferred embodiment of this implementation, and is not a limitation and a fixation on the value of the distance determination threshold in the method of the present invention, and may be set according to the characteristics of the historical index parameters of the power transformer.
Calculating Euclidean distances between each type of fault index data and an initial clustering center, dividing data in the unbalance monitoring data into a lower approximate set or a boundary area of the class clusters according to the distances, and calculating the unbalance degree between the class clusters: firstly, calculating a first Euclidean distance between index parameters corresponding to different types of faults and a normal cluster center and a second Euclidean distance between the index parameters and a fault cluster center, judging the size of the first Euclidean distance and the second Euclidean distance, and acquiring the ratio of a larger value to a smaller value of the first Euclidean distance and the second Euclidean distance; then, comparing the obtained ratio with a distance judgment threshold, and if the ratio is greater than the distance judgment threshold, dividing the index parameters into lower approximate sets of the corresponding class clusters of the smaller Euclidean distance in the first Euclidean distance and the second Euclidean distance; otherwise, dividing the image into boundary areas; finally, respectively calculating the ratio of the number of samples of the lower approximate set in the normal cluster and the fault cluster to the number of samples of all the lower approximate sets in the imbalance monitoring data to obtain the imbalance degree between the normal cluster and the fault cluster; specifically, the imbalance degree is calculated by the formula
Is calculated to obtain wherein
Indicates the number of samples of the approximation set falling on the ith class cluster, C z Representing data samples x j The cross cluster at present.
Step four, calculating the membership degree of the data in the imbalance monitoring data based on the imbalance degree, referring to fig. 3, if not, calculating the membership degree of the data in the imbalance monitoring dataWhen the data in the balanced monitoring data belongs to the lower approximate set in the normal cluster or the fault cluster, the membership value is 1; when the sample data belongs to the boundary area, the membership degree of the sample data needs to be determined by a membership degree formula:
is depicted, wherein u ij Is a sample X j Membership to the ith class cluster; d ij Is a sample X j Euclidean distance to the cluster center; m is a blurring coefficient; k is the number of clusters of the cluster.
Combining the above contents and with fig. 4, the whole clustering process in the method of the present invention includes: firstly, inputting unbalanced monitoring data to be clustered, and randomly selecting two data samples as initial cluster centers; then calculating the distance between each sample and the cluster center, judging a threshold value according to the set distance, and dividing each sample into a lower approximate set or a boundary area of the class cluster; counting the number of samples contained in the lower approximate set of each class cluster, and calculating the imbalance degree; calculating the membership degree of each sample by using the improved membership degree calculation formula; the cluster center is updated until the cluster center does not change any more.
Step five, performing iterative computation on the cluster center according to the clustering result of the data samples in the step three, counting samples of an approximation set and a boundary area under each cluster if the cluster center is not updated any more, and evaluating the state of the transformer; otherwise, returning to the third step: counting approximate set samples under the normal cluster, determining that the samples belong to normal data, marking the data with '1', and indicating that the power transformer corresponding to the data does not have the type fault; counting approximate set samples under the fault cluster, determining that the samples belong to fault samples, and marking the samples with 1 to indicate that the transformer has the fault; counting 'boundary region' samples, and marking '0' on the samples to indicate that the transformer is likely to have the type of fault in the future; and if the cluster center is continuously updated, returning to the step three.
Referring to fig. 5 and 6, the effectiveness of the method of the present invention is further illustrated by performing a test analysis on the method of the present invention by taking 4 sets of normal data and 16 sets of fault data as an example; specifically, the group of test data is processed by a rough fuzzy k-means algorithm and a power transformer state evaluation method based on imbalance measurement rough fuzzy k-means clustering, wherein in the figure, "+" and "o" represent samples which are clustered correctly; "," indicates that samples originally belonging to the majority class cluster are wrongly divided into the minority class cluster; black "□" indicates samples divided into boundary regions; according to the evaluation results of 20 groups of test data by two algorithms, the power transformer state evaluation method based on the unbalanced measurement rough fuzzy k-means clustering only evaluates two groups of data in error, and the rough fuzzy k-means algorithm evaluates six groups of data in error, namely, the method has better accuracy on state evaluation and analysis of the power transformer, and can improve more accurate prediction on the state of the power transformer; therefore, in actual operation, according to the state evaluation cluster analysis of the power transformer, measures can be made in advance and the corresponding maintenance can be carried out on the power transformer, so that the use safety of the power transformer can be improved.
Compared with the prior art, the invention has the beneficial effects that: the clustering effect of the unbalanced monitoring data set of the power transformer is good, the monitoring data are divided to be determined to belong to normal or fault clusters by introducing the concept of a rough set, and the fact that the monitoring data are determined to belong to normal or fault data is shown; the data of uncertain classification are divided into boundary regions, the fact that the monitoring data belong to abnormal data is shown, faults of the type can happen in the future, and the precision of state evaluation of the power transformer is effectively improved.
Claims (5)
1. A transformer state evaluation cluster analysis method based on a data imbalance measurement is characterized by comprising the following steps:
screening index parameters corresponding to different types of fault analysis of the power transformer from unbalanced monitoring data according to a common fault index system of the power transformer, and processing the index parameters by a proportional normalization method;
randomly selecting two groups of data in the index parameters as initial clustering centers, and setting clustering analysis parameters of the unbalanced monitoring data according to historical data characteristics;
calculating Euclidean distances between each type of fault index data and the initial clustering center, dividing data in the unbalance monitoring data into a lower approximate set or a boundary region of the class clusters according to the distances, and calculating the unbalance degree between the class clusters;
the third step comprises the following steps:
calculating a first Euclidean distance between the index parameters corresponding to different types of faults and a normal cluster center and a second Euclidean distance between the index parameters and the fault cluster center, judging the size of the first Euclidean distance and the second Euclidean distance, and comparing the ratio of a larger value to a smaller value;
comparing the ratio with the distance judgment threshold, and if the ratio is greater than the distance judgment threshold, dividing the index parameter into the lower approximate set of the class cluster corresponding to the smaller Euclidean distance of the first Euclidean distance and the second Euclidean distance; otherwise, dividing the image into the boundary area;
respectively calculating the ratio of the number of the lower approximate set samples in the normal cluster and the fault cluster to the number of all the lower approximate set samples in the imbalance monitoring data to obtain the imbalance degree between the normal cluster and the fault cluster;
step four, fusing the unbalance degree to calculate the membership degree of the unbalance monitoring data;
in the fourth step:
when data in the unbalanced monitoring data belong to a lower approximate set in the normal cluster or the fault cluster, the membership value of the data is 1; when the sample data belongs to the boundary area, the membership degree of the sample data needs to be determined by a membership degree formula:
is depicted, wherein u ij Is a sample X j Membership to the ith class cluster; d ij Is a sample X j Euclidean distance to the cluster center; m is a blurring coefficient; k is the number of clusters of the cluster；
Step five, performing iterative computation on the cluster center according to the clustering result of the data samples in the step three, counting samples of an approximation set and a boundary area under each cluster if the cluster center is not updated any more, and evaluating the state of the transformer; otherwise, returning to the third step.
2. The transformer state estimation cluster analysis method based on the data imbalance metric of claim 1, wherein in the first step, the imbalance monitoring data is composed of the index parameters.
3. The transformer state evaluation cluster analysis method based on the data imbalance metric of claim 1, wherein the second step comprises:
randomly selecting two groups of data in the index parameters, wherein one group of data is used as an initial clustering center of a normal-state cluster of the power transformer, and the other group of data is used as an initial clustering center of a fault cluster of the power transformer; and setting a distance judgment threshold and a fuzzy coefficient according to the historical data characteristics.
4. The transformer state evaluation cluster analysis method based on the data imbalance metric as claimed in claim 2, wherein the distance judgment threshold is used for comparing the relative sizes of the Euclidean distance between the sample and the normal cluster center and the Euclidean distance between the sample and the fault cluster center, and dividing the sample into an approximate set or a boundary region under the cluster according to the comparison result; the fuzzy coefficient is a constant coefficient 2 in the rough fuzzy K-means algorithm.
5. The transformer state evaluation cluster analysis method based on the data imbalance metric of claim 1, wherein the fifth step comprises:
performing iterative computation on the cluster center according to the clustering result of the data samples in the step three; if the cluster center is not updated any more, counting samples of the lower approximation set and the boundary area in the corresponding cluster, and evaluating the state of the transformer: counting approximate set samples under the normal cluster, determining that the samples belong to normal data, marking the data with '1', and indicating that the power transformer corresponding to the data does not have the type fault; counting approximate set samples under the fault cluster, determining that the samples belong to fault samples, and marking the samples with 1 to indicate that the transformer has the fault; counting 'boundary region' samples, and marking '0' on the samples to indicate that the transformer is likely to have the type of fault in the future; and if the cluster center is continuously updated, returning to the step three.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910093168.4A CN109816031B (en) | 2019-01-30 | 2019-01-30 | Transformer state evaluation clustering analysis method based on data imbalance measurement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910093168.4A CN109816031B (en) | 2019-01-30 | 2019-01-30 | Transformer state evaluation clustering analysis method based on data imbalance measurement |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109816031A CN109816031A (en) | 2019-05-28 |
| CN109816031B true CN109816031B (en) | 2022-08-05 |
Family
ID=66606017
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910093168.4A Active CN109816031B (en) | 2019-01-30 | 2019-01-30 | Transformer state evaluation clustering analysis method based on data imbalance measurement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109816031B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110133410A (en) * | 2019-05-31 | 2019-08-16 | 国网河北省电力有限公司石家庄供电分公司 | Diagnosis Method of Transformer Faults and system based on Fuzzy C-Means Cluster Algorithm |
| CN110333414A (en) * | 2019-08-02 | 2019-10-15 | 华北电力大学（保定） | The multi-level state evaluating method of power transformer |
| CN111598148B (en) * | 2020-04-29 | 2022-09-16 | 中国电子科技集团公司第二十八研究所 | Capacity evaluation method and device based on historical capacity similarity characteristics |
| CN112560900B (en) * | 2020-09-08 | 2023-01-20 | 同济大学 | Multi-disease classifier design method for sample imbalance |
| CN112599252A (en) * | 2021-03-04 | 2021-04-02 | 北京冠新医卫软件科技有限公司 | Chronic medical record feature classification processing method and device and terminal |
| CN113344347B (en) * | 2021-05-25 | 2023-10-20 | 国网湖南省电力有限公司 | Low-voltage power supply line health state assessment method and system |
| CN113537528B (en) * | 2021-07-28 | 2022-05-03 | 贵州电网有限责任公司 | Preprocessing method and system for state monitoring data of power transmission and transformation equipment |
| CN113869438B (en) * | 2021-09-30 | 2023-04-07 | 自然资源部第三海洋研究所 | Key defense area dividing method for storm surge disaster based on k-means |
| CN116153437B (en) * | 2023-04-19 | 2023-06-30 | 乐百氏(广东)饮用水有限公司 | Water quality safety evaluation and water quality prediction method and system for drinking water source |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107656154A (en) * | 2017-09-18 | 2018-02-02 | 杭州安脉盛智能技术有限公司 | Based on the Diagnosis Method of Transformer Faults for improving Fuzzy C-Means Cluster Algorithm |
| CN109063734A (en) * | 2018-06-29 | 2018-12-21 | 广东工业大学 | The oil-immersed transformer malfunction appraisal procedure clustered in conjunction with multistage local density |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101711028B1 (en) * | 2012-05-04 | 2017-03-13 | 한국전자통신연구원 | Apparatus and method for vihicle outlier monitoring using clustering |
-
2019
- 2019-01-30 CN CN201910093168.4A patent/CN109816031B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107656154A (en) * | 2017-09-18 | 2018-02-02 | 杭州安脉盛智能技术有限公司 | Based on the Diagnosis Method of Transformer Faults for improving Fuzzy C-Means Cluster Algorithm |
| CN109063734A (en) * | 2018-06-29 | 2018-12-21 | 广东工业大学 | The oil-immersed transformer malfunction appraisal procedure clustered in conjunction with multistage local density |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109816031A (en) | 2019-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109816031B (en) | Transformer state evaluation clustering analysis method based on data imbalance measurement | |
| CN110070060B (en) | Fault diagnosis method for bearing equipment | |
| CN109389145B (en) | Electric energy meter manufacturer evaluation method based on metering big data clustering model | |
| CN112911627B (en) | Wireless network performance detection method, device and storage medium | |
| CN108053148B (en) | Efficient fault diagnosis method for power information system | |
| CN113780443B (en) | Threat detection-oriented network security situation assessment method | |
| CN109389325B (en) | Method for evaluating state of electronic transformer of transformer substation based on wavelet neural network | |
| CN115982602B (en) | Photovoltaic transformer electrical fault detection method | |
| CN114004137A (en) | Multi-source meteorological data fusion and pretreatment method | |
| CN115130578A (en) | Incremental rough clustering-based online evaluation method for state of power distribution equipment | |
| CN110674858A (en) | Traffic jam detection method based on space-time association and big data mining | |
| CN102945222A (en) | Poor information measurement data gross error discrimination method based on Grey System Theory | |
| CN110544047A (en) | Bad data identification method | |
| CN111709668A (en) | Power grid equipment parameter risk identification method and device based on data mining technology | |
| CN114266289A (en) | Complex equipment health state assessment method | |
| Vermetten et al. | Is there anisotropy in structural bias? | |
| CN113837591A (en) | Equipment health assessment method oriented to multi-working-condition operation conditions | |
| CN112949735A (en) | Liquid hazardous chemical substance volatile concentration abnormity discovery method based on outlier data mining | |
| CN111767273B (en) | Data intelligent detection method and device based on improved SOM algorithm | |
| CN109784777B (en) | Power grid equipment state evaluation method based on time sequence information fragment cloud similarity measurement | |
| CN114597886A (en) | Power distribution network operation state evaluation method based on interval type two fuzzy clustering analysis | |
| CN111507374A (en) | Power grid mass data anomaly detection method based on random matrix theory | |
| CN112765219B (en) | Stream data abnormity detection method for skipping steady region | |
| CN114547796B (en) | Ball mill feature fusion fault diagnosis method based on optimized BN network | |
| CN116307844A (en) | Low-voltage transformer area line loss evaluation analysis method |
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 |