CN115905990A - Transformer oil temperature abnormity monitoring method based on density aggregation algorithm - Google Patents
Transformer oil temperature abnormity monitoring method based on density aggregation algorithm Download PDFInfo
- Publication number
- CN115905990A CN115905990A CN202211454785.0A CN202211454785A CN115905990A CN 115905990 A CN115905990 A CN 115905990A CN 202211454785 A CN202211454785 A CN 202211454785A CN 115905990 A CN115905990 A CN 115905990A
- Authority
- CN
- China
- Prior art keywords
- oil temperature
- transformer
- cluster
- transformer oil
- 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
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Protection Of Transformers (AREA)
- Housings And Mounting Of Transformers (AREA)
Abstract
The invention provides a transformer oil temperature abnormity monitoring method based on a density aggregation algorithm. And calling density aggregation based on the dynamic time normalization distance to perform clustering processing on the normalized feature matrix, and performing anomaly detection on the oil temperature data of the transformer by using a density aggregation algorithm.
Description
Technical Field
The invention relates to the field of electrical technology detection, in particular to a transformer oil temperature abnormity monitoring method based on a density aggregation algorithm.
Background
The transformer is used as a key device of primary power, the related technology is complex, and the operation condition of the transformer has a large influence on the safety of a power grid. The transformer oil temperature determines the transformer load capacity and the internal insulation aging speed thereof, thereby affecting the service life of the transformer.
The traditional method for identifying the abnormal oil temperature of the transformer mainly comprises methods for predicting the top oil temperature of the oil immersed transformer, diagnosing the oil immersed transformer based on oil chromatography and the like, but the accuracy and the timeliness are difficult to meet at the same time.
Disclosure of Invention
The invention aims to provide a transformer oil temperature abnormity monitoring method based on a density clustering algorithm, and provides a data analysis method for transformer cooling device working condition abnormity analysis.
The invention is realized by the following technical scheme:
a transformer oil temperature abnormity monitoring method based on a density clustering algorithm comprises the following steps:
s1: acquiring data of a transformer to generate a first data set, and cleaning the data in the first data set according to the working characteristics of a transformer cooling device to generate a second data set; extracting characteristic parameters of the transformer from the second data set, and generating a characteristic matrix according to the characteristic parameters according to a time sequence; wherein
The characteristic parameters include: manufacturer, equipment model, ambient temperature and transformer load rate;
s2: normalizing the characteristic matrix, and performing density aggregation input on parameters required by given density aggregation;
s3: calling density aggregation to perform clustering processing on the normalized feature matrix, and outputting a clustering result;
s4: and marking and eliminating the noise sample according to the result of the step S3, and outputting oil temperature data.
As an alternative mode of the present invention, in the step S1, the generating the feature matrix includes the steps of:
determining a time dimension direction, continuously collecting the transformer for N times based on the time dimension direction to obtain N first data sets, cleaning data of each first data set to generate N second data sets, extracting characteristic parameters of each second data set, preprocessing the extracted characteristic parameters, and generating a characteristic matrix; wherein N is a positive integer.
As an optional mode of the present invention, the preprocessing of the extracted feature parameters includes the following steps:
calibrating a characteristic reference set, and performing recursive circulation by using the characteristic reference set and characteristic parameters adjacent to the characteristic reference set in the time dimension direction;
traversing all the characteristic parameters, extracting all the difference degree information, and screening effective characteristic parameters meeting the preset difference degree requirement;
and merging the effective characteristic parameters, and generating a characteristic matrix according to a time sequence.
As an alternative of the present invention, in the step S2, the parameters required for density clustering include neighborhood and core objects, wherein
The neighborhood comprises a transformer oil temperature sample set formed by elements in a normalized feature matrix and any element x in the feature matrix j Samples whose distance is not greater than a threshold value of the selected metric distance;
the core object is determined in the following manner: if x j At least MinPts samples are included in the neighborhood, then x j Is a core object.
As an alternative of the present invention, in the above step S2, the determining the density aggregation input parameter includes the sub-steps of:
s21: packing characteristic parameters of the transformer, including a manufacturer, an equipment model, an environment temperature and a transformer load rate, to generate an original data set;
s22: calculating Euclidean distances between all points in the original data set to generate a one-dimensional distance data set; clustering the one-dimensional distance data set, marking a cluster in a clustering result, and calculating the proportion of the clustered result in the one-dimensional distance data set;
s23: calculating a weighted average value of the sample set on the clustered class, and taking the weighted average value as a value Eps of the density aggregation neighborhood;
s24: and calculating the point number of the mark cluster in the Eps neighborhood, and outputting the value of the input parameter MinPts.
As an optional manner of the present invention, in step S22, a K-Means algorithm is used for clustering, the best K one-dimensional distance classes are output, the ith class cluster is used as a labeled class cluster, the distance data in each labeled class cluster is respectively calculated, the mean value of the distance data is recorded, and the specific gravity of the mean value of each distance data in the one-dimensional distance data set is calculated.
As an optional manner of the present invention, in step S24, after the number of points of the mark class cluster in the vicinity of the Eps is calculated, the minimum number of points is set as the value of the mark class cluster core sample MinPts; and selecting the MinPts with the largest value from the core samples MinPts of all the mark class clusters as the value of the input parameter core sample MinPts.
As an alternative mode of the present invention, in step S3, the clustering process on the feature matrix includes the sub-steps of:
s31: oil temperature sample set D = (x) of input transformer cooling device 1 ,x 2 ,Λ,x m ) Neighborhood, input parameters;
s32: initializing a transformer oil temperature core object set A, the number k of oil temperature cluster, an unaccessed transformer oil temperature sample set B and oil temperature cluster division C, and enabling
k＝0，B＝D，/>
S33: finding out all core objects in a transformer oil temperature sample set D, and setting j =1,2, \8230;, m; through a distance measurement mode, searching out an oil temperature data flow sample x j Neighborhood oil temperature data subsample set N ε (x j ) If | N is satisfied ε (x j ) If | ≧ MinPts, then x is adjusted j The core object of the transformer oil temperature is counted into A, and the A = A ^ U { x ^ x j }；
S34: randomly selecting any oil temperature object a of a core from a transformer oil temperature core object set A; and initializing a current cluster core transformer oil temperature object queue Acur = { a }, a class oil temperature cluster number k = k +1 and a current transformer oil temperature cluster sample set C K = a; updating the sample set of unaccessed transformer oil temperatures B = B- { a };
s35: checking the current oil temperature cluster core object queue of the transformer cooling device if
And if the current cluster Ck is generated completely, updating the cluster division C = { C = { (C) } 1 ,C 2 ,Λ,C K And updating an oil temperature core object set A = A-C K ；
S36: an oil temperature core object a' is taken out from a current oil temperature cluster core object queue Acur of the transformer, and a neighborhood sub-sample set N is found out through a neighborhood distance threshold value ε (a'), let Δ = N ε (a'). Quadrature.B, updating the current transformer oil temperature cluster sample set C K ＝C K U delta, updating an inaccessible transformer oil temperature sample set B = B-delta, updating Acur = Acur @ U (delta ≈ A) -a'; and continuing to check according to the step of the step S35;
s37: output transformer oil temperature cluster C = { C = { (C) } 1 ,C 2 ,Λ,C K And abnormal oil temperature data clusters D-C as noise samples.
In addition, to achieve the above object, the present embodiment further provides an electronic device, which includes a memory and a processor, where the memory stores a computer program executable on the processor, and the processor implements the steps of the transformer oil temperature abnormality monitoring method based on the density aggregation algorithm when executing the computer program.
In addition, to achieve the above object, the present embodiment further provides a storage medium having a computer program stored thereon, where the computer program is executed by a processor to implement the steps in the above-mentioned transformer oil temperature abnormality monitoring method based on the density clustering algorithm.
Compared with the prior art, the invention has the following advantages and beneficial effects:
the invention can find clusters with any shapes by selecting a density clustering algorithm, does not need to determine the number of the clustered clusters in advance, and is insensitive to noise points in data. Therefore, the distance between normal data in the oil temperature data of the transformer cooling device in the field of abnormality detection is close, the distance between abnormal data and normal data is far, the density clustering-based abnormality detection algorithm can perform clustering according to the distribution condition of the density of a data set, and the abnormal transformer can be detected more efficiently and quickly.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention. In the drawings:
fig. 1 is a schematic flow chart of a transformer oil temperature anomaly monitoring method based on a density aggregation algorithm according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to examples and accompanying drawings, and the exemplary embodiments and descriptions thereof are only used for explaining the present invention and are not meant to limit the present invention.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
Thus, the following detailed description of the embodiments of the present invention, provided in the embodiments, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
Examples
Referring to fig. 1, the present embodiment provides a method for monitoring an abnormal oil temperature of a transformer based on a density clustering algorithm, including:
s1: acquiring data of a transformer to generate a first data set, and cleaning the data in the first data set according to the working characteristics of a transformer cooling device to generate a second data set; extracting characteristic parameters of the transformer from the second data set, and generating a characteristic matrix according to the characteristic parameters according to a time sequence; wherein
The characteristic parameters include: manufacturer, equipment model, ambient temperature and transformer load factor. The first data set comprises all parameters and performances of the transformer, but not all parameters or performances are required in the actual production environment of the embodiment, so that the parameters or performances are cleaned, unnecessary features are eliminated, a second data set with a smaller range is obtained, and a manufacturer, an equipment model, an environment temperature and a transformer load rate are extracted from the second data set according to the requirements of a density clustering algorithm and actual application consideration.
Wherein, step S1 further comprises: determining a time dimension direction, continuously collecting the transformer for N times based on the time dimension direction to obtain N first data sets, cleaning data of each first data set to generate N second data sets, extracting characteristic parameters of each second data set, preprocessing the extracted characteristic parameters, and generating a characteristic matrix; wherein N is a positive integer. The temperature information at different time points is collected along with the lapse of time, and a plurality of points with different characteristic parameters of the time change temperature are generated according to the time lapse sequence, so that the effect of real-time detection is achieved.
After the characteristic parameters are obtained, whether errors, false detection or missing detection exist in the characteristic parameters needs to be judged. In this embodiment, the preprocessing of the feature parameters includes: calibrating a characteristic reference set, and performing recursive circulation by using the characteristic reference set and characteristic parameters adjacent to the characteristic reference set in the time dimension direction; traversing all the characteristic parameters, extracting all the difference degree information, and screening effective characteristic parameters meeting the preset difference degree requirement; and merging the effective characteristic parameters, and generating a characteristic matrix according to a time sequence. The reference set can be designed in a user-defined mode at will, and characteristic parameter information of conditions of actual conditions is met in a certain interval. And comparing the data on the left side and the right side of the time axis according to the data to judge whether obvious errors exist. And identifying the difference degree information as an effective parameter in a reasonable and acceptable range, and participating in subsequent matrix synthesis.
S2: and carrying out normalization processing on the characteristic matrix, and inputting density aggregation by giving parameters required by density aggregation. In this embodiment, a density clustering algorithm is adopted, and the density clustering parameters include two definitions:
ε -neighborhood: for element x in normalized feature matrix j And the epsilon-neighborhood of the transformer oil temperature sample set D is composed of elements in the normalized feature matrix and x j Samples for which the distance of (a) is not greater than the threshold value epsilon for the chosen metric distance, i.e.:
N ε (x j )＝{x i ∈D|dist(x i ,x j )≤ε}
core pairLike: if the epsilon-neighborhood of xj contains at least MinPts samples, i.e. | | N ε (x j ) | MinPts, then x j Is a core object.
Then, in order to determine the specific values of the two parameters, determining the density aggregation input parameter comprises the sub-steps of:
s21: packing the characteristic parameters of the transformer, including a manufacturer, an equipment model, an environment temperature and a transformer load rate, to generate an original data set;
s22: calculating Euclidean distances between all points in the original data set to generate a one-dimensional distance data set; clustering the one-dimensional distance data set, marking cluster in the clustering result, and calculating the proportion of the clustered result in the one-dimensional distance data set. In the step, clustering is carried out by adopting a K-Means algorithm, and the optimal K one-dimensional distance classes { G ] are output 1 ,G 2 ,Λ,G K And taking the ith cluster as a mark cluster, respectively calculating the distance data in each mark cluster, recording the mean value of the distance data, and calculating the specific gravity p of the mean value of each distance data in the one-dimensional distance data set i I.e. p i ＝|G i |。
S23: calculating to obtain a data set D in { G 1 ,G 2 ,Λ,G j And (4) weighted average value on the obtained value and taking the weighted average value as a value of the neighborhood of the density clustering algorithm parameter Eps. I.e. Eps = p 1 G 1 +p 2 G 2 +Λ+p j G j 。
S24: and calculating the point number of the mark cluster in the Eps neighborhood, and outputting the value of the input parameter MinPts. After the point number of the mark cluster in the Eps neighborhood is calculated, the minimum point number is set as the value of a mark cluster core sample MinPts; the most significant value of the MinPts among the core samples MinPts i (i =1,2.. Times., j) of all the labeled clusters is selected as the value of the input parameter core sample MinPts. The largest MinPtsi is selected as the global MinPts because, if MinPtsi is too small, noise points between adjacent clusters may also meet the density reachable condition, thereby merging two clusters into one cluster.
S3: and calling density aggregation to perform clustering processing on the feature matrix after normalization processing, and outputting a clustering result. The method comprises the following substeps:
s31: input transformer cooling device oil temperature sample set D = (x) 1 ,x 2 ,Λ,x m ) Neighborhood, input parameters;
s32: initializing a transformer oil temperature core object set A, the number k of oil temperature cluster groups, an unaccessed transformer oil temperature sample set B and oil temperature cluster division C, and enabling
k＝0，B＝D，/>
S33: finding out all core objects in a transformer oil temperature sample set D, and setting j =1,2, \8230;, m; through a distance measurement mode, searching out an oil temperature data flow sample x j Neighborhood oil temperature data subsample set N ε (x j ) If | N is satisfied ε (x j ) If | ≧ MinPts, then x j The core object of the transformer oil temperature is counted into A, and the A = A ^ U { x ^ x j }；
S34: randomly selecting any oil temperature object a of a core from a transformer oil temperature core object set A; and initializing a current cluster core transformer oil temperature object queue Acur = { a }, a class oil temperature cluster number k = k +1 and a current transformer oil temperature cluster sample set C K = a; updating the sample set of unaccessed transformer oil temperatures B = B- { a };
s35: checking the current oil temperature cluster core object queue of the transformer cooling device if
And if the current cluster Ck is generated completely, updating the cluster division C = { C = { (C) } 1 ,C 2 ,Λ,C K And updating an oil temperature core object set A = A-C K ；
S36: an oil temperature core object a' is taken out from a current oil temperature cluster core object queue Acur of the transformer, and a neighborhood sub-sample set N is found out through a neighborhood distance threshold value ε (a'), let Δ = N ε (a'). Quadrature.B, furtherNew current transformer oil temperature cluster sample set C K ＝C K U & ltdelta & gt, updating a sample set B = B-delta of the inaccessible transformer oil temperature, and updating Acur = Acur { (delta ≧ U & ltlambda) — a'; and continuing to check according to the step of the step S35;
s37: output transformer oil temperature cluster C = { C = { (C) } 1 ,C 2 ,Λ,C K And the abnormal oil temperature data cluster D-C are used as noise samples.
Before the step S34, determining whether the oil temperature core object set is empty, and if so, determining whether the oil temperature core object set is empty
The algorithm is ended, otherwise, S34 is executed; in step 35, after the operation is detected and updated, the operation returns to continuously determine whether the oil temperature core object set is empty.
S4: and marking and eliminating the noise sample according to the result of the step S3, and outputting oil temperature data. The noise sample is the transformer with abnormal oil temperature.
In this embodiment, the density clustering algorithm is able to find clusters of arbitrary shapes. And the number of clustering clusters does not need to be determined in advance, and the method is insensitive to noise points in the data. Therefore, the distance between normal data in the oil temperature data of the transformer cooling device used in the abnormal detection field is close, the distance between the abnormal data and the normal data is far, and the abnormal detection algorithm based on density clustering can perform clustering according to the distribution condition of the density of the data set.
In addition, to achieve the above object, the present embodiment further provides an electronic device, which includes a memory and a processor, where the memory stores a computer program executable on the processor, and the processor executes the computer program to implement the steps of the transformer oil temperature abnormality monitoring method based on the density aggregation algorithm.
In addition, to achieve the above object, the present embodiment further provides a storage medium having a computer program stored thereon, where the computer program is executed by a processor to implement the steps in the above-mentioned transformer oil temperature abnormality monitoring method based on the density clustering algorithm.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are merely exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (10)
1. A transformer oil temperature abnormity monitoring method based on a density clustering algorithm is characterized by comprising the following steps:
s1: acquiring data of a transformer to generate a first data set, and cleaning the data in the first data set according to the working characteristics of a transformer cooling device to generate a second data set; extracting characteristic parameters of the transformer from the second data set, and generating a characteristic matrix according to the characteristic parameters according to a time sequence; wherein
The characteristic parameters comprise: manufacturer, equipment model, ambient temperature and transformer load factor;
s2: normalizing the characteristic matrix, and performing density aggregation input on parameters required by given density aggregation;
s3: calling density aggregation to perform clustering processing on the normalized feature matrix, and outputting a clustering result;
s4: and marking and eliminating the noise sample according to the result of the step S3, and outputting oil temperature data.
2. The transformer oil temperature abnormality monitoring method based on the density clustering algorithm as claimed in claim 1, wherein in the step S1, the generating the characteristic matrix comprises the following steps:
determining a time dimension direction, continuously collecting the transformer for N times based on the time dimension direction to obtain N first data sets, cleaning data of each first data set to generate N second data sets, extracting characteristic parameters of each second data set, preprocessing the extracted characteristic parameters, and generating a characteristic matrix; wherein N is a positive integer.
3. The transformer oil temperature abnormity monitoring method based on the density clustering algorithm as claimed in claim 2, wherein the preprocessing of the extracted characteristic parameters comprises the following steps:
calibrating a characteristic reference set, and performing recursive circulation by using the characteristic reference set and the characteristic parameters adjacent to the characteristic reference set in the time dimension direction;
traversing all the characteristic parameters, extracting all the difference degree information, and screening effective characteristic parameters meeting the requirement of preset difference degree;
and merging the effective characteristic parameters, and generating a characteristic matrix according to a time sequence.
4. The method for monitoring abnormal oil temperature of transformer based on density clustering algorithm as claimed in claim 1, wherein in step S2, the parameters required for density clustering include neighborhood and core objects, wherein
The neighborhood comprises a transformer oil temperature sample set formed by elements in the normalized feature matrix and any element x in the feature matrix j Samples whose distance is not greater than a threshold value of the selected metric distance;
the core object is determined in the following manner: if x j At least MinPts samples are included in the neighborhood, then x j Is a core object.
5. The transformer oil temperature abnormality monitoring method based on the density clustering algorithm as claimed in claim 4, wherein in the step S2, the determining the density clustering input parameters comprises the sub-steps of:
s21: packing the characteristic parameters of the transformer, including a manufacturer, an equipment model, an environment temperature and a transformer load rate, to generate an original data set;
s22: calculating Euclidean distances between all points in the original data set to generate a one-dimensional distance data set; clustering the one-dimensional distance data set, marking a cluster in a clustering result, and calculating the proportion of the clustered result in the one-dimensional distance data set;
s23: calculating a weighted average value of the sample set on the clustered class, and taking the weighted average value as a value Eps of a density aggregation neighborhood;
s24: and calculating the point number of the mark cluster in the Eps neighborhood, and outputting the value of the input parameter MinPts.
6. The method for monitoring transformer oil temperature abnormality based on density clustering algorithm according to claim 5, wherein in the step S22, a K-Means algorithm is adopted for clustering, the best K one-dimensional distance classes are output, the ith class cluster is taken as the mark class cluster, the distance data in each mark class cluster are respectively calculated, the mean value of the distance data is recorded, and the specific gravity of the mean value of each distance data in the one-dimensional distance data set is calculated.
7. The method for monitoring transformer oil temperature abnormality based on density clustering algorithm according to claim 6, wherein in the step S24, after the number of points of the mark cluster in the neighborhood of Eps is calculated, the minimum number of points is set as the value of the mark cluster core sample MinPts; and selecting the MinPts with the largest value from the core samples MinPts of all the mark class clusters as the value of the input parameter core sample MinPts.
8. The transformer oil temperature anomaly monitoring method based on the density clustering algorithm as claimed in claim 6, wherein in the step S3, the clustering process of the feature matrix comprises the sub-steps of:
s31: input transformer cooling device oil temperature sample set D = (x) 1 ,x 2 ,Λ,x m ) Neighborhood, input parameters;
s32: initializing a transformer oil temperature core object set A, the number k of oil temperature cluster groups, an unaccessed transformer oil temperature sample set B and oil temperature cluster division C, and enabling
k＝0，B＝D，/>
S33: finding out all core objects in a transformer oil temperature sample set D, and setting j =1,2, \8230;, m; through a distance measurement mode, searching out an oil temperature data flow sample x j Neighborhood oil temperature data subsample set N ε (x j ) If | N is satisfied ε (x j ) If | ≧ MinPts, then x is adjusted j The core object of the transformer oil temperature is counted into A, and A = A { [ x ] is set as j }；
S34: randomly selecting any oil temperature object a of a core from a transformer oil temperature core object set A; and initializing a current cluster core transformer oil temperature object queue Acur = { a }, a serial number k = k +1 of cluster number of category oil temperature clusters and a current transformer oil temperature cluster sample set C K = a; updating the sample set of unaccessed transformer oil temperatures B = B- { a };
s35: checking the current oil temperature cluster core object queue of the transformer cooling device if
And if the current cluster Ck is generated completely, updating the cluster division C = { C = { (C) } 1 ,C 2 ,Λ,C K And updating an oil temperature core object set A = A-C K ；
S36: an oil temperature core object a' is taken out from a current transformer oil temperature cluster core object queue Acur, and a neighborhood sub sample set N is found out through a neighborhood distance threshold value ε (a'), let Δ = N ε (a'). Quadrature.B, updating the current transformer oil temperature cluster sample set C K ＝C K U delta, updating an inaccessible transformer oil temperature sample set B = B-delta, updating Acur = Acur @ U (delta ≈ A) -a'; and continuing to check according to the step of the step S35;
s37: output transformer oil temperature cluster C = { C = { (C) } 1 ,C 2 ,Λ,C K And the abnormal oil temperature data cluster D-C are used as noise samples.
9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program operable on the processor, and wherein the processor implements the steps of the density aggregation algorithm based transformer oil temperature anomaly monitoring method according to any one of claims 1 to 8 when executing the computer program.
10. A storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the transformer oil temperature anomaly monitoring method based on a density aggregation algorithm according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211454785.0A CN115905990A (en) | 2022-11-21 | 2022-11-21 | Transformer oil temperature abnormity monitoring method based on density aggregation algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211454785.0A CN115905990A (en) | 2022-11-21 | 2022-11-21 | Transformer oil temperature abnormity monitoring method based on density aggregation algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115905990A true CN115905990A (en) | 2023-04-04 |
Family
ID=86494404
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211454785.0A Pending CN115905990A (en) | 2022-11-21 | 2022-11-21 | Transformer oil temperature abnormity monitoring method based on density aggregation algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115905990A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116502170A (en) * | 2023-06-29 | 2023-07-28 | 山东浩坤润土水利设备有限公司济宁经济开发区分公司 | Agricultural water conservancy monitoring method and related device based on cloud platform |
| CN116702473A (en) * | 2023-06-08 | 2023-09-05 | 江苏国电南自海吉科技有限公司 | Clustering algorithm-based transformer temperature abnormality early warning method and system |
-
2022
- 2022-11-21 CN CN202211454785.0A patent/CN115905990A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116702473A (en) * | 2023-06-08 | 2023-09-05 | 江苏国电南自海吉科技有限公司 | Clustering algorithm-based transformer temperature abnormality early warning method and system |
| CN116702473B (en) * | 2023-06-08 | 2024-08-27 | 江苏国电南自海吉科技有限公司 | Clustering algorithm-based transformer temperature abnormality early warning method and system |
| CN116502170A (en) * | 2023-06-29 | 2023-07-28 | 山东浩坤润土水利设备有限公司济宁经济开发区分公司 | Agricultural water conservancy monitoring method and related device based on cloud platform |
| CN116502170B (en) * | 2023-06-29 | 2023-09-22 | 山东浩坤润土水利设备有限公司济宁经济开发区分公司 | Agricultural water conservancy monitoring method and related device based on cloud platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chien et al. | Analysing semiconductor manufacturing big data for root cause detection of excursion for yield enhancement | |
| CN115905990A (en) | Transformer oil temperature abnormity monitoring method based on density aggregation algorithm | |
| CN112637132B (en) | Network anomaly detection method and device, electronic equipment and storage medium | |
| JP2020500420A (en) | Yield prediction system and method for machine learning based semiconductor manufacturing | |
| CN109816031B (en) | Transformer state evaluation clustering analysis method based on data imbalance measurement | |
| CN111898647B (en) | Clustering analysis-based low-voltage distribution equipment false alarm identification method | |
| CN108667684B (en) | Data flow anomaly detection method based on local vector dot product density | |
| CN114048870A (en) | Power system abnormity monitoring method based on log characteristic intelligent mining | |
| Kamat et al. | Deep learning-based anomaly-onset aware remaining useful life estimation of bearings | |
| CN112001409A (en) | Power distribution network line loss abnormity diagnosis method and system based on K-means clustering algorithm | |
| US20060184474A1 (en) | Data analysis apparatus, data analysis program, and data analysis method | |
| CN112365060B (en) | Preprocessing method for network Internet of things sensing data | |
| CN111767162B (en) | Fault prediction method for hard disks of different models and electronic device | |
| CN116304909A (en) | Abnormality detection model training method, fault scene positioning method and device | |
| CN113536066A (en) | Data anomaly detection algorithm determination method and device and computer equipment | |
| CN117689661B (en) | Method and system for detecting coating defects on surface of medical breathable material | |
| CN112732690B (en) | Stabilizing system and method for chronic disease detection and risk assessment | |
| CN117994955A (en) | Method and device for building and alarming temperature alarm model of hydroelectric generating set | |
| CN108537249B (en) | Industrial process data clustering method for density peak clustering | |
| Chen et al. | GAIA: delving into gradient-based attribution abnormality for out-of-distribution detection | |
| CN117743803A (en) | Workload perception instant defect prediction method based on evolutionary feature construction | |
| CN113515450A (en) | Environment anomaly detection method and system | |
| CN117196649A (en) | Carbon emission prediction method under data characteristic limited scene | |
| CN116720079A (en) | Wind driven generator fault mode identification method and system based on multi-feature fusion | |
| CN114674511B (en) | Bridge modal anomaly early warning method for eliminating time-varying environmental factor influence |
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 |