CN118013304B - Transformer fault positioning method and system based on clustering algorithm - Google Patents
Transformer fault positioning method and system based on clustering algorithm Download PDFInfo
- Publication number
- CN118013304B CN118013304B CN202410183750.0A CN202410183750A CN118013304B CN 118013304 B CN118013304 B CN 118013304B CN 202410183750 A CN202410183750 A CN 202410183750A CN 118013304 B CN118013304 B CN 118013304B
- Authority
- CN
- China
- Prior art keywords
- data
- fault
- monitoring
- abnormal
- 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
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 18
- 238000012544 monitoring process Methods 0.000 claims abstract description 159
- 238000004458 analytical method Methods 0.000 claims abstract description 121
- 230000002159 abnormal effect Effects 0.000 claims abstract description 115
- 238000012545 processing Methods 0.000 claims abstract description 30
- 239000013598 vector Substances 0.000 claims description 46
- 238000011156 evaluation Methods 0.000 claims description 20
- 230000001186 cumulative effect Effects 0.000 claims description 16
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 230000005484 gravity Effects 0.000 claims description 5
- 230000004807 localization Effects 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- 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
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Locating Faults (AREA)
Abstract
The invention discloses a transformer fault positioning method and system based on a clustering algorithm, which belong to the technical field of transformer fault positioning and comprise a data acquisition module, a data processing module, a monitoring analysis module, a fault analysis module and a fault positioning module; the data acquisition module is used for carrying out real-time data acquisition on the transformer according to each preset monitoring item to obtain monitoring item data corresponding to each monitoring item, and integrating the monitoring item data into monitoring data; establishing a transformer model corresponding to the transformer; inputting the monitoring data into a transformer model for real-time display; the data processing module is used for processing the monitoring data to obtain corresponding abnormal analysis data and monitoring analysis data; the monitoring analysis module is used for processing the monitoring analysis data and determining corresponding abnormal analysis data; the fault analysis module is used for analyzing the abnormal analysis data and determining the corresponding fault type; the fault locating module is used for determining the corresponding fault position according to the fault type.
Description
Technical Field
The invention belongs to the technical field of transformer fault positioning, and particularly relates to a transformer fault positioning method and system based on a clustering algorithm.
Background
In a substation, a transformer is one of the most important power equipment, and its operation condition is directly related to the safety and reliability of a power generation and supply system. Therefore, it is necessary to perform fault diagnosis and fault localization for the transformer.
Conventional fault localization methods are typically based on measurement and analysis of electrical parameters such as current, voltage, and impedance. However, these methods may not accurately determine the location of the fault in some situations, particularly for some complex fault conditions. In addition, some conventional fault localization methods require manual intervention and operation, which not only increases the time and cost of fault handling, but also may lead to misjudgment and missed judgment. Therefore, when the transformer fails, how to accurately determine the failure location is a problem that needs to be solved at present; based on the method, the invention provides a transformer fault positioning method and a system based on a clustering algorithm.
Disclosure of Invention
In order to solve the problems of the scheme, the invention provides a transformer fault positioning method and system based on a clustering algorithm.
The aim of the invention can be achieved by the following technical scheme:
a transformer fault positioning system based on a clustering algorithm comprises a data acquisition module, a data processing module, a monitoring analysis module, a fault analysis module and a fault positioning module;
The data acquisition module is used for acquiring real-time data of the transformer according to each preset monitoring item, obtaining monitoring item data corresponding to each monitoring item, and integrating the monitoring item data into monitoring data; establishing a transformer model corresponding to the transformer; and inputting the monitoring data into the transformer model for real-time display.
The data processing module is used for processing the monitoring data to obtain corresponding abnormal analysis data and monitoring analysis data.
Further, the working method of the data processing module comprises the following steps:
Acquiring historical monitoring data of the transformer, establishing a corresponding abnormal recognition model based on the historical monitoring data, and analyzing monitoring item data through the abnormal recognition model to obtain an abnormal value set corresponding to each monitoring item;
generating an abnormal coordinate graph corresponding to each monitoring item according to the abnormal value set;
identifying each unit section in the abnormal coordinate graph, and marking the unit time length corresponding to each unit section and the unit interval time length between each adjacent unit section;
combining the unit sections to obtain a plurality of abnormal sections;
identifying the abnormal time length corresponding to each abnormal section, and marking the monitoring data with the abnormal time length larger than a threshold value X1 as abnormal analysis data;
and removing the monitoring item data and the abnormal analysis data with the abnormal time length not more than the threshold value X1 from the monitoring data to obtain corresponding monitoring analysis data.
Further, the method for merging the unit sections comprises the following steps:
Step SA1: calculating a merging limit value of each unit segment according to a formula tw=lg (5+T d);
Wherein: TW is the merge limit for the corresponding cell segment; t d is the unit time length of the corresponding unit section; lg is a logarithmic function based on 10;
Step SA2: identifying a corresponding evaluation segment; when the evaluation section is not available, finishing the combination of the unit sections and outputting corresponding abnormal sections;
Step SA3: identifying a merging limit value of the evaluation segment, and comparing the merging limit value of the evaluation segment with the corresponding unit interval duration;
when the merging limit value is not greater than the unit interval duration, marking the corresponding evaluation segment as an abnormal segment, and returning to the step SA2;
when the combination limit value is greater than the unit interval duration, combining the evaluation segment with the corresponding unit segment to form a new unit segment; calculating the merging limit value of the new unit section; returning to step SA2.
The monitoring analysis module is used for processing the monitoring analysis data and determining corresponding abnormal analysis data.
Further, the working method of the monitoring and analyzing module comprises the following steps:
acquiring monitoring analysis data, converting the monitoring analysis data into corresponding monitoring vectors, and merging based on the monitoring vectors to obtain a plurality of data classes;
identifying monitoring analysis data corresponding to each data class, and marking the monitoring analysis data as data class data;
And determining corresponding abnormal analysis data according to the positions of the data types.
Further, the data class acquisition method comprises the following steps:
Step SC1: forming a corresponding sample set d= { a 1,a2,a3,……,an};ai as an i-th monitoring vector according to each monitoring vector, wherein i=1, 2, … …, n and n are positive integers;
step SC2: all sample points in the sample set are regarded as an independent class cluster, and the distance d (C 1,C2),d(C1,C2) between two classes of clusters is calculated to satisfy the following conditions:
step SC3: when d (C 1,C2) is not more than the threshold value X2, merging the two corresponding class clusters to form a new class cluster; returning to the step SC2;
when d (C 1,C2) is greater than the threshold value X2, not merging, returning to the step SC2; until all the class clusters cannot be combined, a plurality of data classes are obtained.
The fault analysis module is used for analyzing the abnormal analysis data and determining the corresponding fault type.
The fault locating module is used for determining a corresponding fault position according to the fault type, identifying the corresponding fault type and determining a corresponding fault range according to the obtained fault type;
Identifying abnormal analysis data corresponding to the fault type, determining fault values of all the transformer components in the fault range based on the abnormal analysis data, and correspondingly marking the obtained fault values of all the transformer components in the transformer model; sequencing the calculated fault values in order from big to small; marking the first-ordered fault value as a reference fault value, and identifying the fault value with the absolute value of the difference value smaller than the threshold value X4 as an auxiliary fault value; and integrating the obtained reference fault value and the auxiliary fault value into positioning data, and determining the corresponding fault position according to the obtained positioning data.
Further, the method for calculating the fault value of each transformer assembly comprises the following steps:
acquiring historical fault data of a transformer, and counting the fault rate of each transformer component according to the historical fault data;
Acquiring reference data of each transformer assembly; converting the obtained reference data into corresponding reference vectors, and setting corresponding reference pictures according to the reference vectors, wherein corresponding reference areas are marked in the reference pictures; the reference area is composed of areas corresponding to the corresponding reference vectors;
Converting the obtained abnormal analysis data into an analysis vector set; marking the obtained analysis vector set in the reference graph, and identifying the proportion of the corresponding region of the analysis vector set in the reference graph; marking the analysis vector set which is not positioned in a reference area in the reference image as a cumulative vector, and identifying the corresponding cumulative proportion of each cumulative vector; the cumulative number is marked as LF j, j=1, 2, … …, m being a positive integer; the cumulative percentage was re-labeled LB j; marking the area specific gravity with QB;
According to the formula Calculating a corresponding fault value;
wherein: GZ is a fault value.
Further, the method for determining the fault location according to the positioning data comprises the following steps:
Establishing a fault location library, wherein the fault location library is used for storing fault location data, and the fault location data comprises historical location data and corresponding fault positions;
identifying a transformer assembly combination corresponding to the positioning data, and inputting the obtained transformer assembly combination into a fault positioning library for matching to obtain corresponding fault positioning data;
and identifying the historical positioning data corresponding to the fault positioning data, calculating the similarity between the positioning data and the historical positioning data, marking the fault positioning data with the highest similarity as target positioning data, identifying the fault position corresponding to the target positioning data, and outputting the obtained fault position.
A transformer fault positioning method based on a clustering algorithm comprises the following steps:
real-time data acquisition is carried out on the transformer according to each preset monitoring item, monitoring item data corresponding to each monitoring item are obtained, and each monitoring item data is integrated into monitoring data; establishing a transformer model corresponding to the transformer; inputting the monitoring data into the transformer model for real-time display;
Processing the monitoring data to obtain corresponding abnormal analysis data and monitoring analysis data; processing the monitoring analysis data to obtain corresponding abnormal analysis data;
Analyzing the abnormal analysis data to determine the corresponding fault type;
Identifying a corresponding fault type, and determining a corresponding fault range according to the obtained fault type;
Identifying abnormal analysis data corresponding to the fault type, determining fault values of all the transformer components in the fault range based on the abnormal analysis data, and correspondingly marking the obtained fault values of all the transformer components in the transformer model; sequencing the calculated fault values in order from big to small; marking the first-ordered fault value as a reference fault value, and identifying the fault value with the absolute value of the difference value smaller than the threshold value X4 as an auxiliary fault value; and integrating the obtained reference fault value and the auxiliary fault value into positioning data, and determining the corresponding fault position according to the obtained positioning data.
Compared with the prior art, the invention has the beneficial effects that:
The monitoring data of the transformer is processed in real time by arranging the data processing module, the influence of abnormal data in the monitoring data on subsequent fault positioning is removed, and particularly, the influence of the abnormal data on a subsequent clustering algorithm is solved; and the fault positioning precision of the transformer is improved. Through the mutual coordination among the data acquisition module, the data processing module, the monitoring analysis module and the fault analysis module, whether faults exist or not and the corresponding fault types are intelligently determined according to the monitoring data, and data support is provided for fault positioning. The fault location module is arranged to determine the corresponding fault position according to the fault type, and the obtained fault value of each component is used for deep analysis, so that the analysis precision of the fault position is further improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive effort to a person skilled in the art.
Fig. 1 is a functional block diagram of the present invention.
Detailed Description
The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1, a transformer fault positioning system based on a clustering algorithm comprises a data acquisition module, a data processing module, a monitoring analysis module, a fault analysis module and a fault positioning module;
The data acquisition module is used for carrying out real-time data acquisition on the transformer to obtain corresponding monitoring data, and carrying out data acquisition according to each preset monitoring item to obtain monitoring data formed by combining the data of each monitoring item; such as temperature, voltage, current, sound, etc., wherein each monitoring item is set according to a fault that the transformer may have and a corresponding manifestation of each fault, so as to determine whether the transformer has the fault or not and a corresponding fault type based on corresponding monitoring data; establishing a corresponding three-dimensional data model of the transformer, which is used for displaying specific structure and other data of the transformer, marking the transformer model, and inserting corresponding display nodes into the transformer model for corresponding monitoring data; and inputting the obtained monitoring data into a transformer model for real-time display.
The data processing module is used for processing the monitoring data acquired by the acquisition module to acquire historical monitoring data of the transformer, and if the data is insufficient, the historical monitoring data of the transformers of the same type can be acquired; establishing a corresponding abnormal recognition model according to the obtained historical monitoring data, and carrying out abnormal analysis on each monitoring item data in the monitoring data through the established abnormal recognition model to obtain an abnormal value set corresponding to each monitoring item; the anomaly identification model is established according to the existing isolated forest algorithm, and the expression isThe input data is xc, c represents a corresponding monitoring item, c=1, 2, … …, v is a positive integer; xc is monitoring item data of the corresponding monitoring item; the output is an outlier, i.e., 1 or 0; abnormal values corresponding to monitoring items at all times can be determined through an abnormal recognition model;
Generating an abnormal coordinate graph corresponding to each monitoring item according to the obtained abnormal value set, wherein the horizontal axis is time, and the vertical axis is abnormal value;
marking a continuous line segment corresponding to an abnormal value of 1 in the abnormal coordinate graph as a unit segment; identifying each unit section in the abnormal coordinate graph, and identifying the unit time length corresponding to each unit section and the unit interval time length between each adjacent unit section;
merging the unit sections to obtain a plurality of abnormal sections;
Identifying the abnormal time length corresponding to each abnormal section, namely, the time length from the starting time to the ending time of the abnormal section is the abnormal time length; integrating the data of each monitoring item corresponding to the abnormal time length which is greater than the threshold value X1 into abnormal analysis data; that is, an abnormal time period corresponding to more than the threshold value X1 is determined, and the monitoring data in the abnormal time period is marked as abnormal analysis data.
And eliminating the monitoring item data and the abnormal analysis data corresponding to the abnormal time length which is not more than the threshold value X1 from the monitoring data to obtain the corresponding monitoring analysis data.
The monitoring data of the transformer is processed in real time by arranging the data processing module, the influence of abnormal data in the monitoring data on subsequent fault positioning is removed, and particularly, the influence of the abnormal data on a subsequent clustering algorithm is solved; and the fault positioning precision of the transformer is improved.
The method for merging the unit sections comprises the following steps:
Step SA1: calculating a merge limit for the corresponding cell segment according to the formula tw=lg (5+T d); wherein: TW is the merge limit for the corresponding cell segment; t d is the unit time length of the corresponding unit section; lg is a logarithmic function based on 10.
Step SA2: starting from the 0 point of the abnormal coordinate graph, evaluating along the horizontal axis direction, identifying a corresponding unit section to be evaluated, marking the unit section as an evaluation section, and marking the first unit section of the non-abnormal section as the evaluation section; when the evaluation section is not available, finishing the combination of the unit sections and outputting corresponding abnormal sections;
Step SA3: identifying a merging limit value of the evaluation segment, and comparing the merging limit value of the evaluation segment with the corresponding unit interval duration;
when the merging limit value is not greater than the unit interval duration, marking the corresponding evaluation segment as an abnormal segment, and returning to the step SA2;
When the combination limit value is greater than the unit interval duration, combining the evaluation segment with the corresponding unit segment to form a new unit segment, namely integrally combining the evaluation segment, the interval segment and the corresponding unit; calculating the merging limit value of the new unit section; returning to step SA2.
The monitoring analysis module is used for processing the monitoring analysis data and determining corresponding abnormal analysis data. The specific process is as follows:
Step SC1: acquiring monitoring analysis data, and converting the acquired monitoring analysis data into corresponding monitoring vectors, namely taking the data corresponding to each monitoring item as corresponding elements in the monitoring vectors into the monitoring vectors to form the monitoring vectors; if the data corresponding to some monitoring items are non-numerical data, the current numerical conversion can be utilized for assignment; forming a corresponding sample set D= { a 1,a2,a3,……,an};ai according to each monitoring vector, wherein i=1, 2, … …, n and n are positive integers;
step SC2: all sample points in the sample set are regarded as an independent class cluster, and the distance d (C 1,C2),d(C1,C2) between two classes of clusters is calculated to satisfy the following conditions:
step SC3: when d (C 1,C2) is not more than the threshold value X2, merging the two corresponding class clusters to form a new class cluster; returning to the step SC2;
When d (C 1,C2) is greater than the threshold value X2, not merging, returning to the step SC2; until all the class clusters can not be combined, a plurality of data classes, namely the rest class clusters, are obtained;
Identifying monitoring analysis data corresponding to each data class, and marking the monitoring analysis data as data class data;
and determining corresponding abnormal analysis data according to the positions of the data types.
Simulating by using a large amount of historical monitoring analysis data, determining a plurality of historical data types of data, and determining the fault probability corresponding to each region in the space according to the corresponding fault results and the positions corresponding to each fault result; marking the region with the fault probability larger than the threshold value X3 as an abnormal region, and marking data class data corresponding to the data class in the abnormal region with abnormal analysis data; in particular, other manners can be used to determine corresponding anomaly analysis data according to the positions of the data classes.
The fault analysis module is used for analyzing the abnormal analysis data and determining the corresponding fault type.
For specific fault types, the determination can be performed by combining the existing fault determination technology, such as the determination of abnormality analysis data determined by data classes according to the corresponding fault types; the other abnormal analysis data can specifically analyze the monitoring item data with abnormal conditions, and determine the corresponding fault type according to the corresponding characteristics; thus, after determining the corresponding anomaly analysis data, the determination of the fault type may be performed in a variety of existing ways.
Through the mutual coordination among the data acquisition module, the data processing module, the monitoring analysis module and the fault analysis module, whether faults exist or not and the corresponding fault types are intelligently determined according to the monitoring data, and data support is provided for fault positioning.
The fault positioning module is used for determining a corresponding fault position according to the fault type, identifying the corresponding fault type, and identifying a transformer component of the fault type possibly occurring in the transformer according to the obtained fault type to form a corresponding fault range, namely the fault range is composed of a plurality of transformer components of possible fault sources;
Determining fault values of all the transformer components in the fault range according to the abnormality analysis data corresponding to the faults, sequencing the calculated fault values in sequence from large to small, marking the fault value in the first sequence as a reference fault value, identifying fault values of which the absolute value of the difference value between other fault values and the reference fault value is smaller than a threshold value X4, and marking the fault values as auxiliary fault values; and integrating the obtained reference fault value and the auxiliary fault value into positioning data, and determining the corresponding fault position according to the obtained positioning data.
The method for determining the fault value of each transformer assembly in the fault range according to the abnormality analysis data corresponding to the fault comprises the following steps:
acquiring historical fault data of the transformer, and counting the fault rate of each transformer component for generating the fault according to the historical fault data;
Counting the corresponding historical monitoring data of each transformer component when the fault occurs, and marking the historical monitoring data as reference data; converting the obtained reference data into corresponding reference vectors according to a conversion mode of the corresponding monitoring vectors; inputting each obtained reference vector of the component into a vector space, marking a corresponding range, and forming a reference diagram corresponding to the component; the region corresponding to the reference vector in the reference picture is marked as a reference region;
Converting the obtained abnormal analysis data according to a conversion mode of the reference vector to form a corresponding analysis vector set; marking the obtained analysis vector set in a reference graph, and identifying the proportion of the analysis vector set in the reference graph, which is positioned in a reference area, and marking the proportion as the area proportion; marking the analysis vectors which are not positioned in a reference area in the reference graph as cumulative vectors, identifying the specific gravity of each cumulative vector, and marking the cumulative vectors as cumulative specific gravity; the cumulative number is marked as LF j, j=1, 2, … …, m being a positive integer; the cumulative percentage was re-labeled LB j; marking the area specific gravity with QB;
According to the formula Calculating a corresponding fault value; wherein: GZ is a fault value.
The method for determining the corresponding fault position according to the obtained positioning data comprises the following steps:
Acquiring historical fault data of transformers of the same type, processing according to the acquired historical fault data to form fault positioning data, namely processing the historical fault data according to the processing process of the positioning data to form corresponding historical positioning data, identifying fault positions corresponding to the corresponding historical fault data, and integrating the fault positions into the fault positioning data; processing according to a large amount of historical fault data to obtain a large amount of fault positioning data, and establishing a corresponding fault positioning library;
Identifying a transformer assembly combination corresponding to the positioning data, inputting the obtained transformer assembly combination into a fault positioning library for matching, and obtaining fault positioning data corresponding to the historical positioning data of the transformer assembly combination only;
Identifying historical positioning data corresponding to the fault positioning data, calculating the similarity between the positioning data and the historical positioning data, and calculating the similarity according to the corresponding auxiliary fault value and the corresponding reference fault value; the cosine similarity algorithm can be utilized for calculation; and marking the fault positioning data with highest similarity as target positioning data, identifying a fault position corresponding to the target positioning data, and outputting the obtained fault position.
The fault location module is arranged to determine the corresponding fault position according to the fault type, and the obtained fault value of each component is used for deep analysis, so that the analysis precision of the fault position is further improved.
A transformer fault positioning method based on a clustering algorithm comprises the following steps:
Acquiring monitoring item data corresponding to each monitoring item, and integrating the monitoring item data into monitoring data; establishing a transformer model corresponding to the transformer; inputting the monitoring data into a transformer model for real-time display;
Acquiring historical monitoring data of the transformer, establishing a corresponding abnormal recognition model based on the historical monitoring data, and analyzing monitoring item data through the abnormal recognition model to obtain an abnormal value set corresponding to each monitoring item;
generating an abnormal coordinate graph corresponding to each monitoring item according to the abnormal value set;
identifying each unit section in the abnormal coordinate graph, and marking the unit time length corresponding to each unit section and the unit interval time length between each adjacent unit section;
combining the unit sections to obtain a plurality of abnormal sections;
identifying the abnormal time length corresponding to each abnormal section, and marking the monitoring data with the abnormal time length larger than a threshold value X1 as abnormal analysis data;
and removing the monitoring item data and the abnormal analysis data with the abnormal time length not more than the threshold value X1 from the monitoring data to obtain corresponding monitoring analysis data.
Acquiring monitoring analysis data, converting the monitoring analysis data into corresponding monitoring vectors, and merging based on each monitoring vector to acquire a plurality of data classes;
Identifying monitoring analysis data corresponding to each data class, and marking the monitoring analysis data as data class data;
and determining corresponding abnormal analysis data according to the positions of the data types.
And analyzing the abnormal analysis data to determine the corresponding fault type.
Identifying a corresponding fault type, and determining a corresponding fault range according to the obtained fault type;
Identifying abnormal analysis data corresponding to the fault types, determining fault values of all the transformer components in the fault range based on the abnormal analysis data, and correspondingly marking the obtained fault values of all the transformer components in a transformer model; sequencing the calculated fault values in order from big to small; marking the first-ordered fault value as a reference fault value, and identifying the fault value with the absolute value of the difference value smaller than the threshold value X4 as an auxiliary fault value; and integrating the obtained reference fault value and the auxiliary fault value into positioning data, and determining the corresponding fault position according to the obtained positioning data.
In a specific unpublished section, reference is made to an embodiment of a transformer fault localization system based on a clustering algorithm.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas which are obtained by acquiring a large amount of data and performing software simulation to obtain the closest actual situation, and preset parameters and preset thresholds in the formulas are set by a person skilled in the art according to the actual situation or are obtained by simulating a large amount of data.
The above embodiments are only for illustrating the technical method of the present invention and not for limiting the same, and it should be understood by those skilled in the art that the technical method of the present invention may be modified or substituted without departing from the spirit and scope of the technical method of the present invention.
Claims (2)
1. The transformer fault positioning system based on the clustering algorithm is characterized by comprising a data acquisition module, a data processing module, a monitoring analysis module, a fault analysis module and a fault positioning module;
The data acquisition module is used for acquiring real-time data of the transformer according to each preset monitoring item, obtaining monitoring item data corresponding to each monitoring item, and integrating the monitoring item data into monitoring data; establishing a transformer model corresponding to the transformer; inputting the monitoring data into the transformer model for real-time display;
The data processing module is used for processing the monitoring data to obtain corresponding abnormal analysis data and monitoring analysis data;
The monitoring analysis module is used for processing the monitoring analysis data and determining corresponding abnormal analysis data;
the fault analysis module is used for analyzing the abnormal analysis data and determining the corresponding fault type;
the fault locating module is used for determining a corresponding fault position according to the fault type, identifying the corresponding fault type and determining a corresponding fault range according to the obtained fault type;
Identifying abnormal analysis data corresponding to the fault type, determining fault values of all the transformer components in the fault range based on the abnormal analysis data, and correspondingly marking the obtained fault values of all the transformer components in the transformer model; sequencing the calculated fault values in order from big to small; marking the first-ordered fault value as a reference fault value, and identifying the fault value with the absolute value of the difference value smaller than the threshold value X4 as an auxiliary fault value; integrating the obtained reference fault value and the auxiliary fault value into positioning data, and determining a corresponding fault position according to the obtained positioning data;
the working method of the data processing module comprises the following steps:
Acquiring historical monitoring data of the transformer, establishing a corresponding abnormal recognition model based on the historical monitoring data, and analyzing monitoring item data through the abnormal recognition model to obtain an abnormal value set corresponding to each monitoring item;
generating an abnormal coordinate graph corresponding to each monitoring item according to the abnormal value set;
identifying each unit section in the abnormal coordinate graph, and marking the unit time length corresponding to each unit section and the unit interval time length between each adjacent unit section;
combining the unit sections to obtain a plurality of abnormal sections;
identifying the abnormal time length corresponding to each abnormal section, and marking the monitoring data with the abnormal time length larger than a threshold value X1 as abnormal analysis data;
removing monitoring item data and abnormal analysis data with abnormal time length not more than a threshold value X1 from the monitoring data to obtain corresponding monitoring analysis data;
The method for merging the unit sections comprises the following steps:
Step SA1: calculating a merging limit value of each unit segment according to a formula tw=lg (5+T d);
Wherein: TW is the merge limit for the corresponding cell segment; t d is the unit time length of the corresponding unit section; lg is a logarithmic function based on 10;
Step SA2: identifying a corresponding evaluation segment; when the evaluation section is not available, finishing the combination of the unit sections and outputting corresponding abnormal sections;
Step SA3: identifying a merging limit value of the evaluation segment, and comparing the merging limit value of the evaluation segment with the corresponding unit interval duration;
when the merging limit value is not greater than the unit interval duration, marking the corresponding evaluation segment as an abnormal segment, and returning to the step SA2;
when the combination limit value is greater than the unit interval duration, combining the evaluation segment with the corresponding unit segment to form a new unit segment; calculating the merging limit value of the new unit section; returning to the step SA2;
The working method of the monitoring and analyzing module comprises the following steps:
acquiring monitoring analysis data, converting the monitoring analysis data into corresponding monitoring vectors, and merging based on the monitoring vectors to obtain a plurality of data classes;
identifying monitoring analysis data corresponding to each data class, and marking the monitoring analysis data as data class data;
Determining corresponding abnormal analysis data according to the positions of the data types;
The data class acquisition method comprises the following steps:
Step SC1: forming a corresponding sample set d= { a 1,a2,a3,……,an};ai as an i-th monitoring vector according to each monitoring vector, wherein i=1, 2, … …, n and n are positive integers;
step SC2: all sample points in the sample set are regarded as an independent class cluster, and the distance d (C 1,C2),d(C1,C2) between two classes of clusters is calculated to satisfy the following conditions:
；
step SC3: when d (C 1,C2) is not more than the threshold value X2, merging the two corresponding class clusters to form a new class cluster; returning to the step SC2;
When d (C 1,C2) is greater than the threshold value X2, not merging, returning to the step SC2; until all the class clusters can not be combined, a plurality of data classes are obtained;
The method for calculating the fault value of each transformer component comprises the following steps:
acquiring historical fault data of a transformer, and counting the fault rate of each transformer component according to the historical fault data;
Acquiring reference data of each transformer assembly; converting the obtained reference data into corresponding reference vectors, and setting corresponding reference pictures according to the reference vectors, wherein corresponding reference areas are marked in the reference pictures; the reference area is composed of areas corresponding to the corresponding reference vectors;
Converting the obtained abnormal analysis data into an analysis vector set; marking the obtained analysis vector set in the reference graph, and identifying the proportion of the corresponding region of the analysis vector set in the reference graph; marking the analysis vector set which is not positioned in a reference area in the reference image as a cumulative vector, and identifying the corresponding cumulative proportion of each cumulative vector; the cumulative number is marked as LF j, j=1, 2, … …, m being a positive integer; the cumulative percentage was re-labeled LB j; marking the area specific gravity with QB;
According to the formula Calculating a corresponding fault value;
Wherein: GZ is a fault value;
The method for determining the fault position according to the positioning data comprises the following steps:
Establishing a fault location library, wherein the fault location library is used for storing fault location data, and the fault location data comprises historical location data and corresponding fault positions;
identifying a transformer assembly combination corresponding to the positioning data, and inputting the obtained transformer assembly combination into a fault positioning library for matching to obtain corresponding fault positioning data;
and identifying the historical positioning data corresponding to the fault positioning data, calculating the similarity between the positioning data and the historical positioning data, marking the fault positioning data with the highest similarity as target positioning data, identifying the fault position corresponding to the target positioning data, and outputting the obtained fault position.
2. A transformer fault locating method based on a clustering algorithm, which is applied to the transformer fault locating system based on the clustering algorithm as claimed in claim 1, wherein the method comprises the following steps:
real-time data acquisition is carried out on the transformer according to each preset monitoring item, monitoring item data corresponding to each monitoring item are obtained, and each monitoring item data is integrated into monitoring data; establishing a transformer model corresponding to the transformer; inputting the monitoring data into the transformer model for real-time display;
Processing the monitoring data to obtain corresponding abnormal analysis data and monitoring analysis data; processing the monitoring analysis data to obtain corresponding abnormal analysis data;
Analyzing the abnormal analysis data to determine the corresponding fault type;
Identifying a corresponding fault type, and determining a corresponding fault range according to the obtained fault type;
Identifying abnormal analysis data corresponding to the fault type, determining fault values of all the transformer components in the fault range based on the abnormal analysis data, and correspondingly marking the obtained fault values of all the transformer components in the transformer model; sequencing the calculated fault values in order from big to small; marking the first-ordered fault value as a reference fault value, and identifying the fault value with the absolute value of the difference value smaller than the threshold value X4 as an auxiliary fault value; and integrating the obtained reference fault value and the auxiliary fault value into positioning data, and determining the corresponding fault position according to the obtained positioning data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410183750.0A CN118013304B (en) | 2024-02-19 | 2024-02-19 | Transformer fault positioning method and system based on clustering algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410183750.0A CN118013304B (en) | 2024-02-19 | 2024-02-19 | Transformer fault positioning method and system based on clustering algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118013304A CN118013304A (en) | 2024-05-10 |
| CN118013304B true CN118013304B (en) | 2024-09-13 |
Family
ID=90955764
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410183750.0A Active CN118013304B (en) | 2024-02-19 | 2024-02-19 | Transformer fault positioning method and system based on clustering algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118013304B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA3177620A1 (en) * | 2021-05-06 | 2022-11-06 | Strong Force Iot Portfolio 2016, Llc | Quantum, biological, computer vision, and neural network systems for industrial internet of things |
| CN116660669A (en) * | 2023-07-26 | 2023-08-29 | 威海双城电气有限公司 | Power equipment fault on-line monitoring system and method |
Family Cites Families (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7930593B2 (en) * | 2008-06-23 | 2011-04-19 | Hewlett-Packard Development Company, L.P. | Segment-based technique and system for detecting performance anomalies and changes for a computer-based service |
| US9606520B2 (en) * | 2009-06-22 | 2017-03-28 | Johnson Controls Technology Company | Automated fault detection and diagnostics in a building management system |
| CN111400930B (en) * | 2020-04-09 | 2022-04-15 | 武汉大学 | Power equipment small sample fault diagnosis method and system based on virtual and real twin space |
| CN113139610A (en) * | 2021-04-29 | 2021-07-20 | 国网河北省电力有限公司电力科学研究院 | Abnormity detection method and device for transformer monitoring data |
| EP4087119A1 (en) * | 2021-05-06 | 2022-11-09 | Siemens Gamesa Renewable Energy A/S | Method for detecting an electrical fault in the stator of an electric machine, especially in the generator of a wind turbine |
| CN115988243A (en) * | 2021-10-14 | 2023-04-18 | 中兴通讯股份有限公司 | Fault positioning method and system, computer readable storage medium |
| CN114417921B (en) * | 2022-01-05 | 2024-04-30 | 中国国家铁路集团有限公司 | Method, device, equipment and storage medium for identifying geometric anomaly data of track |
| CN115130578A (en) * | 2022-06-29 | 2022-09-30 | 南京邮电大学 | Incremental rough clustering-based online evaluation method for state of power distribution equipment |
| CN115099280A (en) * | 2022-07-11 | 2022-09-23 | 国网湖北省电力有限公司襄阳供电公司 | Prony iterative analysis method, computing equipment, storage medium and power grid system |
| CN116861354A (en) * | 2023-07-17 | 2023-10-10 | 国家电网有限公司华北分部 | Abnormality identification and mode distinguishing method and system for monitoring data of power transformer |
| CN116703915B (en) * | 2023-08-07 | 2023-09-29 | 天津医科大学眼科医院 | Diabetes retinopathy consultation method and system based on multiple image fusion |
| CN117540262A (en) * | 2023-09-28 | 2024-02-09 | 贵州电网有限责任公司 | Transformer fault prediction modeling method based on feedforward and feedback combination |
| CN117439101B (en) * | 2023-12-21 | 2024-06-04 | 武汉市豪迈电力自动化技术有限责任公司 | Intelligent network for interaction of new energy and flexible load in power grid |
-
2024
- 2024-02-19 CN CN202410183750.0A patent/CN118013304B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA3177620A1 (en) * | 2021-05-06 | 2022-11-06 | Strong Force Iot Portfolio 2016, Llc | Quantum, biological, computer vision, and neural network systems for industrial internet of things |
| CN116660669A (en) * | 2023-07-26 | 2023-08-29 | 威海双城电气有限公司 | Power equipment fault on-line monitoring system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118013304A (en) | 2024-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11624731B2 (en) | Method for predicting remaining life of numerical control machine tool | |
| CN111459700B (en) | Equipment fault diagnosis method, diagnosis device, diagnosis equipment and storage medium | |
| EP3385889A1 (en) | Abnormality detection system, abnormality detection method, abnormality detection program, and method for generating learned model | |
| US11782430B2 (en) | Abnormality diagnosis method, abnormality diagnosis device and non-transitory computer readable storage medium | |
| JP2015114967A (en) | Abnormality detection method and abnormality detection device | |
| CN112955837A (en) | Abnormality diagnosis device, abnormality diagnosis method, and program | |
| JP3718765B2 (en) | Plant diagnostic equipment | |
| CN110222765A (en) | A kind of permanent magnet synchronous motor health status monitoring method and system | |
| CN116520068B (en) | Diagnostic method, device, equipment and storage medium for electric power data | |
| CN116308883A (en) | Regional power grid data overall management system based on big data | |
| CN117498225B (en) | Unmanned aerial vehicle intelligent power line inspection system | |
| CN117791856B (en) | Power grid fault early warning method and device based on inspection robot | |
| CN118013304B (en) | Transformer fault positioning method and system based on clustering algorithm | |
| CN108574290B (en) | Method, device, terminal and readable storage medium for positioning oscillation source of forced oscillation | |
| CN112734637B (en) | Thermal infrared image processing method and system for monitoring temperature of lead | |
| EP4002021A1 (en) | System for monitoring a circuit breaker | |
| CN111352365B (en) | Dustproof ventilation type electric power and electrical equipment cabinet and control method | |
| CN117036732A (en) | Electromechanical equipment detection system, method and equipment based on fusion model | |
| CN115293303B (en) | High-voltage transmission line network monitoring method, system, equipment and medium | |
| Wu et al. | Hilbert ID considering multi-window feature extraction for transformer deep vision fault positioning | |
| CN113255771A (en) | Fault diagnosis method and system based on multi-dimensional heterogeneous difference analysis | |
| Xu | Research on a Small Target Object Detection Algorithm for Electric Transmission Lines Based on Convolutional Neural Network | |
| CN113065219A (en) | Method and device for detecting abnormality of train traction transformer cooling system | |
| WO2023105725A1 (en) | Time series data processing method | |
| CN117909120B (en) | Semantic understanding and logic reasoning method of digital intelligent expert |
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 |