CN116467655A - Capacitor fault prediction method, device, equipment and medium - Google Patents

Capacitor fault prediction method, device, equipment and medium Download PDF

Info

Publication number
CN116467655A
CN116467655A CN202310440205.0A CN202310440205A CN116467655A CN 116467655 A CN116467655 A CN 116467655A CN 202310440205 A CN202310440205 A CN 202310440205A CN 116467655 A CN116467655 A CN 116467655A
Authority
CN
China
Prior art keywords
capacitor
fault
candidate
attribute
state quantity
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
Application number
CN202310440205.0A
Other languages
Chinese (zh)
Inventor
黄书健
杨茂强
施理成
卢先锋
寨战争
纪经涛
陈晓鹏
蔡素雄
刘焕辉
张焕燊
吕志鹏
赖咏
李海发
邱睿
张玄
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Power Grid Co Ltd
Huizhou Power Supply Bureau of Guangdong Power Grid Co Ltd
Original Assignee
Guangdong Power Grid Co Ltd
Huizhou Power Supply Bureau of Guangdong Power Grid Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangdong Power Grid Co Ltd, Huizhou Power Supply Bureau of Guangdong Power Grid Co Ltd filed Critical Guangdong Power Grid Co Ltd
Priority to CN202310440205.0A priority Critical patent/CN116467655A/en
Publication of CN116467655A publication Critical patent/CN116467655A/en
Pending legal-status Critical Current

Links

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
    • G01R31/003—Environmental or reliability tests
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/20—Administration of product repair or maintenance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • 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

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Data Mining & Analysis (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • Quality & Reliability (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Artificial Intelligence (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Probability & Statistics with Applications (AREA)
  • Environmental & Geological Engineering (AREA)
  • Development Economics (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a capacitor fault prediction method, device, equipment and medium. The capacitor fault prediction method comprises the following steps: acquiring real-time operation data of fault state quantity of a capacitor in a transformer substation in a real-time working process; the fault state quantity is a state quantity which is in an abnormal state in the fault process of the capacitor; and determining a fault prediction result of the capacitor according to the real-time operation data based on the prediction probability of the candidate category in the fault prediction model. According to the embodiment of the invention, the real-time operation data of the capacitor related to the fault state quantity in the real-time working process is input into the fault prediction model, so that the operation state of the capacitor is predicted in advance, the power failure accident caused by the capacitor fault is reduced, and the utilization rate of the capacitor is improved.

Description

Capacitor fault prediction method, device, equipment and medium
Technical Field
The present invention relates to the field of power systems, and in particular, to a method, an apparatus, a device, and a medium for predicting a capacitor fault.
Background
Along with the rapid development of economy, the demand for electric power also increases sharply, the system scale also increases continuously, the demands of people on electricity quality and power supply reliability also increase continuously, and higher demands are also put forward on system voltage and reactive power adjustment. Therefore, the use of capacitance compensation devices is particularly important. The power capacitor is used as the reactive power compensation device with the most application in the transformer substation, and the fault rate is greatly improved. Therefore, it is important to detect the operation state of the capacitor.
The traditional state detection method of the capacitor is usually a power failure test, and cannot generate faults in real time, so that the workload of operation and maintenance staff is increased, the utilization rate of the capacitor is reduced, and the capacity of compensating the system voltage and reactive power is weakened. In addition, the capacitor is in a non-operation voltage and load state during the power failure test, and the insulation condition and the development trend of faults during the actual operation of the capacitor cannot be truly reflected. Therefore, the operation state of the capacitor needs to be subjected to fault prediction, so that the optimal maintenance strategy of the capacitor is effectively guided, and further accidents are avoided.
Disclosure of Invention
The invention provides a capacitor fault prediction method, device, equipment and medium, which are used for realizing fault prediction in the real-time operation process of a capacitor.
According to an aspect of the present invention, there is provided a capacitor failure prediction method including:
acquiring real-time operation data of fault state quantity of a capacitor in a transformer substation in a real-time working process; the fault state quantity is an abnormal state quantity in the fault process of the capacitor;
determining a fault prediction result of the capacitor according to the real-time operation data based on the prediction probability of the candidate class in the fault prediction model;
The fault prediction model is obtained through training in the following mode: determining the fault state quantity and the historical operation data of the fault state quantity from the historical operation data of the capacitor as training samples; and training an initial fault prediction model according to the training sample by adopting an integrated classifier of attribute importance to obtain the fault prediction model.
According to another aspect of the present invention, there is provided a capacitor failure prediction apparatus including:
the real-time data acquisition module is used for acquiring real-time operation data of fault state quantity of the capacitor in the transformer substation in the real-time working process; the fault state quantity is an abnormal state quantity in the fault process of the capacitor;
the fault prediction module is used for determining a fault prediction result of the capacitor according to the real-time operation data based on the prediction probability of the candidate category in the fault prediction model;
the fault prediction model is obtained through training in the following mode: determining the fault state quantity and the historical operation data of the fault state quantity from the historical operation data of the capacitor as training samples; and training an initial fault prediction model according to the training sample by adopting an integrated classifier of attribute importance to obtain the fault prediction model.
According to another aspect of the present invention, there is provided an electronic apparatus including:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the capacitor fault prediction method according to any one of the embodiments of the present invention.
According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to implement the capacitor fault prediction method according to any one of the embodiments of the present invention when executed.
According to the technical scheme, the historical operation data of the capacitor in the transformer substation in the historical fault process are analyzed, the fault state quantity affecting the capacitor is determined, the integrated classifier of the attribute importance is adopted, the fault prediction model is constructed according to the historical operation data of the fault state quantity, the structural redundancy of the fault prediction model is reduced, and the training efficiency of the fault prediction model on the data is improved; and inputting real-time operation data of the capacitor related to the fault state quantity in the real-time working process into a fault prediction model, and predicting the operation state of the capacitor in advance, so that the power failure accident caused by the capacitor fault is reduced, and the utilization rate of the capacitor is improved.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for predicting capacitor failure according to a first embodiment of the present invention;
fig. 2 is a flowchart of a capacitor fault prediction method according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a capacitor failure prediction device according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device implementing a capacitor failure prediction method according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent 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 present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "candidate," "target," and the like in the description and claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. In the technical scheme of the invention, the collection, storage, use, processing, transmission, provision, disclosure and the like of the related data accord with the regulations of related laws and regulations, and the public order is not violated.
Example 1
Fig. 1 is a flowchart of a capacitor fault prediction method according to an embodiment of the present invention, where the method may be implemented by a capacitor fault prediction device, and the capacitor fault prediction device may be implemented in hardware and/or software, and the capacitor fault prediction device may be configured in various general-purpose computing devices. As shown in fig. 1, the method includes:
S110, acquiring real-time operation data of fault state quantity of the capacitor in the transformer substation in a real-time working process. The fault state quantity is a state quantity which is in an abnormal state in the fault process of the capacitor.
Alternatively, the fault state quantity may include at least one of: the capacitance value of the capacitor, the dielectric loss value of the capacitor, the harmonic current of the capacitor and the operating voltage of the capacitor.
For example, the capacitance value of the capacitor may directly reflect the operation condition of the capacitor, and the types of faults occurring in the capacitor may be different, so that the capacitance values of the capacitor may be different. For example, the capacitance value of the capacitor may beWhen breakdown short-circuit fault occurs in the capacitor, the capacitance value of the capacitor can be +>When the capacitor has fuse blowing fault, the capacitance value of the capacitor can be +>Wherein n is the number of capacitance elements connected in parallel in the capacitor, m is the number of capacitance elements connected in series in the capacitor, C 0 The capacitance value of the single capacitive element is x, and the number of the capacitive elements connected in series when breakdown short-circuit fault occurs to the capacitor.
Illustratively, the dielectric loss value of the capacitor may reflect the magnitude of the active power consumed by the capacitor. Alternatively, the larger the dielectric loss value of the capacitor is, the larger the active power consumed by the capacitor is, and the more serious the temperature rise phenomenon of the capacitor is. Alternatively, the dielectric loss value of the capacitor may be expressed as Where ρ is the resistivity of the insulating medium, ε is the dielectric constant, f is the current frequency, and δ is the dielectric loss angle.
By way of example, the harmonic current generated by the capacitor can increase the active power loss of the capacitor, and the higher the active power loss of the capacitor, the higher the temperature of the capacitor, so that the probability of the capacitor to fail is increased, the aging process of the capacitor is accelerated, and the service life of the capacitor is shortened. Therefore, harmonic current generated by the capacitor needs to be collected and counted to be used as a fault factor affecting the normal operation of the capacitor.
For example, if the capacitor is in an overvoltage operation state for a long time, the loop current of the capacitor increases to raise the temperature of the capacitor, which causes aging of the insulating medium and is liable to cause insulation breakdown failure of the capacitor. Therefore, the operation voltage of the capacitor needs to be collected and counted as a fault factor affecting the normal operation of the capacitor.
Specifically, the operation data of the capacitor in the transformer substation about the fault state quantity in the real-time working process can be obtained as the real-time operation data of the capacitor.
S120, determining a fault prediction result of the capacitor according to the real-time operation data based on the prediction probability of the candidate category in the fault prediction model.
The fault prediction model can be obtained through training in the following way: determining a fault state quantity and historical operation data of the fault state quantity in the historical operation data of the capacitor as training samples; and training the initial fault prediction model according to the training sample by adopting an integrated classifier of attribute importance to obtain the fault prediction model.
The historical operation data of the capacitor may refer to operation data of the capacitor in a historical operation process. The historical operating data for the fault state quantity may refer to operating data of the capacitor during historical operation that is related to the fault state quantity. Attribute importance may refer to the importance of attributes in an integrated classifier. Alternatively, the integrated classifier may refer to a random forest algorithm (Random Forest Algorithm). Alternatively, the initial fault prediction model may be a model pre-trained with an integrated classifier that combines the importance of the attributes.
Optionally, determining the fault state quantity and the historical operation data of the fault state quantity according to the historical operation data of the capacitor is used as a training sample, and includes: analyzing the operation data of the state quantity of the capacitor in the historical failure process, and taking the state quantity of the operation data in the historical failure process as the failure state quantity affecting the capacitor failure according to the analysis result; the normal operation data of the fault state quantity of the capacitor in the normal working process is extracted from the historical operation data of the capacitor to serve as a positive training sample, and the abnormal operation data of the fault state quantity of the capacitor in the fault process is extracted from the historical operation data of the capacitor to serve as a negative training sample. Further, the number of positive training samples and negative training samples may be determined according to a preset ratio. Alternatively, the preset ratio may be set by those skilled in the art according to actual circumstances.
Specifically, the integrated classifier combined with the attribute importance degree can be pre-trained to obtain an initial fault prediction model, and the initial fault prediction model is trained according to the training sample to obtain the fault prediction model.
The prediction probability refers to a prediction parameter obtained by training a fault prediction model. The candidate category is used for representing the running state of the capacitor in the fault prediction model, and has corresponding prediction probability. By way of example, the operating conditions may include a normal operating condition and an abnormal operating condition, which may refer to a type of fault transmitted by the capacitor, such as a breakdown short circuit fault, a fuse blowing fault, and the like.
Specifically, the real-time operation data of the capacitor can be input into a fault prediction model, the fault prediction model can process the real-time operation data to obtain candidate categories corresponding to the real-time operation data, and a fault prediction result of the capacitor is determined according to the prediction probability of the candidate categories in the fault prediction model.
Optionally, determining the fault prediction result of the capacitor according to the real-time operation data based on the prediction probability of the candidate class in the fault prediction model includes: based on the fault prediction model, determining at least two corresponding candidate categories for real-time operation data, taking the corresponding candidate categories as candidate fault prediction results, and taking the prediction probability of the corresponding candidate categories as the prediction probability of the candidate fault results; sequencing at least two candidate prediction results according to the prediction probability of the candidate fault prediction results; and determining a target fault prediction result according to the sequencing result.
The candidate category may refer to a category corresponding to each leaf node of the fuzzy decision tree in the fault prediction model, and the candidate fault prediction result may refer to a category corresponding to real-time operation data input to the fault prediction model in the fuzzy decision tree. The target fault prediction result may refer to a final prediction result of real-time operation data determined from all candidate fault results.
Specifically, based on the fault prediction model, real-time operation data of the capacitor may be input into the fault prediction model, the fault prediction model may output at least two candidate categories corresponding to the real-time operation data as candidate fault prediction results, and a prediction probability of the corresponding candidate categories is used as a prediction probability of the candidate fault prediction results; sequencing at least two candidate fault prediction results according to the prediction probability corresponding to each candidate fault prediction result from small to large or from large to small; and selecting the candidate fault prediction result with the maximum prediction probability value as a target fault prediction result according to the sequencing result.
Alternatively, the prediction probability of the target fault prediction result may be determined by the following formula:
Wherein x represents real-time operation data to be predicted, F (x) represents the prediction probability of target fault prediction results of the real-time operation data, B represents the number of candidate fault prediction results, F b (x) The prediction probability of the b-th candidate fault prediction result is represented.
The candidate fault prediction results output by the fault prediction model are ordered according to the prediction probability of the candidate fault prediction results, so that the target fault prediction result can be rapidly determined from the candidate fault prediction results, and the determination efficiency of the target fault prediction result is improved.
In an embodiment of the present invention, optionally, after determining the target fault prediction result, the method further includes: judging the fault degree of the target fault prediction result; and (5) formulating a corresponding emergency scheme according to the fault degree of the target fault prediction result and uploading the emergency scheme. Therefore, operation and maintenance personnel can quickly formulate an optimal maintenance strategy of the capacitor according to a target prediction result of the capacitor real-time operation data, the fault degree of the target fault prediction result and a corresponding emergency scheme, and further expansion of accidents is avoided.
It should be noted that the training process and the using process of the fault prediction model in the present invention may be performed on the same electronic device, or may be performed on different electronic devices, and those skilled in the art may select the training process and the using process according to needs, which is not limited in the embodiments of the present invention.
According to the technical scheme, the historical operation data of the capacitor in the transformer substation in the historical fault process are analyzed, the fault state quantity affecting the capacitor is determined, the integrated classifier of the attribute importance is adopted, the fault prediction model is constructed according to the historical operation data of the fault state quantity, the structural redundancy of the fault prediction model is reduced, and the training efficiency of the fault prediction model on the data is improved; and inputting real-time operation data of the capacitor related to the fault state quantity in the real-time working process into a fault prediction model, and predicting the operation state of the capacitor in advance, so that the power failure accident caused by the capacitor fault is reduced, and the utilization rate of the capacitor is improved.
Example two
Fig. 2 is a flowchart of a capacitor fault prediction method provided in a second embodiment of the present invention, where the embodiment is further refined based on the foregoing embodiment, and provides a specific step of training an initial fault prediction model according to a training sample to obtain a fault prediction model by using an integrated classifier with attribute importance. As shown in fig. 2, the method includes:
s210, generating at least two sample subsets according to training samples.
The data range of the sample data in the sample subset is smaller than that of the sample data in the training sample, and the data ranges of the sample data in each sample subset are different from each other.
Specifically, at least two sample subsets may be partitioned according to a data range of sample data in the training samples.
Optionally, the process of generating at least two sample subsets from training samples includes: and setting corresponding categories for the sample data included in the sample subset respectively. For example, the class to which the sample data corresponds may refer to an operational state in which the capacitor operates under the operational condition of the sample data, e.g., the class may be set to normal or a specific type of fault.
S220, determining the attribute importance of at least one candidate attribute in the sample subset.
The candidate attribute may refer to an attribute feature that can distinguish sample data in a subset of samples. For example, the candidate attribute may be represented as a data range of the fault state quantity greater than X and less than Y, where X and Y may be used to represent two possible values of the fault state quantity, and where X is less than Y.
The attribute importance may refer to a comparison result of importance of the information gain contribution among all candidate attributes.
Specifically, at least one candidate attribute and an attribute importance of the candidate attribute are determined according to sample data in the sample subset.
Optionally, determining the attribute importance of at least one candidate attribute in the sample subset includes: determining at least one candidate attribute contained in the sample subset, and calculating an attribute importance of each candidate attribute, which may be determined by the following formula:
wherein (D, a) represents the attribute importance of attribute a, d= { (x) 1 ，y 1 )，(x 2 ，y 2 )…，(x i ，y i ) The sample subset is denoted (x) i ，y i ) Representing the ith sample data and corresponding class in the sample subset, I (D) being the entropy of information, D v The V-th of the V possible values representing the candidate attribute a and contained in D, G (, a) represents the number of sample data belonging to the candidate attribute a in the sample subset.
S230, selecting a target attribute from the candidate attributes according to the attribute importance, and constructing a fuzzy decision tree by combining the target attribute serving as a split node with the sample subset by adopting an integrated classifier.
The target attribute may refer to an attribute determined from candidate attributes.
Specifically, based on a fuzzy random forest algorithm, a candidate attribute with the largest attribute importance degree is selected from at least one candidate attribute of a sample subset according to the attribute importance degree of the candidate attribute to serve as a target attribute, the target attribute is taken as a splitting node, sample data conforming to the target attribute and sample data not conforming to the target attribute are partitioned from the sample subset according to the target attribute, the process is repeated until the last candidate attribute is selected as the splitting node, and a fuzzy decision tree corresponding to the sample subset is constructed.
S240, determining a fault prediction model according to the fuzzy decision tree corresponding to each sample subset.
Alternatively, the number of sample subsets may represent the number of iterations in the failure prediction model process. Specifically, a fault prediction model is determined according to a fuzzy decision tree constructed by each sample subset
Optionally, before selecting the target attribute from the candidate attributes according to the attribute importance degree, the method further includes: determining average attribute importance according to the attribute importance; and filtering out candidate attributes with attribute importance less than the average attribute importance.
Specifically, the average attribute importance of the candidate attributes can be determined according to the attribute importance of each candidate attribute, the candidate attributes with the attribute importance smaller than the average attribute importance are filtered according to the average attribute importance of the candidate attributes, and the candidate attributes with the attribute importance greater than the average attribute importance are reserved.
Through screening candidate attributes, the candidate attributes with high attribute importance are reserved, redundancy of attributes with low importance to the fault prediction model can be effectively reduced, the structure of the fault prediction model can be optimized, the training efficiency and the working efficiency of the fault prediction model are improved, and meanwhile, the memory occupation in the working process of the fault prediction model is reduced.
Optionally, in an embodiment of the present invention, the method further includes: determining a sample subset corresponding to the fuzzy decision tree in a training sample and the number of sample data belonging to the candidate category in the sample subset aiming at the candidate category corresponding to each leaf node of the fuzzy decision tree in the fault prediction model, and determining the candidate category characteristics of the fuzzy decision tree according to the number; and determining the maximum fuzzy probability of each candidate category in the fuzzy decision tree according to the candidate category characteristics and the weight of each candidate category in the fuzzy decision tree, and taking the maximum fuzzy probability as the prediction probability of the candidate category in the fault prediction model. Alternatively, the above-described process may follow step S230 or step S240, and is not particularly limited herein.
The candidate category corresponding to the leaf node may refer to a category corresponding to the sample data meeting the same attribute condition. The sample subset is split continuously according to the target attribute represented by the splitting node of the fuzzy decision tree, sample data with the same attribute in the sample subset are divided together, and the category corresponding to the sample data with the same attribute is used as the candidate category corresponding to the leaf node.
Wherein, candidate category characteristics can be used for representing category distribution conditions of the fuzzy decision tree, and optionally, the candidate category characteristics can be represented by the following formula:
wherein K represents candidate category characteristics of the fuzzy decision tree, |z i The I represents that sample data in a sample subset corresponding to the fuzzy decision tree belongs to a candidate category z i I represents the number of candidate categories divided by the fuzzy decision tree. Alternatively, the candidate class features may be represented in a matrix form.
Specifically, according to the candidate category corresponding to the leaf node of the fuzzy decision tree in the fault prediction model and the number of sample data belonging to the candidate category in the sample subset corresponding to the fuzzy decision tree, determining the category characteristics of the fuzzy decision tree.
The maximum ambiguity probability can be used to represent the coincidence degree of the candidate category determined in the fuzzy decision tree and the corresponding operation condition of the real-time operation data. Alternatively, the maximum ambiguity probability can be calculated by the following formula:
P＝argmax{w(y i )·K}。
wherein P represents the maximum ambiguity probability, w (y i ) And a weight function representing candidate category, wherein K represents candidate category characteristics of the fuzzy decision tree. Alternatively, the weighting function may be set by those skilled in the art according to the actual situation.
Specifically, according to the candidate category characteristics of the fuzzy decision tree and the weights set for the candidate categories corresponding to the leaf nodes in the fuzzy decision tree, the maximum fuzzy probability of the candidate categories corresponding to the leaf nodes in the fuzzy decision tree is determined and used as the prediction probability of the candidate categories in the fault prediction model.
According to the embodiment of the invention, the candidate attribute with the largest attribute importance is selected as the splitting node to construct the fuzzy decision tree according to the attribute importance of at least one candidate attribute in the sample subset, so that the condition that the attribute with more attribute values is selected as the splitting node to construct the fuzzy decision tree by the traditional random forest algorithm is avoided, the influence of subjective factors is reduced, meanwhile, the influence of the irrelevant attribute of the attribute importance on the fault prediction model is avoided by combining the attribute importance, the processing amount of data in the training process of the fault prediction model is reduced, and the redundancy of the fault prediction model caused by taking the irrelevant attribute as the splitting node is reduced.
Example III
Fig. 3 is a schematic structural diagram of a capacitor fault prediction device according to a third embodiment of the present invention. As shown in fig. 3, the apparatus includes:
a real-time data acquisition module 310, configured to acquire real-time operation data of a fault state quantity of a capacitor in a substation during a real-time working process; the fault state quantity is a state quantity which is in an abnormal state in the fault process of the capacitor.
A fault prediction module 320, configured to determine a fault prediction result of the capacitor according to the real-time operation data based on a prediction probability of the candidate class in the fault prediction model; the fault prediction model is obtained through training in the following mode: determining a fault state quantity and historical operation data of the fault state quantity in the historical operation data of the capacitor as training samples; and training the initial fault prediction model according to the training sample by adopting an integrated classifier of attribute importance to obtain the fault prediction model.
According to the technical scheme, the historical operation data of the capacitor in the transformer substation in the historical fault process are analyzed, the fault state quantity affecting the capacitor is determined, the integrated classifier of the attribute importance is adopted, the fault prediction model is constructed according to the historical operation data of the fault state quantity, the structural redundancy of the fault prediction model is reduced, and the training efficiency of the fault prediction model on the data is improved; and inputting real-time operation data of the capacitor related to the fault state quantity in the real-time working process into a fault prediction model, and predicting the operation state of the capacitor in advance, so that the power failure accident caused by the capacitor fault is reduced, and the utilization rate of the capacitor is improved.
Optionally, determining the fault state quantity and the historical operation data of the fault state quantity according to the historical operation data of the capacitor as training samples includes:
analyzing the operation data of the state quantity of the capacitor in the historical failure process, and taking the state quantity of the operation data in the historical failure process as the failure state quantity affecting the capacitor failure according to the analysis result;
the method comprises the steps of extracting normal operation data of a fault state quantity of a capacitor in a normal working process from historical operation data of the capacitor to serve as a positive training sample, and extracting abnormal operation data of the fault state quantity of the capacitor in the fault process from the historical operation data of the capacitor to serve as a negative training sample.
Optionally, the fault state quantity includes at least one of: the capacitance value of the capacitor, the dielectric loss value of the capacitor, the harmonic current of the capacitor and the operating voltage of the capacitor.
Optionally, an integrated classifier of attribute importance is adopted, and an initial fault prediction model is trained according to a training sample to obtain the fault prediction model, which comprises the following steps:
generating at least two sample subsets from the training samples;
determining the attribute importance of at least one candidate attribute in the sample subset;
Selecting a target attribute from candidate attributes according to the attribute importance, and constructing a fuzzy decision tree by combining the target attribute serving as a splitting node with the sample subset by adopting the integrated classifier;
and determining a fault prediction model according to the fuzzy decision tree corresponding to each sample subset.
Optionally, determining the attribute importance of at least one candidate attribute in the sample subset includes:
determining at least one candidate attribute contained in the sample subset, and calculating attribute importance of each candidate attribute:
where G (D, a) represents the attribute importance of attribute a in the sample subset, d= { (x) 1 ，y 1 )，(x 2 ，y 2 )…，(x i ，y i ) The sample subset is denoted (x) i ，y i ) Representing the ith sample data and corresponding class in the sample subset, I (D) being the entropy of information, D v The V-th of the V possible values representing the candidate attribute a and contained in D, G (a) represents the number of sample data belonging to the candidate attribute a in the sample subset.
Optionally, before selecting the target attribute from the candidate attributes according to the attribute importance degree, the method further includes:
determining average attribute importance according to the attribute importance;
and filtering out candidate attributes with attribute importance less than the average attribute importance.
Optionally, the training process of the fault prediction model further includes:
Determining a sample subset corresponding to the fuzzy decision tree in a training sample and the number of sample data belonging to the candidate category in the sample subset aiming at the candidate category corresponding to each leaf node of the fuzzy decision tree in the fault prediction model, and determining the candidate category characteristics of the fuzzy decision tree according to the number;
and determining the maximum fuzzy probability of each candidate category in the fuzzy decision tree according to the candidate category characteristics and the weight of each candidate category in the fuzzy decision tree, and taking the maximum fuzzy probability as the prediction probability of the candidate category in the fault prediction model.
Alternatively, the fault prediction module 320 may be specifically configured to:
based on the fault prediction model, determining at least two corresponding candidate categories for real-time operation data, taking the corresponding candidate categories as candidate fault prediction results, and taking the prediction probability of the corresponding candidate categories as the prediction probability of the fault prediction results;
sequencing at least two candidate prediction results according to the prediction probability of the candidate fault prediction results;
and determining a target fault prediction result according to the sequencing result.
The capacitor fault prediction device provided by the embodiment of the invention can execute the capacitor fault prediction method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example IV
Fig. 4 shows a schematic diagram of the structure of an electronic device 10 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 4, the electronic device 10 includes at least one processor 11, and a memory, such as a Read Only Memory (ROM) 12, a Random Access Memory (RAM) 13, etc., communicatively connected to the at least one processor 11, in which the memory stores a computer program executable by the at least one processor, and the processor 11 may perform various appropriate actions and processes according to the computer program stored in the Read Only Memory (ROM) 12 or the computer program loaded from the storage unit 18 into the Random Access Memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 may also be stored. The processor 11, the ROM 12 and the RAM 13 are connected to each other via a bus 14. An input/output (I/O) interface 15 is also connected to bus 14.
Various components in the electronic device 10 are connected to the I/O interface 15, including: an input unit 16 such as a keyboard, a mouse, etc.; an output unit 17 such as various types of displays, speakers, and the like; a storage unit 18 such as a magnetic disk, an optical disk, or the like; and a communication unit 19 such as a network card, modem, wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The processor 11 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as the capacitor failure prediction method.
In some embodiments, the capacitor fault prediction method may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 10 via the ROM 12 and/or the communication unit 19. One or more of the steps of method XXX described above may be performed when the computer program is loaded into RAM 13 and executed by processor 11. Alternatively, in other embodiments, the processor 11 may be configured to perform the capacitor failure prediction method in any other suitable way (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.

Claims (11)

1. A method for predicting capacitor failure, comprising:
acquiring real-time operation data of fault state quantity of a capacitor in a transformer substation in a real-time working process; the fault state quantity is an abnormal state quantity in the fault process of the capacitor;
determining a fault prediction result of the capacitor according to the real-time operation data based on the prediction probability of the candidate class in the fault prediction model;
The fault prediction model is obtained through training in the following mode: determining the fault state quantity and the historical operation data of the fault state quantity from the historical operation data of the capacitor as training samples; and training an initial fault prediction model according to the training sample by adopting an integrated classifier of attribute importance to obtain the fault prediction model.
2. The method of claim 1, wherein determining the fault state quantity and the historical operating data of the fault state quantity from the historical operating data of the capacitor as training samples comprises:
analyzing the operation data of the state quantity of the capacitor in the historical failure process, and taking the state quantity of the operation data in the historical failure process as the failure state quantity affecting the capacitor failure according to the analysis result;
extracting normal operation data of the fault state quantity of the capacitor in the normal working process from the historical operation data of the capacitor to serve as a positive training sample, and extracting abnormal operation data of the fault state quantity of the capacitor in the fault process from the historical operation data of the capacitor to serve as a negative training sample.
3. The method of claim 2, the fault state quantity comprising at least one of: the capacitance value of the capacitor, the dielectric loss value of the capacitor, the harmonic current of the capacitor and the operating voltage of the capacitor.
4. The method of claim 1, wherein training the initial failure prediction model based on the training samples using the integrated classifier of attribute importance to obtain the failure prediction model comprises:
generating at least two sample subsets from the training samples;
determining attribute importance of at least one candidate attribute in the sample subset;
selecting a target attribute from the candidate attributes according to the attribute importance, and constructing a fuzzy decision tree by combining the target attribute serving as a splitting node with the sample subset by adopting the integrated classifier;
and determining the fault prediction model according to the fuzzy decision tree corresponding to each sample subset.
5. The method of claim 4, wherein determining the attribute importance of at least one candidate attribute in the sample subset comprises:
determining at least one candidate attribute contained in the sample subset, and calculating attribute importance of each candidate attribute:
Wherein G (D, a) represents the attribute importance of attribute a, d= { (x) 1 ，y 1 )，(x 2 ，y 2 )…,(x i ，y i ) The sample subset is denoted (x) i ，y i ) Representing the ith sample data and corresponding class in the sample subset, I (D) being the entropy of information, D v The V-th of the V possible values representing the candidate attribute a and contained in D, G (a) represents the number of sample data belonging to the candidate attribute a in the sample subset.
6. The method of claim 4, further comprising, prior to selecting a target attribute from the candidate attributes based on the attribute importance:
determining an average attribute importance according to the attribute importance;
and filtering out candidate attributes with the attribute importance less than the average attribute importance.
7. The method according to any one of claims 1-6, further comprising:
determining a sample subset corresponding to the fuzzy decision tree in the training sample and the number of sample data belonging to the candidate category in the sample subset aiming at the candidate category corresponding to each leaf node of the fuzzy decision tree in the fault prediction model, and determining the candidate category characteristics of the fuzzy decision tree according to the number;
and determining the maximum fuzzy probability of each candidate category in the fuzzy decision tree as the prediction probability of the candidate category in the fault prediction model according to the candidate category characteristics and the weight of each candidate category in the fuzzy decision tree.
8. The method of claim 1, wherein determining a fault prediction result for the capacitor from the real-time operational data based on a prediction probability of a candidate class in a fault prediction model comprises:
determining at least two corresponding candidate categories for the real-time operation data based on a fault prediction model, taking the corresponding candidate categories as candidate fault prediction results, and taking the prediction probability of the corresponding candidate categories as the prediction probability of the candidate fault prediction results;
sequencing at least two candidate prediction results according to the prediction probability of the candidate fault prediction results;
and determining a target fault prediction result according to the sequencing result.
9. A capacitor failure prediction apparatus, comprising:
the real-time data acquisition module is used for acquiring real-time operation data of fault state quantity of the capacitor in the transformer substation in the real-time working process; the fault state quantity is an abnormal state quantity in the fault process of the capacitor;
the fault prediction module is used for determining a fault prediction result of the capacitor according to the real-time operation data based on the prediction probability of the candidate category in the fault prediction model;
The fault prediction model is obtained through training in the following mode: determining the fault state quantity and the historical operation data of the fault state quantity from the historical operation data of the capacitor as training samples; and training an initial fault prediction model according to the training sample by adopting an integrated classifier of attribute importance to obtain the fault prediction model.
10. An electronic device, the electronic device comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the capacitor fault prediction method of any one of claims 1-8.
11. A computer readable storage medium storing computer instructions for causing a processor to implement the capacitor failure prediction method of any one of claims 1-8 when executed.
CN202310440205.0A 2023-04-21 2023-04-21 Capacitor fault prediction method, device, equipment and medium Pending CN116467655A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310440205.0A CN116467655A (en) 2023-04-21 2023-04-21 Capacitor fault prediction method, device, equipment and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310440205.0A CN116467655A (en) 2023-04-21 2023-04-21 Capacitor fault prediction method, device, equipment and medium

Publications (1)

Publication Number Publication Date
CN116467655A true CN116467655A (en) 2023-07-21

Family

ID=87178579

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310440205.0A Pending CN116467655A (en) 2023-04-21 2023-04-21 Capacitor fault prediction method, device, equipment and medium

Country Status (1)

Country Link
CN (1) CN116467655A (en)

Similar Documents

Publication Publication Date Title
CN114841619A (en) State evaluation method and device for isolating switch, electronic equipment and medium
CN116962272B (en) Abnormality detection method, device, equipment and storage medium for network index
CN116957539A (en) Cable state evaluation method, device, electronic equipment and storage medium
CN117455067A (en) Electric quantity consumption prediction method and device, electronic equipment and storage medium
CN116755974A (en) Cloud computing platform operation and maintenance method and device, electronic equipment and storage medium
CN116703109A (en) Method, device, equipment and storage medium for selecting power distribution network project
CN116467655A (en) Capacitor fault prediction method, device, equipment and medium
CN116127326B (en) Composite insulator detection method and device, electronic equipment and storage medium
CN115373449B (en) Data processing method, device, equipment and storage medium
CN117974363A (en) Distribution transformer running state evaluation method, device, equipment and medium
CN116822740A (en) Power distribution network operation and maintenance scheme determining method and device, electronic equipment and storage medium
CN115129538A (en) Event processing method, device, equipment and medium
CN116840696A (en) Method and device for predicting remaining life of storage battery, electronic equipment and storage medium
CN117907752A (en) Fault early warning method and device for high-voltage transmission line, electronic equipment and storage medium
CN118095952A (en) New energy grid-connected power quality assessment method, device, equipment and medium
CN116186536A (en) Risk prediction method, risk prediction device, electronic equipment and storage medium
CN117493060A (en) Database component anomaly detection method, device, equipment and medium
CN116308284A (en) Operation data detection method, device and equipment of pumped storage equipment
CN118657252A (en) Training method, device, equipment and medium for lightning arrester fault prediction model
CN116402250A (en) Method and device for evaluating equipment state, electronic equipment and storage medium
CN116304635A (en) Operation data analysis method, device and equipment of pumped storage equipment
CN117035405A (en) Power system risk index assessment method and device, electronic equipment and medium
CN118540200A (en) Alarm compression noise reduction strategy arrangement method and device, electronic equipment and medium
CN117422108A (en) Method, device, equipment and storage medium for determining convolution kernel
CN116842837A (en) Transformer fault diagnosis method and device, electronic equipment and storage medium

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