Method and system for evaluating comprehensive state of zinc oxide arrester

Abstract

The invention belongs to the technical field of electric power engineering. The invention provides a zinc oxide arrester comprehensive state evaluation method and system, and aims at any fault, different types of index data sets, including a historical data set, a current data set and a future data set, corresponding to the fault are obtained, association rule mining is carried out, and multiple types of index data sets with the highest association degree with the fault are obtained; scoring the comprehensive state information of the zinc oxide arrester according to the multiple types of index data sets, and respectively obtaining scores corresponding to the historical data set, the current data set and the future data set; and according to the scores corresponding to the obtained historical data set, the current data set and the future data set, in combination with a pre-trained Bayesian network model, obtaining a conditional probability corresponding to each comprehensive state level of the zinc oxide arrester. According to the invention, the state of the zinc oxide arrester can be evaluated more accurately.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power engineering, and in particular to a method and system for evaluating the comprehensive state of a zinc oxide lightning arrester. Background Art

[0002] As a key component of the power system, the safe operation of the distribution network has a significant impact on the entire system. Due to their low insulation levels and the general lack of lightning protection lines, distribution network lines are particularly susceptible to lightning strikes, making them vulnerable to damage, affecting users' normal power supply and personal safety. Therefore, lightning protection for distribution networks is of paramount importance.

[0003] Currently, installing lightning arresters is a key method of lightning protection for distribution networks. They effectively limit the impact of transient high voltages such as lightning overvoltage and switching overvoltage on power system equipment, significantly improve the lightning resistance of distribution network lines, and effectively reduce the probability of lightning tripping. Among them, zinc oxide lightning arresters, due to their excellent nonlinear resistance characteristics, fast response speed, high discharge and lightning resistance capabilities, have become the most widely used lightning arresters in distribution networks. However, in actual power system operation, they are exposed to air for long periods of time and are affected by factors such as changes in atmospheric temperature and humidity, as well as overvoltages caused by various reasons. This can cause the lightning arrester structure to age and degrade, potentially losing its protective effect on power equipment and even causing explosions, posing safety hazards. Therefore, to prevent operational accidents, it is necessary to comprehensively consider various equipment status information to accurately assess the operating status of zinc oxide lightning arresters.

[0004] There are some problems with the existing zinc oxide arrester status assessment algorithms, which limit the accuracy and reliability of the assessment, including: (1) some assessment algorithms may only be applicable to specific types of zinc oxide arresters or specific working environments, resulting in limitations in the practical application of the algorithms; (2) some assessment algorithms may involve complex mathematical calculations or model construction, which increases the complexity and computational cost of the algorithms and is not conducive to real-time online monitoring and rapid assessment; (3) when using certain assessment models, accurate model parameters need to be input. However, due to the influence of various factors, the acquisition of these parameters may be difficult or erroneous, thus affecting the accuracy and reliability of the models. Summary of the Invention

[0005] In order to address the shortcomings of the existing technology, the present invention provides a comprehensive state assessment method and system for zinc oxide lightning arresters. First, association rules are used to mine the correlation of state parameters under various fault modes. Then, a five-level state assessment method is adopted to establish a hierarchical model for quantifying the health state of the lightning arrester. The historical, current and predicted state information of the lightning arrester is evaluated. The data characteristics of these three types of information are comprehensively considered to perform conditional probability table self-learning, and a zinc oxide state assessment model based on a Bayesian network is established to achieve a more accurate assessment of the zinc oxide lightning arrester state.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a method for evaluating the comprehensive state of a zinc oxide lightning arrester, comprising:

[0008] For any fault, different types of indicator datasets corresponding to the fault are obtained. The indicator datasets include historical indicator datasets, current indicator datasets, and future indicator datasets. Association rule mining is performed on the indicator datasets to obtain multiple types of indicator datasets with the highest degree of correlation with the fault. The future indicator datasets are obtained based on Bayesian prediction.

[0009] Scoring the comprehensive status information of the zinc oxide lightning arrester based on multiple types of indicator data sets with the highest degree of correlation with the fault, and obtaining scores for each comprehensive status information of the zinc oxide lightning arrester corresponding to the historical indicator data set, the current indicator data set, and the future indicator data set; the comprehensive status information of the zinc oxide lightning arrester includes multiple comprehensive status levels;

[0010] Based on the scoring, the number of samples of historical, current and future comprehensive state combinations is counted through the Bayesian learning algorithm, and the conditional probability corresponding to each comprehensive state level of the zinc oxide lightning arrester is calculated.

[0011] As a further limitation of the first aspect of the present invention, the faults include: aging faults, contamination faults, moisture faults and other faults;

[0012] The indicator data set corresponding to the aging fault includes: action count increment data, action count data, operation years data, temperature data, lightning data, leakage current data, resistive leakage current data and infrared temperature data;

[0013] The indicator data set corresponding to the contamination fault includes: haze condition data, leakage current data, resistive leakage current data, infrared temperature data, manufacturer data and jacket type data;

[0014] The indicator data set corresponding to the moisture fault includes: air humidity data, rainfall data, leakage current data, resistive leakage current data, infrared temperature data and jacket type data;

[0015] The indicator data sets corresponding to the other faults include: leakage current data, resistive leakage current data, infrared temperature data and appearance data.

[0016] As a further limitation of the first aspect of the present invention, the comprehensive status information of the zinc oxide lightning arrester includes five levels: normal, caution, minor, abnormal, and emergency.

[0017] As a further limitation of the first aspect of the present invention, the scoring of the zinc oxide lightning arrester comprehensive status information based on multiple types of indicator data sets with the highest degree of correlation with the fault includes:

[0018] For indicator data with a positive correlation between the score and the indicator data value, the score is calculated using the ascending half-step model; for indicator data with an inverse correlation between the score and the indicator data value, the score is calculated using the descending half-step model;

[0019] When the scores of multiple types of indicators in a certain comprehensive state are not less than 0.7, the arithmetic mean of the scores of multiple types of indicators is taken as the final score of the comprehensive state; otherwise, the minimum value of the scores of multiple types of indicators is taken as the final score of the comprehensive state;

[0020] The scores of zinc oxide lightning arresters under various comprehensive conditions are calculated for the historical indicator data set, current indicator data set, and future indicator data set.

[0021] As a further limitation of the first aspect of the present invention, for indicators whose indicator data values are equal to the model threshold, the score is calculated by a fuzzy mathematical membership function; the model refers to the half-step-up model and the half-step-down model.

[0022] As a further limitation of the first aspect of the present invention, the method of counting the number of samples of historical, current and future comprehensive state combinations through a Bayesian learning algorithm based on the score, and calculating the conditional probability corresponding to each comprehensive state level of the zinc oxide lightning arrester, includes:

[0023] ,

[0024] in, For the parent node Take the first combination, at the same time The conditional probability of the combined state is For the parent node For the The confidence level obtained by mining association rules in a comprehensive state is For the parent node Take the first combination, at the same time The number of samples in the comprehensive state, for The parent node includes three nodes: history, current and future; the comprehensive status includes five levels: normal, attention, minor, abnormal and emergency; the combination is a comprehensive status combination of three nodes.

[0025] In a second aspect, the present invention provides a zinc oxide arrester comprehensive status assessment system, comprising:

[0026] An association mining unit is configured to: for any fault, obtain different types of indicator data sets corresponding to the fault, the indicator data sets including historical indicator data sets, current indicator data sets, and future indicator data sets, perform association rule mining on the indicator data sets to obtain multiple types of indicator data sets with the highest degree of correlation with the fault; the future indicator data sets are obtained based on Bayesian prediction;

[0027] a scoring unit configured to score the comprehensive status information of the zinc oxide lightning arrester based on multiple types of indicator data sets with the highest degree of correlation with the fault, and obtain scores for each comprehensive status information of the zinc oxide lightning arrester corresponding to the historical indicator data set, the current indicator data set, and the future indicator data set; the comprehensive status information of the zinc oxide lightning arrester includes multiple comprehensive status levels;

[0028] The comprehensive evaluation unit is configured to: count the number of samples of historical, current and future comprehensive state combinations based on the scores through a Bayesian learning algorithm, and calculate the conditional probability corresponding to each comprehensive state level of the zinc oxide lightning arrester.

[0029] In a third aspect, the present invention provides a computer device comprising: a processor and a computer-readable storage medium;

[0030] a processor adapted to execute a computer program;

[0031] A computer-readable storage medium having a computer program stored therein, wherein when the computer program is executed by the processor, the method for evaluating the comprehensive state of a zinc oxide lightning arrester according to the first aspect of the present invention is implemented.

[0032] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, wherein the computer program is suitable for being loaded by a processor and executing the method for comprehensive status assessment of zinc oxide lightning arresters as described in the first aspect of the present invention.

[0033] In a fifth aspect, the present invention provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the comprehensive status assessment method of the zinc oxide lightning arrester as described in the first aspect of the present invention.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] The present invention proposes a comprehensive state assessment method for zinc oxide lightning arresters. First, association rules are used to mine the correlation of state parameters under various fault modes. Then, a five-level state assessment method is adopted to establish a hierarchical model for quantifying the health state of the lightning arrester. The historical, current and predicted state information of the lightning arrester is evaluated. The data characteristics of these three types of information are comprehensively considered to perform conditional probability table self-learning, and a zinc oxide state assessment model based on a Bayesian network is established, which realizes a more accurate assessment of the zinc oxide lightning arrester state. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A schematic flow chart of a method for evaluating the comprehensive state of a zinc oxide lightning arrester provided in Example 1 of the present invention;

[0037] Figure 2 This is a fault item frequency diagram provided in Example 2 of the present invention;

[0038] Figure 3 The transaction matrix heat map provided in Example 2 of the present invention;

[0039] Figure 4 A schematic diagram of a comprehensive status assessment system for zinc oxide lightning arresters provided in Example 3 of the present invention;

[0040] Figure 5 A schematic diagram of a computer device provided in Example 4 of the present invention. DETAILED DESCRIPTION

[0041] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0042] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0043] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0044] Example 1

[0045] This embodiment 1 proposes a comprehensive state evaluation method for zinc oxide lightning arresters, see Figure 1 , including the following processes:

[0046] For any fault, different types of indicator data sets corresponding to the fault are obtained, where the indicator data sets include historical indicator data sets, current indicator data sets, and future indicator data sets. Association rule mining is performed on the indicator data sets to obtain multiple types of indicator data sets with the highest degree of correlation with the current fault;

[0047] Based on multiple types of indicator data sets with the highest degree of correlation with the current fault, the comprehensive status information of the zinc oxide lightning arrester is scored, and the scores of the comprehensive status information of the zinc oxide lightning arrester corresponding to the historical indicator data set, the current indicator data set, and the future indicator data set are obtained respectively. Based on the scores, the number of samples of each comprehensive status of the zinc oxide lightning arrester corresponding to the historical indicator data set, the current indicator data set, and the future indicator data set is determined. The comprehensive status includes five levels: normal, caution, minor, abnormal, and emergency;

[0048] Based on the number of samples of each comprehensive state of zinc oxide arrester corresponding to the historical indicator data set, current indicator data set and future indicator data set, combined with the pre-trained Bayesian network model, the conditional probability corresponding to each comprehensive state level of zinc oxide arrester is obtained.

[0049] It should be noted that the future data set is obtained based on Bayesian prediction.

[0050] Specifically, association rule mining is used to find meaningful associations in large data sets, and the discovered associations can be given in the form of association rules or frequent item sets. In this embodiment, the Apriori algorithm in the association rule mining algorithm is used to quantify the association knowledge between specific indicators and the occurrence of faults, realize the combination of the influence of multiple variables and parameters, and initialize the prior probability through the correlation of variables. The Apriori algorithm uses binary representation, that is, the data is encoded as 0 or 1, each row corresponds to the test data of each indicator obtained from a test, and each column corresponds to the test data obtained from each test of a certain indicator. If the data exceeds the corresponding threshold, the value is 1, indicating that the arrester has failed for this indicator; otherwise, it is 0.

[0051] In order to describe the fault types of zinc oxide lightning arresters in detail, this embodiment divides them into four types of faults, as shown in Table 1.

[0052] Table 1 Fault classification

[0053]

[0054] Specifically, the indicator data sets corresponding to aging faults include: action count increment data, action count data, operation years data, temperature data, lightning data, leakage current data, resistive leakage current data, and infrared temperature data;

[0055] The indicator data sets corresponding to contamination faults include: haze data, leakage current data, resistive leakage current data, infrared temperature data, manufacturer data, and jacket type data;

[0056] The indicator data sets corresponding to moisture faults include: air humidity data, rainfall data, leakage current data, resistive leakage current data, infrared temperature data, and jacket type data;

[0057] Other fault indicator data sets include: leakage current data, resistive leakage current data, infrared temperature data, and appearance data.

[0058] For the historical status information of the equipment, the Apriori algorithm is first applied to set the total fault set to be mined as . Contains all historical status information of multiple measurements, including Transactions , which can be expressed as:

[0059] ;

[0060] in, is the fault item set containing various indicators each time a device fails. , if it contains Item is called Itemset, this The item is the reason for each failure, that is, the corresponding evaluation indicators, which can be expressed as:

[0061] ;

[0062] For a certain type of failure that occurs once, the corresponding fault item set ,support for:

[0063] ;

[0064] Define an occurrence different from The fault item set when the fault occurs is , which is defined by the association rule as → The expression of ∩ = ø. Then for the fault association rule, its support is:

[0065] ;

[0066] The support in formula (4) reflects the two fault item sets and The probability of simultaneous occurrence has the same support as that of frequent itemsets. Similarly, the confidence of the fault association rule is:

[0067] ;

[0068] The confidence described in formula (5) reflects the confidence that the fault item set contains Also includes Therefore, association rules with high support and confidence are mainly mined by defining the values of min_support and min_confidence. Association rule discovery is to discover association rules for a given , find those rules that satisfy s ≥ min_support_limit and c ≥ min_confidence_limit, where min_support_limit and min_confidence_limit are the thresholds for support and confidence.

[0069] The status information of electrical equipment is mainly divided into historical status information, current status information and predicted status information. These three types of status information jointly determine the comprehensive status of the equipment.

[0070] In the past, electrical equipment status was classified into only two levels: qualified or unqualified. Although this is intuitive and easy to understand, it does not meet the requirements of multi-level status classification. In this embodiment, a single yes / no system is no longer used. The overall status of the lightning arrester equipment is divided into five levels: normal, caution, minor, abnormal, and emergency. These levels are denoted as F = {F1, F2, F3, F4, F5}. The specific descriptions are shown in Table 2.

[0071] Table 2 Equipment operation and maintenance requirements at different levels

[0072]

[0073] In this embodiment, a hierarchical evaluation model for the comprehensive state of a lightning arrester is established as shown in Table 3, and the comprehensive state information scores of the zinc oxide lightning arrester corresponding to the historical indicator data set, the current indicator data set, and the future indicator data set are calculated based on the evaluation model and various indicator data.

[0074] The diagram shows a parent layer and its corresponding parallel child layers. The characteristic value is the child layer, and the fault type is the parent layer. The comprehensive status score of each parent layer is calculated by combining the scores of the corresponding child layers. If the child layer scores are all at least 0.7, the arithmetic mean of the child layers is used as the comprehensive status score of the parent layer. Otherwise, the minimum value of the child layers is used as the comprehensive status score of the parent layer. The comprehensive status is divided into five levels: normal, caution, minor, abnormal, and emergency.

[0075] Table 3. Status assessment model and characteristic quantity selection of zinc oxide arrester

[0076]

[0077] Table 3 above establishes a hierarchical comprehensive state assessment model for zinc oxide lightning arresters, which determines the operating status of the lightning arrester by analyzing recent data. In reality, the current state of the lightning arrester is often affected by its historical status, and recent state data often make it difficult to accurately obtain the comprehensive state of the lightning arrester. Therefore, in this embodiment, the historical, current, and future states are considered when performing the comprehensive state assessment of the lightning arrester, and the scores of the comprehensive state information of the zinc oxide lightning arrester corresponding to the historical data set, the current data set, and the future data set are respectively obtained.

[0078] Furthermore, the probability weight of the sample belonging to a certain comprehensive state (such as grade A) is determined according to the score value. During the training phase, the number of samples belonging to each comprehensive state level in the historical, current, and future states is counted.

[0079] In the embodiment of the present invention, the conditional probabilities corresponding to the various comprehensive status levels of the zinc oxide arrester are obtained in combination with the pre-trained Bayesian network model.

[0080] A Bayesian network is a network model used for reasoning and analyzing uncertainty problems. It is used to represent a graph pattern of connection probabilities between variables. It is currently the most effective theoretical model in the field of uncertain knowledge expression and reasoning. It can discover the relationship between a large number of variables and is a powerful tool for data prediction and classification. In this embodiment, future data sets are obtained based on Bayesian prediction, and the Bayesian network is applied to solve the uncertainty problem in the status assessment of zinc oxide lightning arresters.

[0081] The Bayesian network structure is a directed acyclic graph, which is part of the qualitative knowledge representation in the Bayesian network and is used to describe the dependency relationship between nodes. Each node in the Bayesian network The set of nodes formed by other descendant nodes that are conditionally independent of the parent node. The probability distribution is the conditional probability table associated with each node. The conditional probability table can be used It represents the conditional probability between a node and its parent node, as shown in formula (6).

[0082] ;

[0083] The modeling of Bayesian networks includes three elements: selecting appropriate sets of variables and important factors; the structure of the Bayesian network is determined on the basis of nodes, and determining the node conditional probability table.

[0084] By mining the fault state parameter association rules, the correlation between the fault state parameters is obtained and used as the prior initial probability of the Bayesian network model. Then, the Bayesian network conditional probability table is self-learned. The embodiment of the present invention uses the Bayesian statistical learning method to determine the conditional probability table of the Bayesian network model:

[0085] ;

[0086] In formula (7), is a random variable The first value in the parent node value combination indivual, for The number of values of . According to the probability normalization theory, It can be seen that according to the conditional expectation estimation, the learning formula of the conditional probability table in the Bayesian network model can be derived as follows:

[0087] ;

[0088] In formula (8), is the conditional probability value of association rule mining, that is, the confidence of each association rule mined. For the parent node Take the first combination, at the same time For the The number of samples in the comprehensive state, for Therefore, for the present invention, the parent node The combination refers to the comprehensive status combination of the three nodes. In this embodiment of the present invention, the arrester comprehensive status is divided into five levels: normal, caution, minor, abnormal, and emergency. Combinations, ,thus When it corresponds to state A in Table 2, It corresponds to state B in Table 2, and so on.

[0089] Formula (8) is the basic formula of conditional probability of Bayesian estimation, where In the initial concept, in order to avoid the problem that some conditional probabilities may be zero due to maximum likelihood estimation, Bayesian estimation is used to smooth the conditional probabilities. Therefore, a smoothing parameter (a very small constant, usually 1, also called Laplace smoothing) is introduced to avoid the situation where some conditional probabilities are zero. In this embodiment, in order to associate the Bayesian network with the above-mentioned association probability learning and construction, the Take the confidence value mined in the association rule mining as the conditional probability value. , Refers to the parent node corresponding to the state", and thus Refers to the parent node For example, if the parent node is "historical state", k is 1, which means state A. That is, when the historical state is A, the confidence that can be obtained by mining association rules using the historical data set is used. This value is used as Calculate the conditional probability. In addition, here is another explanation of association rules: Association rule mining is to analyze a known data set to mine the possible related indicator pairs and their confidence levels, regardless of the indicator order.

[0090] The calculation method is as follows: When the parent node combination is "historical state = A, current state = A, future state = B", the conditional probability when the comprehensive state is A, B, C, D, and E respectively is:

[0091] Assuming that the Bayesian learning algorithm calculates that the number of samples with the comprehensive states of A, B, C, D, and E under the full parent node combination is 38, 0, 0, 0, 0, the value of each conditional probability can be calculated:

[0092] p(A)=1+38 / (1+38)+(1+0)+(1+0)+(1+0)+(1+0)=0.906977;

[0093] p(B)=1+0 / (1+38)+(1+0)+(1+0)+(1+0)+(1+0)=0.023256;

[0094] Similarly, p(C), p(D), and p(E) can be calculated.

[0095] The operating status of the zinc oxide lightning arrester mainly integrates the above-mentioned historical status, current status, and future status information. These three types of information have different timeliness, and their information also differs when describing the impact of faults on the operating status of the equipment. The embodiment of the present invention uses a half-ladder model to score the zinc oxide lightning arrester status information, including an ascending half-ladder model and a descending half-ladder model. For indicators with larger values, the better (i.e., the dependent variable is proportional to the independent variable), the ascending half-ladder model is used, as shown in formula (9); and for indicators with smaller values, the better (i.e., the dependent variable is inversely proportional to the independent variable), the descending half-ladder model is used, as shown in formula (10).

[0096] ;

[0097] For the current status information, it can be assumed that the current fault status is linearly related to the evaluation of the equipment operating status and described by a step distribution.

[0098] ;

[0099] in 、 、 、 are model thresholds, is the value of the scoring parameter, i.e., the characteristic value. For the historical state dataset, this step distribution is used to obtain the historical score; for the current state dataset, the same step distribution is used to obtain the current score; and using the Bayesian algorithm for prediction, the future score can be obtained.

[0100] Taking the YH5WZ-17 / 45, 10kV distribution type lightning arrester as an example, the model threshold of the typical scoring items is taken as an example, see Table 4.

[0101] Table 4. Thresholds of zinc oxide arrester scoring model

[0102]

[0103] But considering the When the value approaches the threshold, it is difficult to accurately describe the specific state of the device at this time. Therefore, a fuzzy mathematical membership function is introduced here to soften the hard boundary conditions. The embodiment of the present invention uses a trapezoidal membership function, see Table 5. Finally, the state evaluation set is established:

[0104] .

[0105] Table 5 Membership function table

[0106]

[0107] The Bayesian network method proposed in this paper was tested and analyzed based on the state variables of a zinc oxide lightning arrester, demonstrating its feasibility and effectiveness. The state assessment results provide insights into the overall operating status of the device. When the device is in an abnormal state, the method can be used to diagnose and locate the fault, or to further troubleshoot any potential problems.

[0108] Example 2

[0109] This embodiment 2 takes a 10kV distribution type arrester model YH5WZ-17 / 45 as an example of an aging failure.

[0110] Taking 10,000 pieces of state data from a particular aging-related failure as the fault item set, we first perform association rule mining. We primarily focus on three items: leakage current, resistive leakage current, and relative infrared temperature. Due to the large variety of normal parameters, the support should not be too large. To ensure the accuracy of the rules, a high confidence level is required. Therefore, the initial minimum support and minimum confidence levels are set as follows:

[0111] ;

[0112] ;

[0113] In each iteration, if the generated rules fail to meet these thresholds, the algorithm will gradually lower the thresholds and try different support and confidence levels until the lowest limit is reached. For this fault item set, the adjustment is to:

[0114] ;

[0115] ;

[0116] The association rule table obtained by mining is shown in Table 6.

[0117] Table 6 Association rule mining under aging failure mode

[0118]

[0119] This embodiment obtains a visual output of the association rules, where Figure 2 The corresponding fault item frequency graph shows the frequency of each fault item in all transactions. Figure 3 The corresponding transaction matrix heat map shows whether each transaction contains a specific fault item. The darker the color, the more frequently the fault item occurs.

[0120] Conditional probability self-learning was performed. 10,000 pieces of status data of a 10kV distribution type arrester of model YH5WZ-17 / 45 were selected to evaluate the status of the device. According to the above scoring method, the status information of the zinc oxide arrester was scored as the training sample of the Bayesian network status evaluation model. According to the learning method based on the Bayesian network, the final conditional probability of the quantity model was calculated. Table 7 lists some conditional probability tables of the comprehensive status nodes of the zinc oxide arrester.

[0121] Table 7 Partial comprehensive state conditional probability table

[0122]

[0123] In summary, the embodiment of the present invention first uses association rules to mine the correlation of state parameters under various fault modes, and then adopts a five-level state evaluation method to establish a hierarchical model for quantifying the health status of the lightning arrester, evaluates the historical, current and predicted state information of the lightning arrester, and comprehensively considers the data characteristics of these three types of information to perform conditional probability table self-learning, and finally establishes a zinc oxide state evaluation model based on a Bayesian network, and provides a method for evaluating the equipment operation status; finally, the present invention uses the actual operation data of the equipment based on the above method for case application and verification, and the effect is good.

[0124] Example 3

[0125] Based on the same inventive concept, this embodiment 3 provides a comprehensive status evaluation system for zinc oxide lightning arresters, such as Figure 4 Shown, including:

[0126] An association mining unit is configured to: for any fault, obtain different types of indicator data sets corresponding to the fault, the indicator data sets including historical indicator data sets, current indicator data sets, and future indicator data sets, perform association rule mining on the indicator data sets to obtain multiple types of indicator data sets with the highest degree of correlation with the fault; the future indicator data sets are obtained based on Bayesian prediction;

[0127] a scoring unit configured to score the comprehensive status information of the zinc oxide lightning arrester based on multiple types of indicator data sets with the highest degree of correlation with the fault, and obtain scores for each comprehensive status information of the zinc oxide lightning arrester corresponding to the historical indicator data set, the current indicator data set, and the future indicator data set; the comprehensive status information of the zinc oxide lightning arrester includes multiple comprehensive status levels;

[0128] The comprehensive evaluation unit is configured to: count the number of samples of historical, current and future comprehensive state combinations based on the scores through a Bayesian learning algorithm, and calculate the conditional probability corresponding to each comprehensive state level of the zinc oxide lightning arrester.

[0129] It is understandable that each of the above-mentioned units can be separately or completely combined into one or several other units to constitute, or one (or some) of the units can be further divided into multiple functionally smaller units to constitute, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present application. The above-mentioned units are divided based on logical functions. In actual applications, the functions of one unit can also be implemented by multiple units, or the functions of multiple units can be implemented by one unit. In other embodiments of the present application, the system may also include other units. In actual applications, these functions can also be implemented with the assistance of other units, and can be implemented by the collaboration of multiple units.

[0130] According to another embodiment of the present application, the system described in this embodiment can be constructed and the method of Example 1 of the present application can be implemented by running a computer program (including program code) capable of executing the steps involved in the corresponding method described in Example 1 on a general-purpose computing device such as a computer, which includes processing elements and storage elements such as a central processing unit (CPU), random access memory (RAM), and read-only memory (ROM). The computer program can be recorded on, for example, a computer-readable recording medium, and loaded into the above-mentioned computing device through the computer-readable recording medium and run therein.

[0131] Example 4

[0132] This embodiment 4 provides an electronic device, such as Figure 5 As shown, the electronic device includes a processor 1001, a communication interface 1002, and a computer-readable storage medium 1003. The processor 1001, the communication interface 1002, and the computer-readable storage medium 1003 may be connected via a bus or other means.

[0133] Among them, the communication interface 1002 is used to receive and send data, the computer-readable storage medium 1003 can be stored in the memory of the electronic device, the computer-readable storage medium 1003 is used to store a computer program, the computer program includes program instructions, and the processor 1001 is used to execute the program instructions stored in the computer-readable storage medium 1003.

[0134] The processor 1001 (also called CPU (Central Processing Unit)) is the computing core and control core of the electronic device, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to implement corresponding method processes or corresponding functions.

[0135] The processor 1001 is configured to execute the following process:

[0136] For any fault, obtain different types of indicator data sets corresponding to the fault, including historical data sets, current data sets, and future data sets, and perform association rule mining on the indicator data sets to obtain multiple types of indicator data sets with the highest degree of correlation with the fault;

[0137] Based on multiple types of indicator data sets, the zinc oxide lightning arrester status information is scored, and the scores corresponding to the historical data set, current data set, and future data set are obtained respectively;

[0138] According to the scores corresponding to the historical data set, current data set and future data set, combined with the pre-trained Bayesian network model, the conditional probability corresponding to each comprehensive fault level of the zinc oxide lightning arrester is obtained.

[0139] Example 5

[0140] This embodiment 5 provides a computer-readable storage medium (Memory). This computer-readable storage medium is a memory device in an electronic device that is used to store programs and data. It is understood that the computer-readable storage medium herein can include both built-in storage media in the electronic device and, of course, extended storage media supported by the electronic device. The computer-readable storage medium provides storage space that stores the processing system of the electronic device.

[0141] Furthermore, the storage space also stores one or more instructions suitable for being loaded and executed by the processor. These instructions may be one or more computer programs (including program code). It should be noted that the computer-readable storage medium herein may be a high-speed RAM memory or a non-volatile memory, such as at least one disk storage device; alternatively, it may be at least one computer-readable storage medium located remotely from the processor.

[0142] In one embodiment, the computer-readable storage medium stores one or more instructions; the processor loads and executes the one or more instructions stored in the computer-readable storage medium to implement the following process:

[0143] For any fault, obtain different types of indicator data sets corresponding to the fault, including historical data sets, current data sets, and future data sets, and perform association rule mining on the indicator data sets to obtain multiple types of indicator data sets with the highest degree of correlation with the fault;

[0144] Based on multiple types of indicator data sets, the zinc oxide lightning arrester status information is scored, and the scores corresponding to the historical data set, current data set, and future data set are obtained respectively;

[0145] According to the scores corresponding to the historical data set, current data set and future data set, combined with the pre-trained Bayesian network model, the conditional probability corresponding to each comprehensive fault level of the zinc oxide lightning arrester is obtained.

[0146] Example 6

[0147] This embodiment 6 provides a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the following process:

[0148] For any fault, obtain different types of indicator data sets corresponding to the fault, including historical data sets, current data sets, and future data sets, and perform association rule mining on the indicator data sets to obtain multiple types of indicator data sets with the highest degree of correlation with the fault;

[0149] Based on multiple types of indicator data sets, the zinc oxide lightning arrester status information is scored, and the scores corresponding to the historical data set, current data set, and future data set are obtained respectively;

[0150] According to the scores corresponding to the historical data set, current data set and future data set, combined with the pre-trained Bayesian network model, the conditional probability corresponding to each comprehensive fault level of the zinc oxide lightning arrester is obtained.

[0151] Those skilled in the art will appreciate that the units and algorithmic steps of each example described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technical personnel may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0152] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions according to the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted via a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data processing device such as a server or data center that integrates one or more available media. Available media can be magnetic media (e.g., floppy disk, hard disk, tape), optical media (e.g., DVD), or semiconductor media (e.g., solid-state drive (SSD)).

[0153] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A method for comprehensive status assessment of zinc oxide lightning arrester, characterized in that: include: For any fault, different types of indicator datasets corresponding to the fault are obtained. The indicator datasets include historical indicator datasets, current indicator datasets, and future indicator datasets. Association rule mining is performed on the indicator datasets to obtain multiple types of indicator datasets with the highest degree of correlation with the fault. The future indicator datasets are obtained based on Bayesian prediction. Scoring the comprehensive status information of the zinc oxide lightning arrester based on multiple types of indicator data sets with the highest degree of correlation with the fault, and obtaining scores for each comprehensive status information of the zinc oxide lightning arrester corresponding to the historical indicator data set, the current indicator data set, and the future indicator data set; The zinc oxide arrester comprehensive status information includes multiple comprehensive status levels; Based on the scoring, the number of samples of historical, current and future comprehensive state combinations is counted through the Bayesian learning algorithm, and the conditional probability corresponding to each comprehensive state level of the zinc oxide lightning arrester is calculated.

2. A zinc oxide arrester comprehensive status assessment method according to claim 1, characterized in that: The said faults include: aging faults, contamination faults, moisture faults and other faults; The indicator data set corresponding to the aging fault includes: action count increment data, action count data, operation years data, temperature data, lightning data, leakage current data, resistive leakage current data and infrared temperature data; The indicator data set corresponding to the contamination fault includes: haze condition data, leakage current data, resistive leakage current data, infrared temperature data, manufacturer data and jacket type data; The indicator data set corresponding to the moisture fault includes: air humidity data, rainfall data, leakage current data, resistive leakage current data, infrared temperature data and jacket type data; The indicator data sets corresponding to the other faults include: leakage current data, resistive leakage current data, infrared temperature data and appearance data.

3. A zinc oxide arrester comprehensive status assessment method according to claim 1, characterized in that: The comprehensive status information of the zinc oxide lightning arrester includes five levels: normal, caution, minor, abnormal, and emergency.

4. A zinc oxide arrester comprehensive status assessment method according to claim 3, characterized in that: The scoring of the zinc oxide lightning arrester comprehensive status information based on multiple types of indicator data sets with the highest degree of correlation with the fault includes: For indicator data with a positive correlation between the score and the indicator data value, the score is calculated using the ascending half-step model; for indicator data with an inverse correlation between the score and the indicator data value, the score is calculated using the descending half-step model; When the scores of multiple types of indicators in a certain comprehensive state are not less than 0.7, the arithmetic mean of the scores of multiple types of indicators is taken as the final score of the comprehensive state; otherwise, the minimum value of the scores of multiple types of indicators is taken as the final score of the comprehensive state; The scores of zinc oxide lightning arresters under various comprehensive conditions are calculated for the historical indicator data set, current indicator data set, and future indicator data set.

5. A method for comprehensive status assessment of a zinc oxide lightning arrester according to claim 4, characterized in that: For indicators whose indicator data values are equal to the model threshold, the score is calculated using a fuzzy mathematical membership function; the model refers to a half-step-up model and a half-step-down model.

6. A method for comprehensive status assessment of a zinc oxide lightning arrester according to claim 4, characterized in that: The scoring-based method uses a Bayesian learning algorithm to count the number of samples of historical, current, and future comprehensive state combinations, and calculates the conditional probability corresponding to each comprehensive state level of the zinc oxide lightning arrester, including: ， in, For the parent node Take the first combination, at the same time The conditional probability of the combined state is For the parent node For the The confidence level obtained by mining association rules in a comprehensive state is For the parent node Take the first combination, at the same time The number of samples in the comprehensive state, for The parent node includes three nodes: history, current and future; the comprehensive status includes five levels: normal, attention, minor, abnormal and emergency; the combination is a comprehensive status combination of three nodes.

7. A zinc oxide arrester comprehensive status assessment system, characterized in that: include: An association mining unit is configured to: for any fault, obtain different types of indicator data sets corresponding to the fault, the indicator data sets including historical indicator data sets, current indicator data sets, and future indicator data sets, perform association rule mining on the indicator data sets to obtain multiple types of indicator data sets with the highest degree of correlation with the fault; the future indicator data sets are obtained based on Bayesian prediction; A scoring unit is configured to score the comprehensive status information of the zinc oxide lightning arrester based on multiple types of indicator data sets with the highest degree of correlation with the fault, and obtain scores for each comprehensive status information of the zinc oxide lightning arrester corresponding to the historical indicator data set, the current indicator data set, and the future indicator data set; The zinc oxide arrester comprehensive status information includes multiple comprehensive status levels; The comprehensive evaluation unit is configured to: count the number of samples of historical, current and future comprehensive state combinations based on the scores through a Bayesian learning algorithm, and calculate the conditional probability corresponding to each comprehensive state level of the zinc oxide lightning arrester.

8. A computer device, characterized in that: include: a processor and a computer-readable storage medium; a processor adapted to execute a computer program; A computer-readable storage medium having a computer program stored therein, wherein when the computer program is executed by the processor, the method for comprehensive status assessment of a zinc oxide lightning arrester according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing the method for comprehensive status assessment of a zinc oxide lightning arrester according to any one of claims 1 to 6.

10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, the method for comprehensive status assessment of a zinc oxide lightning arrester according to any one of claims 1 to 6 is implemented.

Images