Power distribution network single-phase grounding cause diagnosis method based on confidence and fuzziness evaluation

Abstract

The application discloses a power distribution network single-phase grounding reason diagnosis method based on confidence and fuzziness evaluation. The method comprises the following steps: obtaining original diagnosis data corresponding to a to-be-detected waveform; determining a confidence level corresponding to the original diagnosis data according to a preset grading rule based on the similarity distance corresponding to each of a plurality of neighbor sample waveforms in the original diagnosis data; determining a plurality of candidate reasons based on the single-phase grounding reasons corresponding to each of the plurality of neighbor sample waveforms; determining the occurrence frequency corresponding to each of the plurality of candidate reasons based on the number of the plurality of neighbor sample waveforms; determining a target fuzziness based on the occurrence frequency corresponding to each of the plurality of candidate reasons and a reason label semantic correlation matrix; and adaptively generating a diagnosis result based on the confidence level and the target fuzziness. The application solves the technical problem that, due to the insufficient information completeness of the output conclusion of the current analysis method, the operation and inspection personnel lack objective judgment basis for the reliability of the conclusion when making decisions.

Description

Technical Field

[0001] This invention relates to the field of power distribution network technology, and more specifically, to a method for diagnosing the causes of single-phase grounding in power distribution networks based on confidence and ambiguity assessment. Background Technology

[0002] For distribution networks with low-current grounding systems, multiple short-term grounding events often precede permanent grounding faults. In recent years, transient waveform fault indicators have been widely deployed in distribution networks, effectively capturing such precursory signals and providing a data foundation for proactive fault elimination. However, these transient signals are often weak and fleeting. Traditional methods relying on manual line inspections suffer from inherent drawbacks such as low efficiency, strong subjectivity, and poor accuracy. Therefore, researching efficient and accurate intelligent analysis methods has significant engineering value.

[0003] Intelligent diagnostic methods based on waveform similarity matching, as a case-based reasoning approach, effectively utilize accumulated high-quality labeled samples, and their performance can be iteratively expanded with the incremental knowledge base, making them an advanced and practical technical solution in this field. This method compares the waveform to be tested with a historical waveform sample library containing a large number of labeled causes based on morphological similarity, and makes a comprehensive judgment by referring to the labels of the most similar set of historical cases. However, this method still has the following problems: First, it only outputs the most probable diagnostic conclusion, lacking a quantitative evaluation mechanism for the reliability of the conclusion itself; second, when faced with ambiguous conclusions, it cannot qualitatively distinguish the nature of the ambiguity.

[0004] The aforementioned defects result in insufficient information completeness in the output conclusions, leaving operation and maintenance personnel without objective basis for judging the reliability of the conclusions when making decisions.

[0005] There is currently no effective solution to the above problems. Summary of the Invention

[0006] This invention provides a method for diagnosing the cause of single-phase grounding in distribution networks based on confidence and ambiguity assessment, in order to at least solve the technical problem that the lack of completeness of the information output by current analysis methods leads to a lack of objective basis for judging the reliability of conclusions when making decisions by operation and maintenance personnel.

[0007] According to one aspect of the present invention, a method for diagnosing the cause of single-phase grounding in a distribution network based on confidence and ambiguity assessment is provided, comprising: acquiring original diagnostic data corresponding to a waveform under test, wherein the original diagnostic data includes the cause of single-phase grounding and similarity distance corresponding to each of multiple neighboring sample waveforms corresponding to the waveform under test, the similarity distance representing the similarity distance between the corresponding neighboring sample waveform and the waveform under test; determining the confidence level corresponding to the original diagnostic data based on the similarity distance corresponding to each of the multiple neighboring sample waveforms in the original diagnostic data; determining multiple candidate causes corresponding to the waveform under test based on the cause of single-phase grounding corresponding to each of the multiple neighboring sample waveforms, wherein the multiple candidate causes are different; determining the occurrence frequency corresponding to each of the multiple candidate causes based on the number of multiple neighboring sample waveforms; determining the target ambiguity corresponding to the original diagnostic data based on the occurrence frequency corresponding to each of the multiple candidate causes and a preset cause label semantic association matrix; and determining the diagnostic result corresponding to the waveform under test based on the confidence level and the target ambiguity.

[0008] Optionally, based on the similarity distances between the waveforms of multiple nearest neighbor samples in the original diagnostic data, the confidence level corresponding to the original diagnostic data is determined, including: determining the extreme values ​​of the similarity distances between the waveforms of multiple nearest neighbor samples; determining the local confidence level based on the extreme values ​​of the similarity distances; calculating the mean of the similarity distances between the waveforms of multiple nearest neighbor samples; determining the global confidence level based on the mean of the similarity distances; calculating the standard deviation of the similarity distances between the waveforms of multiple nearest neighbor samples; determining the confidence level stability based on the standard deviation; and determining the confidence level corresponding to the original diagnostic data according to a preset grading rule based on the local confidence level, global confidence level, and confidence level stability.

[0009] Optionally, based on the occurrence frequencies of multiple candidate causes and a preset semantic association matrix of cause labels, the target ambiguity corresponding to the original diagnostic data is determined, including: based on the occurrence frequencies of multiple candidate causes, calculating the information entropy corresponding to the original diagnostic data according to a preset formula to determine the basic ambiguity; if the information entropy corresponding to the original diagnostic data exceeds a preset information entropy threshold, querying the semantic association matrix of cause labels to determine the semantic association degree between multiple candidate causes; and correcting the basic ambiguity based on the semantic association degree to determine the target ambiguity.

[0010] Optionally, the preset formula is as follows:

[0011] ,

[0012] in, The information entropy corresponding to the original diagnostic data. The set representing the waveforms of multiple nearest neighbor samples, where m is the number of multiple candidate causes; Let be the frequency of occurrence of the i-th candidate reason among multiple candidate reasons.

[0013] Optionally, it also includes: collecting multiple historical diagnostic data and diagnostic results corresponding to the multiple historical diagnostic data, wherein the historical diagnostic data includes the single-phase grounding cause and similarity distance corresponding to multiple historical neighbor sample waveforms corresponding to the historical waveform; calculating the information entropy corresponding to each of the multiple historical diagnostic data; and determining a preset information entropy threshold based on the information entropy corresponding to each of the multiple historical diagnostic data and the diagnostic results corresponding to each of the multiple historical diagnostic data.

[0014] Optionally, based on the confidence level and target ambiguity, the diagnostic result corresponding to the waveform under test is determined, including: based on the confidence level and target ambiguity, querying and locating in a preset two-dimensional diagnostic state matrix to determine the diagnostic scenario state corresponding to the waveform under test; based on the diagnostic scenario state corresponding to the waveform under test, matching the corresponding output strategy from a preset conclusion template library to generate the diagnostic result corresponding to the waveform under test.

[0015] According to another aspect of the present invention, a single-phase grounding cause diagnosis device for a distribution network based on confidence and ambiguity assessment is also provided, comprising: an acquisition module, configured to acquire original diagnostic data corresponding to a waveform under test, wherein the original diagnostic data includes single-phase grounding causes and similarity distances corresponding to multiple neighboring sample waveforms corresponding to the waveform under test, the similarity distance representing the similarity distance between the corresponding neighboring sample waveforms and the waveform under test; a first determination module, configured to determine the confidence level corresponding to the original diagnostic data based on the similarity distances corresponding to the multiple neighboring sample waveforms in the original diagnostic data; a second determination module, configured to determine multiple candidate causes corresponding to the waveform under test based on the single-phase grounding causes corresponding to the multiple neighboring sample waveforms, wherein the multiple candidate causes are different; a third determination module, configured to determine the occurrence frequency corresponding to each of the multiple candidate causes based on the number of multiple neighboring sample waveforms; a fourth determination module, configured to determine the target ambiguity corresponding to the original diagnostic data based on the occurrence frequency corresponding to each of the multiple candidate causes and a preset cause label semantic association matrix; and a fifth determination module, configured to determine the diagnostic result corresponding to the waveform under test based on the confidence level and the target ambiguity.

[0016] According to another aspect of the present invention, a non-volatile storage medium is also provided, the non-volatile storage medium including a stored program, wherein, when the program is running, the device where the non-volatile storage medium is located executes any of the above-described methods for diagnosing the cause of single-phase grounding in a power distribution network based on confidence and ambiguity assessment.

[0017] According to another aspect of the present invention, a computer device is also provided, the computer device including a processor, the processor being configured to run a program, wherein the program, when running, executes any of the above-described methods for diagnosing single-phase grounding causes in a power distribution network based on confidence and ambiguity assessment.

[0018] According to another aspect of the present invention, a computer program product is also provided, including a computer program that, when executed by a processor, implements any of the above-described methods for diagnosing single-phase grounding causes in a distribution network based on confidence and ambiguity assessment.

[0019] In this embodiment of the invention, a method for diagnosing single-phase grounding causes in a distribution network based on confidence and ambiguity assessment is adopted. This involves acquiring original diagnostic data corresponding to the waveform under test. The original diagnostic data includes the single-phase grounding causes and similarity distances corresponding to multiple neighboring sample waveforms of the waveform under test. The similarity distance characterizes the similarity distance between the corresponding neighboring sample waveforms and the waveform under test. Based on the similarity distances corresponding to the multiple neighboring sample waveforms in the original diagnostic data, the confidence level corresponding to the original diagnostic data is determined. Based on the single-phase grounding causes corresponding to the multiple neighboring sample waveforms, multiple candidate causes corresponding to the waveform under test are determined. The reasons are different; based on the number of waveforms of multiple nearest neighbor samples, the occurrence frequency of each of the multiple candidate reasons is determined; based on the occurrence frequency of each of the multiple candidate reasons and the preset semantic association matrix of the reason labels, the target ambiguity corresponding to the original diagnostic data is determined; based on the confidence level and the target ambiguity, the diagnostic result corresponding to the waveform to be tested is determined, thus achieving the purpose of quality assessment and intelligent interpretation of the diagnostic conclusion, thereby realizing the technical effect of improving the usability and credibility of the diagnostic results, and thus solving the technical problem that the lack of information completeness in the output conclusions of the current analysis methods causes the operation and maintenance personnel to lack objective judgment basis for the reliability of the conclusions when making decisions. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0021] Figure 1 A hardware block diagram of a computer terminal for implementing a method for diagnosing the cause of single-phase grounding in a distribution network based on confidence and ambiguity assessment is shown.

[0022] Figure 2 This is a flowchart illustrating a method for diagnosing single-phase grounding causes in a distribution network based on confidence and ambiguity assessment, according to an embodiment of the present invention.

[0023] Figure 3This is a schematic diagram of a method for diagnosing single-phase grounding causes in a distribution network based on confidence and ambiguity assessment, provided by an optional embodiment of the present invention.

[0024] Figure 4 This is a structural block diagram of a single-phase grounding cause diagnosis device for a distribution network based on confidence and ambiguity assessment, provided according to an embodiment of the present invention. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0027] According to an embodiment of the present invention, a method embodiment for diagnosing the cause of single-phase grounding in a distribution network based on confidence and ambiguity assessment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0028] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware block diagram of a computer terminal for implementing a method for diagnosing single-phase grounding causes in a distribution network based on confidence and ambiguity assessment is shown. Figure 1As shown, the computer terminal 10 may include one or more processors (shown as 102a, 102b, ..., 102n in the figure) (the processor may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0029] It should be noted that the aforementioned one or more processors and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10. As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0030] The memory 104 can be used to store software programs and modules for application software, such as the program instructions / data storage device corresponding to the distribution network single-phase grounding cause diagnosis method based on confidence and ambiguity assessment in this embodiment of the invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the above-mentioned application program for the distribution network single-phase grounding cause diagnosis method based on confidence and ambiguity assessment. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0031] The display can be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10.

[0032] Figure 2This is a flowchart illustrating a method for diagnosing single-phase grounding causes in a distribution network based on confidence and ambiguity assessment, according to an embodiment of the present invention. Figure 2 As shown, the method includes the following steps:

[0033] Step S202: Obtain the original diagnostic data corresponding to the waveform to be tested. The original diagnostic data includes the single-phase grounding cause and similarity distance of each of the multiple neighboring sample waveforms corresponding to the waveform to be tested. The similarity distance represents the similarity distance between the corresponding neighboring sample waveform and the waveform to be tested.

[0034] This step first relies on the pre-existing intelligent diagnostic module based on waveform similarity matching. This module, based on an established high-quality historical waveform sample library, uses elastic matching algorithms or similarity measurement methods, such as Euclidean distance, Manhattan distance, Dynamic Time Warping (DTW), or other suitable algorithms, to calculate the similarity distance between the waveform to be tested and each historical waveform in the library. Then, the module employs a K-nearest neighbor (KNN) strategy to select the K most similar historical samples to the waveform to be tested as the "nearest neighbor sample waveforms."

[0035] Specifically, each nearest neighbor waveform in the raw diagnostic data has a unique identifier for quickly locating and retrieving detailed information about that sample in a database or sample library. The raw data also needs to include a single-phase grounding cause label provided by the preliminary diagnostic module based on the nearest neighbor waveforms, representing the actual fault cause corresponding to the waveform considered most similar to the waveform under test. These causes can be various factors such as tree obstructions, bird damage, foreign metallic objects, and insulator contamination, each corresponding to specific physical phenomena and electrical characteristics. The raw diagnostic data also needs to include a similarity distance value, representing the similarity distance between each nearest neighbor waveform and the waveform under test. This is typically a numerical value; the smaller the value, the more similar the two are. The calculation method for the similarity distance depends on the similarity measurement algorithm used. For example, the DTW algorithm considers the time-series characteristics of the waveforms and calculates the minimum cumulative distance between the two sequences.

[0036] Once this raw diagnostic data is acquired, it will be passed to subsequent analysis modules for further multidimensional confidence and conclusion ambiguity assessment.

[0037] Step S204: Based on the similarity distances between the waveforms of multiple nearest neighbor samples in the original diagnostic data, determine the confidence level corresponding to the original diagnostic data.

[0038] In this step, the similarity distance values ​​of the K nearest neighbor waveforms most similar to the waveform to be tested are first extracted from the acquired raw diagnostic data. These distance values ​​reflect the degree of morphological similarity between the nearest neighbor waveforms and the waveform to be tested. The minimum value among these K similarity distance values ​​is found and denoted as . , as a quantitative indicator of local confidence. The smaller the value, the higher the local confidence, indicating that there is at least one very similar sample waveform.

[0039] Furthermore, based on Calculating local confidence can be done using simple mathematical functions, such as linear or exponential functions. Mapped to confidence levels. For example, if Below a certain set threshold, the local confidence level can be considered very high; conversely, above it, it is low. Then, the average waveform similarity distance of the K nearest neighbor samples is calculated, denoted as... This is used to evaluate the average matching degree between the waveform under test and a group of similar cases. Based on The size of the value can determine the global confidence level. The smaller the value, the higher the global confidence, indicating that the waveform under test is not only similar to one or a few specific sample waveforms, but also has a high similarity to the entire group of similar samples. The standard deviation of the K similarity distances also needs to be calculated, denoted as . To measure the consistency of waveform similarity among nearest-neighbor samples, The smaller the value, the higher the stability of the confidence level, indicating that the waveform similarity distribution of neighboring samples is more concentrated, and the diagnostic results are less affected by individual abnormal samples.

[0040] Finally, by comprehensively considering local confidence, global confidence, and confidence stability indicators, a pre-defined rule system is used to make a comprehensive judgment to derive the overall confidence level corresponding to the original diagnostic data. For example, if both local and global confidence levels are high and the stability is good, then the overall confidence level will also be very high. The comprehensive judgment rules can include setting weights for different indicators and formulating criteria for dividing confidence levels, ensuring that the final confidence assessment is comprehensive and reasonable.

[0041] Step S206: Based on the single-phase grounding causes corresponding to the waveforms of multiple neighboring samples, determine multiple candidate causes corresponding to the waveform under test, wherein the multiple candidate causes are different.

[0042] In this step, the cause labels for all neighboring sample waveforms are first collected and organized. These cause labels are provided by experts from the historical waveform library, reflecting the experts' judgments on the causes of single-phase grounding faults in historical waveforms. The single-phase grounding causes of the collected neighboring sample waveforms are then deduplicated to ensure the independence and completeness of the candidate causes. This deduplication step ensures that no duplicate cause types appear in the candidate cause set, thus making subsequent frequency statistics more accurate.

[0043] Furthermore, the frequency of occurrence of each different single-phase grounding cause in the nearest neighbor samples is statistically analyzed. Specifically, for each candidate cause, the number of times it appears in the K nearest neighbor samples is calculated. Based on the above statistical results, a candidate cause set is generated. Each element in this set not only contains the type of single-phase grounding cause but also its frequency of occurrence in the nearest neighbor samples. For example, if the single-phase grounding causes identified from the waveforms in the nearest neighbor samples include "metal grounding," "tree contact," "insulator flashover," and "foreign object contact," and each cause appears 3 times, 1 time, 4 times, and 1 time in the nearest neighbor samples, respectively, then the generated candidate cause set will be "metal grounding" (3 times), "tree contact" (1 time), "insulator flashover" (4 times), and "foreign object contact" (1 time). In this way, not only can multiple candidate grounding causes of the waveform under test be identified, but the distribution of these causes in historical data can also be clarified.

[0044] Step S208: Based on the number of waveforms of multiple neighboring samples, determine the occurrence frequency of each of the multiple candidate causes.

[0045] In this step, waveform similarity matching technology is first used to find multiple nearest-neighbor sample waveforms that are most similar to the waveform under test. These nearest-neighbor sample waveforms may each correspond to different single-phase grounding causes, and these causes constitute a set of candidate causes for the waveform under test. For example, suppose the found nearest-neighbor sample waveforms cover four different single-phase grounding causes, namely C1, C2, C3, and C4.

[0046] Furthermore, the occurrence frequency of candidate causes is counted by iterating through all found nearest-neighbor waveform samples and recording the frequency of each candidate cause across all nearest-neighbor waveform samples. Based on the above assumptions, let's say C1 occurs 5 times, C2 occurs 3 times, C3 occurs 1 time, and C4 occurs 2 times. Divide the occurrence frequency of each candidate cause by the total number of nearest-neighbor waveform samples to obtain the frequency of each candidate cause. Assuming the total number of found nearest-neighbor waveform samples is K, then for the above example, the frequency of C1 is 5 / K, the frequency of C2 is 3 / K, the frequency of C3 is 1 / K, and the frequency of C4 is 2 / K. The calculated frequencies of candidate causes are then sorted to facilitate ambiguity evaluation in subsequent steps. The sorting can be descending or ascending, depending on the requirements of the subsequent processing logic.

[0047] After determining the frequency of occurrence of all candidate causes, the candidate cause with the highest frequency is selected as the preliminary diagnostic result.

[0048] Step S210: Based on the occurrence frequency of each of the multiple candidate causes and the preset semantic association matrix of cause labels, determine the target ambiguity corresponding to the original diagnostic data.

[0049] In this step, firstly, based on the single-phase grounding causes corresponding to the multiple nearest-neighbor sample waveforms obtained in step S206, the frequency of each candidate cause appearing in the K nearest-neighbor samples is counted. Since the K nearest-neighbor sample waveforms are used to diagnose the grounding cause of the waveform under test, the single-phase grounding cause labels they carry constitute a candidate set of possible causes for the waveform under test. By calculating the frequency of each cause appearing in the nearest-neighbor samples, the occurrence frequency of each candidate cause is obtained.

[0050] Furthermore, by utilizing the frequency of occurrence of candidate causes, ambiguity can be quantified by calculating information entropy. Information entropy is commonly used to measure the uncertainty of information. In this step, information entropy is used to measure which of multiple candidate causes is the most probable cause and the uncertainty among these causes. The formula for calculating information entropy is as follows:

[0051]

[0052] in, The information entropy corresponding to the original diagnostic data. The set representing the waveforms of multiple nearest neighbor samples, where m is the number of multiple candidate causes; Let be the frequency of occurrence of the i-th candidate reason among multiple candidate reasons.

[0053] Specifically, the higher the information entropy value, the more evenly the distribution of multiple candidate causes is, and the higher the uncertainty, i.e., the greater the ambiguity; conversely, if the information entropy value is small, it indicates that there is one or a few dominant causes, and the diagnosis result is clearer.

[0054] Finally, based on the calculated information entropy, if the information entropy is higher than a preset information entropy threshold, indicating high ambiguity in the diagnostic result, a preset cause-label semantic association matrix will be further introduced to analyze the inherent connections and similarities between different fault causes. Based on the analysis results of the inherent connections and similarities between different fault causes, the ambiguity level is adjusted. If competing causes have high semantic similarity, the ambiguity level may be downgraded from "high" to "medium," indicating that although multiple competing causes exist, they point to similar problems, and the ambiguity nature of the diagnostic result is qualitatively interpreted.

[0055] Through the above steps, based on the frequency distribution of fault cause labels and the preset semantic association matrix, this invention can perform in-depth analysis and correction of the ambiguity of the original diagnostic data, thereby providing a more accurate and comprehensive quality assessment of the diagnostic results and providing strong support for adaptive diagnosis of short-term grounding causes in distribution networks.

[0056] Step S212: Based on the confidence level and target ambiguity, determine the diagnostic result corresponding to the waveform to be tested.

[0057] After obtaining the confidence level and ambiguity index of the waveform under test, this step first maps these two indices to a predefined two-dimensional diagnostic state matrix. The rows and columns of this matrix represent the confidence level (high, medium, low) and ambiguity level (high, medium, low), respectively. Each cell corresponds to a specific diagnostic scenario state, such as a high confidence low ambiguity state, a medium confidence medium ambiguity state, etc.

[0058] Furthermore, the aforementioned two-dimensional diagnostic state matrix is ​​positioned to a specific cell, which represents the current diagnostic scenario. The positioned cell contains a pre-defined optimal output strategy and text template. Based on the strategy of that cell, and combined with specific information about the waveform under test (such as candidate causes, confidence indices, ambiguity indices, etc.), the system fills the corresponding text template to generate the final diagnostic conclusion text that matches the current diagnostic scenario state.

[0059] Through the above steps, the technical effect of improving the usability and reliability of diagnostic results can be achieved, thereby solving the technical problem that the lack of completeness of information in the conclusions output by current analysis methods leads to a lack of objective basis for judging the reliability of conclusions when making decisions.

[0060] As an optional embodiment, this can be achieved through the following steps: determining the confidence level of the original diagnostic data based on the similarity distances corresponding to the waveforms of multiple nearest neighbor samples in the original diagnostic data, including: determining the extreme values ​​of the similarity distances among the similarity distances corresponding to the waveforms of multiple nearest neighbor samples; determining the local confidence level based on the extreme values ​​of the similarity distances; calculating the mean of the similarity distances corresponding to the waveforms of multiple nearest neighbor samples; determining the global confidence level based on the mean of the similarity distances; calculating the standard deviation of the similarity distances corresponding to the waveforms of multiple nearest neighbor samples; determining the confidence stability based on the standard deviation; and determining the confidence level of the original diagnostic data according to a preset grading rule based on the local confidence level, global confidence level, and confidence stability.

[0061] In this step, the similarity distance values ​​of the K nearest neighbor waveforms of the waveform to be tested are first extracted from the acquired raw diagnostic data. The smaller these distance values ​​are, the higher the morphological match with the waveform to be tested, and the greater the reliability. Local confidence is measured by finding the minimum value among these distance values.

[0062] Specifically, let the set of distance values ​​for the K nearest neighbor waveforms of the waveform to be measured be denoted as . ,in This represents the similarity distance between the waveform of the i-th nearest neighbor sample and the waveform to be tested. It is a core indicator of local confidence, representing the extreme value (minimum) of the similarity distance. The calculation formula is as follows:

[0063] ,

[0064] in, The smaller the value, the higher the local confidence level.

[0065] Furthermore, a global confidence score can be calculated. This score is measured by assessing the degree of match between the waveform under test and the overall "group of similar cases," reflecting the general applicability of the diagnostic conclusion. Specifically, this involves calculating the distance set. The arithmetic mean of all distance values ​​is used as the global confidence index. :

[0066] ,

[0067] in, For set The i-th distance value in the equation. The smaller the value, the higher the global confidence level.

[0068] Furthermore, it is necessary to assess the consistency of nearest neighbor matching, that is, the internal stability of diagnostic evidence. This can be done by calculating the standard deviation of the similarity distance, as shown in the following formula:

[0069] ,

[0070] in, The smaller the value, the higher the confidence stability.

[0071] Based on an evaluation of three dimensions—local confidence, global confidence, and confidence stability—a pre-defined rule system is used to determine the confidence level of the original diagnostic data. This process involves threshold setting and grading rules to comprehensively determine the level of confidence. Preferably, this embodiment uses the percentile method to determine the confidence index through statistical analysis of massive historical sample data. , , The evaluation threshold T for "excellent" and "poor" of the three indicators that are considered "better the smaller the value". 优 and T 差 It can be set using the following formula:

[0072] ;

[0073] ;

[0074] in, P represents the set of distributions of a certain indicator on a historical dataset. 30 With P 70These represent the 30th and 70th percentiles of the set, respectively. When the index value is below T... 优 At that time, it was rated as "excellent"; higher than T. 差 When the condition is "poor", it is rated as "poor"; when it falls between the two, it is rated as "good".

[0075] Finally, the rule system is used to comprehensively evaluate the three qualitative levels to output the final confidence level. In this optional embodiment, Figure 3 This is a schematic diagram of a single-phase grounding cause diagnosis method for distribution networks based on confidence and ambiguity assessment, provided by an optional embodiment of the present invention. Figure 3 As shown, if the global confidence level is poor (indicating a low overall match with historical cases), or the local confidence level is good but the confidence stability level is poor (indicating a conflict of evidence where there are individual similarities but large group differences), then the overall confidence level is determined to be low. If none of the above low confidence conditions are met, and the local confidence level is good and the global confidence level is good, or the global confidence level is good and the confidence stability level is good, then the final confidence level is determined to be high. If none of the above conditions are met, then the final confidence level is determined to be medium.

[0076] As an optional embodiment, this can be achieved through the following steps: determining the target ambiguity of the original diagnostic data based on the occurrence frequency of each of the multiple candidate causes and a preset semantic association matrix of cause labels, including: calculating the information entropy of the original diagnostic data according to a preset formula based on the occurrence frequency of each of the multiple candidate causes to determine the basic ambiguity; querying the semantic association matrix of cause labels to determine the semantic association degree between the multiple candidate causes when the information entropy of the original diagnostic data exceeds a preset information entropy threshold; and correcting the basic ambiguity based on the semantic association degree to determine the target ambiguity.

[0077] In this step, the first step is to record the cause of a single-phase grounding for each of the K nearest neighbor waveform samples obtained from the original diagnostic data. During this process, it is necessary to identify all different candidate causes and count the number of times each candidate cause appears in the K nearest neighbor waveform samples. If there are m different candidate causes, then an m-dimensional frequency vector will be obtained.

[0078] Based on this frequency vector, information entropy is calculated. Information entropy is a statistical indicator that measures the degree of confusion or uncertainty of information. It can be used to assess the clarity or ambiguity of diagnostic results. Based on the relationship between information entropy and a preset information entropy threshold, the level of ambiguity is determined, and the numerical value of information entropy is converted into an easily understood and applicable level of ambiguity, such as: low, medium, and high.

[0079] Specifically, a basic ambiguity is determined based on a defined information entropy. Then, an information entropy threshold is preset to distinguish between clear and ambiguous diagnostic results. This threshold can be determined based on the distribution of historical data and through statistical analysis. If the calculated information entropy exceeds the preset threshold, the diagnostic result is considered to have high ambiguity and requires further analysis. A preset semantic association matrix of cause labels can be queried, reflecting the degree of correlation between various fault causes defined by domain experts. The semantic association degree between the two candidate causes with the most votes is calculated. A semantic association degree threshold can be preset to distinguish whether the association between labels is strong enough to be considered to point to the same or similar fault type. If the semantic association degree is higher than the semantic association degree threshold, it is considered that although there are multiple competing causes, due to their inherent correlation in physical causes or discharge processes, the ambiguity level can be downgraded to a lower level, indicating that the ambiguity is caused by different manifestations of the same type of problem. Conversely, if the semantic association degree does not exceed the semantic association degree threshold, the ambiguity level remains unchanged or increases, indicating that there are essentially different competing fault causes, and the diagnostic result is more ambiguous.

[0080] Through these steps, the present invention not only quantifies the basic ambiguity caused by the uncertainty of the result distribution, but also intelligently corrects the basic ambiguity based on the semantic correlation between cause labels, thereby more accurately determining the target ambiguity. This makes the interpretation of diagnostic results closer to actual engineering applications, improving the accuracy and practicality of the diagnosis.

[0081] As an optional embodiment, it can be achieved through the following steps: The preset formula is as follows:

[0082] ,

[0083] in, The information entropy corresponding to the original diagnostic data. The set representing the waveforms of multiple nearest neighbor samples, where m is the number of multiple candidate causes; Let be the frequency of occurrence of the i-th candidate reason among multiple candidate reasons.

[0084] The preset formula in this embodiment is the information entropy calculation formula. After calculating the information entropy using the preset formula, its value is analyzed. Information Entropy The numerical range of is from 0 to the maximum possible value, which occurs when all candidate causes occur with equal frequency, i.e., when This applies to all values ​​of i. A lower value indicates more concentrated information and a clearer diagnostic conclusion (lower ambiguity); a higher value indicates more dispersed information and a more ambiguous diagnostic conclusion.

[0085] Furthermore, to distinguish different levels of ambiguity, an information entropy threshold can be set. This threshold can be determined by statistically analyzing the information entropy distribution of historical diagnostic data, such as by finding inflection points in the information entropy distribution, or by selecting an appropriate value based on experience. Then, based on the calculations... To determine the level of ambiguity. If If the ambiguity level is low, the diagnostic result is relatively clear; if If the ambiguity level is high, the diagnostic result may have multiple interpretations.

[0086] Furthermore, in-depth semantic correlation analysis can be performed. By incorporating domain expert knowledge, it can determine whether multiple competing candidate causes are inherently related in terms of physical origin. Specifically, a pre-defined semantic correlation matrix M of fault causes can be queried. semantic The final ambiguity level is determined by combining basic ambiguity and semantic relevance analysis. The specific process is as follows:

[0087] Step 1: Set the information entropy threshold T to distinguish between "low ambiguity" and "high ambiguity". H The threshold is determined using the following steps:

[0088] First, determine the "boundary votes" that mark the beginning of ambiguity. :

[0089] ;

[0090] ;

[0091] in, The minimum number of votes required to represent a "clear majority" For floor operations, The coefficient is set in advance.

[0092] Secondly, determine the information entropy threshold. The information entropy corresponding to the boundary vote count is used as the information entropy threshold, calculated using the following formula:

[0093]

[0094]

[0095]

[0096] Step 2, if information entropy Regardless of semantic relevance, the final ambiguity level is directly determined as "low".

[0097] Step 3, if information entropy Then calculate the reasons for the two candidates with the most votes (denoted as L). a and L b ) in M semantic The semantic relevance sim(L) a , L b If sim(L) a , L b If the ambiguity level is higher than the preset semantic threshold, the final ambiguity level correction is determined to be "medium" (indicating that there is a divergence, but it points to similar issues); otherwise, the final ambiguity level is determined to be "high" (indicating that there is a large divergence, pointing to different issues).

[0098] As an optional embodiment, this can be achieved through the following steps: It further includes: collecting multiple historical diagnostic data and corresponding diagnostic results, wherein the historical diagnostic data includes the single-phase grounding cause and similarity distance corresponding to multiple historical neighbor sample waveforms corresponding to historical waveforms; calculating the information entropy corresponding to each of the multiple historical diagnostic data; and determining a preset information entropy threshold based on the information entropy corresponding to each of the multiple historical diagnostic data and the corresponding diagnostic results.

[0099] In this step, multiple historical diagnostic data sets and their corresponding diagnostic results are collected. Here, "historical diagnostic data" refers to transient waveform records where the cause of a single-phase grounding has been diagnosed and confirmed in the past. This historical data includes not only the historical waveforms but also the single-phase grounding causes and similarity distances corresponding to multiple matching historical neighboring waveform samples, as well as the final confirmed diagnostic results.

[0100] Furthermore, the information entropy corresponding to each of the multiple historical diagnostic data points is calculated. After collecting sufficient historical diagnostic data, the ambiguity of each historical diagnostic data point is quantified using the aforementioned information entropy calculation formula. This information entropy calculation considers the frequency distribution of different single-phase grounding causes in the historical diagnostic data, thereby reflecting the clarity or ambiguity of the diagnostic results. Based on the information entropy corresponding to each of the multiple historical diagnostic data points and the corresponding diagnostic results, a preset information entropy threshold is determined. This process involves the correlation analysis between the information entropy of historical data and the diagnostic results. Through observation and analysis, the relationship between different numerical ranges of information entropy and the correctness, clarity, and confidence of the diagnostic results can be identified. For example, an information entropy threshold can be set; when the calculated information entropy is below this threshold, the reliability and clarity of the diagnostic results can be determined. The determination of the information entropy threshold can employ methods including, but not limited to, percentile methods, statistical analysis methods, or machine learning methods.

[0101] As an optional embodiment, this can be achieved through the following steps: determining the diagnostic result corresponding to the waveform under test based on the confidence level and the target ambiguity, including: querying and locating in a preset two-dimensional diagnostic state matrix based on the confidence level and the target ambiguity to determine the diagnostic scenario state corresponding to the waveform under test; and matching the corresponding output strategy from a preset conclusion template library based on the diagnostic scenario state corresponding to the waveform under test to generate the diagnostic result corresponding to the waveform under test.

[0102] In this step, we first need to design a two-dimensional diagnostic state matrix, where the rows and columns represent different levels of confidence and ambiguity, respectively. For example, confidence can be divided into three levels: high, medium, and low, and ambiguity can also be divided into three levels: high, medium, and low. Each cell in the matrix corresponds to a specific diagnostic scenario state, which describes the possible quality of the diagnostic result or the necessary follow-up actions under the combined conditions of confidence and ambiguity.

[0103] Furthermore, the diagnostic results of the waveform under test are mapped to the corresponding cells in a two-dimensional diagnostic state matrix. For example, if the obtained confidence level is "high" and the ambiguity level is "low," then the "high confidence clarity state" is located in the matrix. After locating the specific state, the optimal output strategy and text template matching that state are selected from a pre-set conclusion template library. The conclusion template library contains text templates and suggested follow-up actions for each diagnostic scenario state. For example, for the "high confidence clarity state," the template could be: "Diagnostic Conclusion: [Specific Cause]. High diagnostic confidence, strong consistency with historical evidence. Recommendation: No further investigation required, implement the [Specific Cause] remediation plan." Each template aims to provide maintenance personnel with a clear, specific, and detailed guidance to help them make decisions.

[0104] Finally, the specific information of this diagnosis, such as the most likely cause of grounding, confidence level, and ambiguity level, can be filled into the selected template to generate the final diagnostic conclusion text. For example, if the most likely cause of grounding is "lightning strike," the confidence level is "high," and the ambiguity level is "low," then the system will generate the text: "Diagnosis Conclusion: Lightning strike. High confidence level, strong consistency with historical evidence." and output it to the operation and maintenance personnel.

[0105] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0106] Through the above description of the embodiments, those skilled in the art can clearly understand that the single-phase grounding cause diagnosis method for distribution networks based on confidence and fuzziness assessment according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0107] According to embodiments of the present invention, a device for diagnosing single-phase grounding causes in a distribution network based on confidence and ambiguity assessment is also provided for implementing the above-described method for diagnosing single-phase grounding causes in a distribution network based on confidence and ambiguity assessment. Figure 4 This is a structural block diagram of a distribution network single-phase grounding cause diagnosis device based on confidence and ambiguity assessment according to an embodiment of the present invention, as shown below. Figure 4 As shown, the distribution network single-phase grounding cause diagnosis device based on confidence and ambiguity assessment includes: acquisition module 402, first determination module 404, second determination module 406, third determination module 408, fourth determination module 410 and fifth determination module 412. The distribution network single-phase grounding cause diagnosis device based on confidence and ambiguity assessment will be described below.

[0108] The acquisition module 402 is used to acquire the original diagnostic data corresponding to the waveform under test. The original diagnostic data includes the single-phase grounding cause and similarity distance of each of the multiple neighboring sample waveforms corresponding to the waveform under test. The similarity distance characterizes the similarity distance between the corresponding neighboring sample waveform and the waveform under test.

[0109] The first determining module 404, connected to the acquiring module 402, is used to determine the confidence level of the original diagnostic data based on the similarity distance between the waveforms of multiple nearest neighbor samples in the original diagnostic data.

[0110] The second determining module 406 is connected to the first determining module 404 and is used to determine multiple candidate causes corresponding to the waveform under test based on the single-phase grounding causes corresponding to multiple neighboring sample waveforms, wherein the multiple candidate causes are different.

[0111] The third determining module 408, connected to the second determining module 406, is used to determine the occurrence frequency of each of the multiple candidate causes based on the number of waveforms of multiple nearest neighbor samples.

[0112] The fourth determining module 410, connected to the third determining module 408, is used to determine the target ambiguity corresponding to the original diagnostic data based on the occurrence frequency of each of the multiple candidate causes and the preset cause label semantic association matrix.

[0113] The fifth determining module 412, connected to the fourth determining module 410, is used to determine the diagnostic result corresponding to the waveform under test based on the confidence level and the target ambiguity.

[0114] It should be noted that the aforementioned acquisition module 402, first determination module 404, second determination module 406, third determination module 408, fourth determination module 410, and fifth determination module 412 correspond to steps S202 to S212 in the embodiments. Multiple modules implement the same instances and application scenarios as their corresponding steps, but are not limited to the content disclosed in the above embodiments. It should also be noted that the aforementioned modules, as part of the device, can run on the computer terminal 10 provided in the embodiments.

[0115] Embodiments of the present invention may provide a computer device. Optionally, in this embodiment, the computer device may be located in at least one of a plurality of network devices in a computer network. The computer device includes a memory and a processor.

[0116] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the method and device for diagnosing single-phase grounding causes in distribution networks based on confidence and ambiguity assessment in this embodiment of the invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby realizing the aforementioned method for diagnosing single-phase grounding causes in distribution networks based on confidence and ambiguity assessment. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to a computer terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0117] The processor can invoke information and application programs stored in the memory via a transmission device to perform the following steps: acquiring the original diagnostic data corresponding to the waveform under test, wherein the original diagnostic data includes the single-phase grounding cause and similarity distance of each of the multiple neighboring sample waveforms corresponding to the waveform under test, and the similarity distance characterizes the similarity distance between the corresponding neighboring sample waveform and the waveform under test; determining the confidence level corresponding to the original diagnostic data based on the similarity distance of each of the multiple neighboring sample waveforms in the original diagnostic data; determining multiple candidate causes corresponding to the waveform under test based on the single-phase grounding cause of each of the multiple neighboring sample waveforms, wherein the multiple candidate causes are different; determining the occurrence frequency of each of the multiple candidate causes based on the number of multiple neighboring sample waveforms; determining the target ambiguity corresponding to the original diagnostic data based on the occurrence frequency of each of the multiple candidate causes and a preset cause label semantic association matrix; and determining the diagnostic result corresponding to the waveform under test based on the confidence level and the target ambiguity.

[0118] Optionally, the processor may also execute program code for the following steps: determining the confidence level of the original diagnostic data based on the similarity distances between the waveforms of multiple nearest neighbor samples in the original diagnostic data, including: determining the extreme values ​​of the similarity distances between the waveforms of multiple nearest neighbor samples; determining the local confidence level based on the extreme values ​​of the similarity distances; calculating the mean of the similarity distances between the waveforms of multiple nearest neighbor samples; determining the global confidence level based on the mean of the similarity distances; calculating the standard deviation of the similarity distances between the waveforms of multiple nearest neighbor samples; determining the confidence level stability based on the standard deviation; and determining the confidence level of the original diagnostic data according to a preset grading rule based on the local confidence level, the global confidence level, and the confidence level stability.

[0119] Optionally, the processor may also execute program code for the following steps: determining the target ambiguity of the original diagnostic data based on the occurrence frequencies of multiple candidate causes and a preset semantic association matrix of cause labels, including: calculating the information entropy of the original diagnostic data according to a preset formula based on the occurrence frequencies of multiple candidate causes to determine the basic ambiguity; querying the semantic association matrix of cause labels to determine the semantic association degree between multiple candidate causes when the information entropy of the original diagnostic data exceeds a preset information entropy threshold; and correcting the basic ambiguity based on the semantic association degree to determine the target ambiguity.

[0120] Optionally, the processor can also execute program code with the following steps: The preset formula is as follows:

[0121] ,

[0122] in, The information entropy corresponding to the original diagnostic data. The set representing the waveforms of multiple nearest neighbor samples, where m is the number of multiple candidate causes; Let be the frequency of occurrence of the i-th candidate reason among multiple candidate reasons.

[0123] Optionally, the processor may also execute program code that includes the following steps: acquiring multiple historical diagnostic data and diagnostic results corresponding to the multiple historical diagnostic data, wherein the historical diagnostic data includes the single-phase grounding cause and similarity distance corresponding to each of the multiple historical neighbor sample waveforms corresponding to the historical waveform; calculating the information entropy corresponding to each of the multiple historical diagnostic data; and determining a preset information entropy threshold based on the information entropy corresponding to each of the multiple historical diagnostic data and the diagnostic results corresponding to each of the multiple historical diagnostic data.

[0124] Optionally, the processor may also execute program code for the following steps: determining the diagnostic result corresponding to the waveform under test based on the confidence level and target ambiguity, including: querying and locating in a preset two-dimensional diagnostic state matrix based on the confidence level and target ambiguity to determine the diagnostic scenario state corresponding to the waveform under test; and matching the corresponding output strategy from a preset conclusion template library based on the diagnostic scenario state corresponding to the waveform under test to generate the diagnostic result corresponding to the waveform under test.

[0125] This invention provides a scheme for diagnosing single-phase grounding causes in distribution networks based on confidence and ambiguity assessment. The method involves acquiring original diagnostic data corresponding to the waveform under test, including the single-phase grounding causes and similarity distances of multiple neighboring waveforms. The similarity distance characterizes the similarity between the neighboring waveforms and the waveform under test. Based on the similarity distances of the neighboring waveforms, the confidence level of the original diagnostic data is determined. Based on the single-phase grounding causes of the neighboring waveforms, multiple candidate causes are identified, each with different causes. Based on the number of neighboring waveforms, the frequency of occurrence of each candidate cause is determined. Based on the frequency of occurrence and a preset semantic association matrix of cause labels, the target ambiguity of the original diagnostic data is determined. Finally, based on the confidence level and target ambiguity, the diagnostic result for the waveform under test is determined. This achieves the purpose of quality assessment and intelligent interpretation of the diagnostic conclusions, thereby solving the technical problem in related technologies where insufficient information completeness in the output conclusions of current analysis methods leads to a lack of objective judgment criteria for the reliability of conclusions when making decisions.

[0126] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a non-volatile storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0127] Embodiments of the present invention also provide a non-volatile storage medium. Optionally, in this embodiment, the aforementioned non-volatile storage medium can be used to store the program code executed by the distribution network single-phase grounding cause diagnosis method based on confidence and ambiguity assessment provided in the above embodiments.

[0128] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0129] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: acquiring the original diagnostic data corresponding to the waveform under test, wherein the original diagnostic data includes the single-phase grounding cause and similarity distance corresponding to each of the multiple neighboring sample waveforms corresponding to the waveform under test, and the similarity distance characterizes the similarity distance between the corresponding neighboring sample waveform and the waveform under test; determining the confidence level corresponding to the original diagnostic data based on the similarity distance corresponding to each of the multiple neighboring sample waveforms in the original diagnostic data; determining multiple candidate causes corresponding to the waveform under test based on the single-phase grounding cause corresponding to each of the multiple neighboring sample waveforms, wherein the multiple candidate causes are different; determining the occurrence frequency corresponding to each of the multiple candidate causes based on the number of multiple neighboring sample waveforms; determining the target ambiguity corresponding to the original diagnostic data based on the occurrence frequency corresponding to each of the multiple candidate causes and the preset cause label semantic association matrix; and determining the diagnostic result corresponding to the waveform under test based on the confidence level and the target ambiguity.

[0130] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: determining the confidence level of the original diagnostic data based on the similarity distances corresponding to the waveforms of multiple nearest neighbor samples in the original diagnostic data, including: determining the extreme values ​​of the similarity distances among the similarity distances corresponding to the waveforms of multiple nearest neighbor samples; determining the local confidence level based on the extreme values ​​of the similarity distances; calculating the mean of the similarity distances corresponding to the waveforms of multiple nearest neighbor samples; determining the global confidence level based on the mean of the similarity distances; calculating the standard deviation of the similarity distances corresponding to the waveforms of multiple nearest neighbor samples; determining the confidence level stability based on the standard deviation; and determining the confidence level of the original diagnostic data according to a preset grading rule based on the local confidence level, the global confidence level, and the confidence level stability.

[0131] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: determining the target ambiguity corresponding to the original diagnostic data based on the occurrence frequency of each of the multiple candidate causes and a preset semantic association matrix of cause labels, including: calculating the information entropy corresponding to the original diagnostic data according to a preset formula based on the occurrence frequency of each of the multiple candidate causes to determine the basic ambiguity; querying the semantic association matrix of cause labels to determine the semantic association degree between the multiple candidate causes when the information entropy corresponding to the original diagnostic data exceeds a preset information entropy threshold; and correcting the basic ambiguity based on the semantic association degree to determine the target ambiguity.

[0132] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: The preset formula is as follows:

[0133] ,

[0134] in, The information entropy corresponding to the original diagnostic data. The set representing the waveforms of multiple nearest neighbor samples, where m is the number of multiple candidate causes; Let be the frequency of occurrence of the i-th candidate reason among multiple candidate reasons.

[0135] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: This further includes: collecting multiple historical diagnostic data and diagnostic results corresponding to the multiple historical diagnostic data, wherein the historical diagnostic data includes the single-phase grounding cause and similarity distance corresponding to each of the multiple historical neighbor sample waveforms corresponding to the historical waveform; calculating the information entropy corresponding to each of the multiple historical diagnostic data; and determining a preset information entropy threshold based on the information entropy corresponding to each of the multiple historical diagnostic data and the diagnostic results corresponding to each of the multiple historical diagnostic data.

[0136] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: determining the diagnostic result corresponding to the waveform under test based on the confidence level and the target ambiguity, including: querying and locating in a preset two-dimensional diagnostic state matrix based on the confidence level and the target ambiguity to determine the diagnostic scenario state corresponding to the waveform under test; and matching the corresponding output strategy from a preset conclusion template library based on the diagnostic scenario state corresponding to the waveform under test to generate the diagnostic result corresponding to the waveform under test.

[0137] Embodiments of the present invention also provide a computer program product, including a computer program. Optionally, in this embodiment, when the computer program is executed by a processor, it can: acquire original diagnostic data corresponding to a waveform under test, wherein the original diagnostic data includes the single-phase grounding cause and similarity distance corresponding to each of multiple neighboring sample waveforms corresponding to the waveform under test, the similarity distance representing the similarity distance between the corresponding neighboring sample waveform and the waveform under test; determine the confidence level corresponding to the original diagnostic data based on the similarity distance corresponding to each of the multiple neighboring sample waveforms in the original diagnostic data; determine multiple candidate causes corresponding to the waveform under test based on the single-phase grounding cause corresponding to each of the multiple neighboring sample waveforms, wherein the multiple candidate causes are different; determine the occurrence frequency corresponding to each of the multiple candidate causes based on the number of multiple neighboring sample waveforms; determine the target ambiguity corresponding to the original diagnostic data based on the occurrence frequency corresponding to each of the multiple candidate causes and a preset cause label semantic association matrix; and determine the diagnostic result corresponding to the waveform under test based on the confidence level and the target ambiguity.

[0138] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0139] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0140] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0141] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0142] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0143] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a non-volatile storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0144] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for diagnosing the causes of single-phase grounding in distribution networks based on confidence and fuzziness assessment, characterized in that, include: Obtain the original diagnostic data corresponding to the waveform under test, wherein the original diagnostic data includes the single-phase grounding cause and similarity distance of each of the multiple neighboring sample waveforms corresponding to the waveform under test, and the similarity distance characterizes the similarity distance between the corresponding neighboring sample waveform and the waveform under test; Based on the similarity distances between the waveforms of the multiple nearest neighbor samples in the original diagnostic data, the confidence level corresponding to the original diagnostic data is determined. Based on the single-phase grounding causes corresponding to the waveforms of the multiple neighboring samples, multiple candidate causes corresponding to the waveform under test are determined, wherein the multiple candidate causes are different; Based on the number of waveforms of the multiple nearest neighbor samples, the occurrence frequency of each of the multiple candidate causes is determined; Based on the occurrence frequency of each of the multiple candidate causes and the preset semantic association matrix of cause labels, the target ambiguity corresponding to the original diagnostic data is determined. Based on the confidence level and the target ambiguity, the diagnostic result corresponding to the waveform under test is determined.

2. The method according to claim 1, characterized in that, The step of determining the confidence level of the original diagnostic data based on the similarity distances between the waveforms of the multiple nearest neighbor samples in the original diagnostic data includes: Determine the extreme value of the similarity distance among the similarity distances corresponding to the waveforms of the plurality of nearest neighbor samples, wherein the extreme value of the similarity distance is the minimum value; Based on the extreme values ​​of the similarity distance, the local confidence level is determined; Calculate the mean similarity distance between the waveforms of the multiple nearest neighbor samples; The global confidence level is determined based on the mean similarity distance. Calculate the standard deviation of the similarity distance between the waveforms of the multiple nearest neighbor samples; Based on the standard deviation, the stability of the confidence level is determined; Based on the local confidence level, the global confidence level, and the confidence level stability, the confidence level corresponding to the original diagnostic data is determined according to a preset grading rule.

3. The method according to claim 1, characterized in that, The step of determining the target ambiguity corresponding to the original diagnostic data based on the occurrence frequency of each of the multiple candidate causes and a preset semantic association matrix of cause labels includes: Based on the occurrence frequency of each of the multiple candidate causes, the information entropy corresponding to the original diagnostic data is calculated according to a preset formula to determine the basic ambiguity. If the information entropy corresponding to the original diagnostic data exceeds a preset information entropy threshold, the semantic association matrix of the cause label is queried to determine the semantic association degree between the multiple candidate causes. Based on the semantic relevance, the basic ambiguity is corrected to determine the target ambiguity.

4. The method according to claim 3, characterized in that, The preset formula is as follows: ， in, The information entropy corresponding to the original diagnostic data. This represents the set corresponding to the waveforms of the multiple nearest neighbor samples. m The number of the plurality of candidate reasons; The first of the multiple candidate reasons i The frequency of occurrence of each candidate reason.

5. The method according to claim 3, characterized in that, Also includes: Collect multiple historical diagnostic data and corresponding diagnostic results, wherein the historical diagnostic data includes the single-phase grounding cause and similarity distance of each of the multiple historical neighbor sample waveforms corresponding to the historical waveform; Calculate the information entropy corresponding to each of the multiple historical diagnostic data; Based on the information entropy corresponding to each of the multiple historical diagnostic data and the diagnostic results corresponding to each of the multiple historical diagnostic data, the preset information entropy threshold is determined.

6. The method according to claim 1, characterized in that, The step of determining the diagnostic result corresponding to the waveform under test based on the confidence level and the target ambiguity includes: Based on the confidence level and the target ambiguity, a query is performed in the preset two-dimensional diagnostic state matrix to determine the diagnostic scenario state corresponding to the waveform to be tested; Based on the diagnostic scenario state corresponding to the waveform under test, the corresponding output strategy is matched from the preset conclusion template library to generate the diagnostic result corresponding to the waveform under test.

7. A single-phase grounding cause diagnosis device for distribution networks based on confidence and fuzziness assessment, characterized in that, include: The acquisition module is used to acquire the original diagnostic data corresponding to the waveform under test. The original diagnostic data includes the single-phase grounding cause and similarity distance of each of the multiple neighboring sample waveforms corresponding to the waveform under test. The similarity distance represents the similarity distance between the corresponding neighboring sample waveform and the waveform under test. The first determining module is used to determine the confidence level of the original diagnostic data based on the similarity distance between the waveforms of the multiple nearest neighbor samples in the original diagnostic data. The second determining module is used to determine multiple candidate causes corresponding to the waveform under test based on the single-phase grounding causes corresponding to the multiple neighboring sample waveforms, wherein the multiple candidate causes are different; The third determining module is used to determine the occurrence frequency of each of the multiple candidate causes based on the number of waveforms of the multiple neighboring samples; The fourth determining module is used to determine the target ambiguity corresponding to the original diagnostic data based on the occurrence frequency of each of the multiple candidate causes and the preset cause label semantic association matrix. The fifth determining module is used to determine the diagnostic result corresponding to the waveform under test based on the confidence level and the target ambiguity.

8. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the non-volatile storage medium to perform the single-phase grounding cause diagnosis method for distribution networks based on confidence and ambiguity assessment as described in any one of claims 1 to 6.

9. A computer device, characterized in that, include: Memory and processor The memory stores computer programs; The processor is configured to execute a computer program stored in the memory, wherein when the computer program is executed, the processor performs the single-phase grounding cause diagnosis method for distribution networks based on confidence and ambiguity assessment as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the single-phase grounding cause diagnosis method for distribution networks based on confidence and ambiguity assessment as described in any one of claims 1 to 6.

Images