Transformer area line loss anomaly detection method based on consistency fuzzy verification

Abstract

The invention discloses a transformer area line loss anomaly detection method based on consistency fuzzy verification, and belongs to the technical field of electric digital data processing. The method comprises the following steps: firstly, constructing a reference archive set and a to-be-analyzed archive set of attribute fields and decision fields with a causal relationship according to an anomaly detection requirement; then judging the advantage or disadvantage relationship between the archive records, dynamically constructing an upper approximate set and a lower approximate set based on the selected relationship type, and taking the difference set as a boundary domain representing logic inconsistency; and finally, according to an application scene, carrying out abnormal score sorting or correct rule comparison on records in the boundary domain, and outputting an abnormal analysis result. According to the method, non-outlier anomalies can be identified, the dependence of a traditional clustering method on the overall distance is overcome, efficient research and judgment can be realized only by a small number of typical samples, the data and computing power requirements are reduced, and the identification precision and the operation and maintenance troubleshooting efficiency of the abnormal mode of the complex archive are improved.

Description

Technical Field

[0001] This invention belongs to the field of electrical digital data processing technology, specifically relating to a method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification. Background Technology

[0002] In the field of power distribution big data analysis, the assessment of line losses in power distribution areas is a crucial link in ensuring the economical operation of the power grid. After meter installation or replacement, key information such as CT / PT configuration, line formulas, and installation / removal times must be accurately entered into the power distribution area archive. The system then calculates theoretical line losses based on real-time collected data and the configuration information in the archive. In actual operation, archive problems such as incorrect comprehensive ratios, incompatible wiring methods, confused affiliations, deviations in caliber formulas, inconsistent meter reading cycles, or inaccurate recording of meter installation and removal times can directly lead to inaccurate calculations of input and output power, thereby causing abnormal management line losses.

[0003] Traditional systems generally lack the ability to automatically verify changes to records, and can only respond passively after line loss anomalies occur. Furthermore, when maintenance personnel investigate, due to a lack of effective technical means, they often need to manually compare every field of all suspected records, or rely on unreliable experience-based judgment, which is not only inefficient but also makes it difficult to guarantee the success rate of error location.

[0004] To improve the efficiency of identifying archival anomalies, those skilled in the art have proposed various solutions for analyzing and investigating abnormal archival records, but all of them have limitations:

[0005] 1. Clustering-based method: Although this method is simple to implement, its mechanism of relying on outlier identification has inherent defects: when only a single field in a multi-field archive record is wrong, its whole may not have a significant outlier distance from normal records, making it difficult to effectively detect such "non-outlier anomalies".

[0006] 2. Rule-based approach: This approach relies on static rules that are pre-set manually or derived statistically. When faced with the diversity of file structures and the dynamic changes in business scenarios, the coverage and adaptability of the rules are often insufficient, affecting the accuracy of analysis and judgment.

[0007] 3. Supervised learning method based on graph network classification: This method is limited by the current situation of inconsistent business structure of power archives and scarce labeled samples, and its practical application scenarios are limited.

[0008] 4. AI-based large model approach: Although this approach improves data processing capabilities, it has high deployment costs and poses a risk of model illusion in business contexts where archival data is highly ambiguous, making it difficult to guarantee reliability.

[0009] In summary, existing technical solutions are insufficient in terms of accuracy and adaptability in detecting power line anomalies, and cannot meet the actual needs of refined management of power distribution area line losses. Summary of the Invention

[0010] This invention proposes a method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification. Its objectives are: first, to accurately identify various types of problematic files, including non-outlier anomalies; second, to dynamically adapt to the diversity of file structures and changes in business scenarios without relying on preset static rules; and third, to effectively assess suspected abnormal files with only a small number of typical samples and without relying on large-scale labeled data and complex models.

[0011] The technical solution of this invention is as follows:

[0012] A method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification includes the following steps:

[0013] Step S1. Construct a reference file set T and a file set U to be analyzed to meet the anomaly detection requirements;

[0014] The reference file set T contains one or more reference file records, and the file set U to be analyzed contains one or more file records to be analyzed. The reference file records and the file records to be analyzed have the same field structure. The field structure contains several attribute fields and one decision field. There is a causal relationship between the attribute fields and the decision field. The attribute fields are the "cause" and the decision field is the "effect".

[0015] Step S2. Perform fuzzy analysis on the consistency of the reference archive set T and the archive set U to be analyzed to obtain the consistency boundary region. The boundary region is a subset of the archive set U to be analyzed and is used to characterize the uncertain state of the two archive records in the logical relationship between attribute fields and decision fields.

[0016] Step S3. Perform anomaly detection analysis on the archive records to be analyzed in the boundary domain obtained in Step S2, and output the analysis results.

[0017] As a further improvement to the above-mentioned method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification, the consistency fuzzy judgment in step S2 includes:

[0018] Step S21. Traverse all the records to be analyzed, and perform the following steps S22-S23 for each record to be analyzed;

[0019] Step S22. Perform fuzzy relation judgments on the current file record to be analyzed and all reference file records respectively. If each fuzzy relation judgment meets the above approximation judgment condition, then add the current file record to be analyzed to the above approximation set.

[0020] Step S23. Perform fuzzy relation judgments on the current file record to be analyzed and all reference file records respectively. If each fuzzy relation judgment meets the next approximation judgment condition, then add the current file record to be analyzed to the next approximation set.

[0021] Step S24. After the traversal is completed, the upper approximation set and the lower approximation set are obtained, and the difference between the upper approximation set and the lower approximation set is taken as the boundary domain.

[0022] As a further improvement to the above-mentioned method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification, for any record to be analyzed... and any reference file record The method for determining the fuzzy relationship between the two is as follows: calculate the records to be analyzed separately. and reference records The fuzzy relation degree between the same field values ​​is determined, and then the relationship between the fuzzy relation degree of each field and the preset threshold in the corresponding upper or lower approximation judgment condition is used to determine whether the current fuzzy relation judgment meets the upper or lower approximation judgment condition.

[0023] As a further improvement to the above-mentioned method for detecting abnormal line loss in transformer areas based on consistency fuzzy verification: the fuzzy relationship judgment includes two types: dominant relationship judgment and inferior relationship judgment; the fuzzy relationship judgment types performed in steps S22 and S23 must be the same.

[0024] As a further improvement to the above-mentioned method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification, for any file to be analyzed... and any reference file :

[0025] The method for determining dominance relationships is as follows:

[0026] ;

[0027] In the above equation, the condition on the right-hand side is satisfied if and only if the condition on the right-hand side is satisfied. The assessment of the dominant relationship is correct; , These are the collections of text-type attribute fields and the collection of numeric attribute fields in the archive records, respectively; when hour, and These represent the files to be analyzed. Attribute fields vectors and reference files Attribute fields The vector; when hour, and These represent the files to be analyzed. Attribute fields Numerical values ​​and reference files Attribute fields The value; express and Fuzzy similarity, express and fuzzy order membership degree; and These represent the files to be analyzed. Values ​​and reference files for decision fields The value of the decision field in the middle. express and fuzzy order membership degree; , , These are the preset attribute fuzzy similarity advantage threshold, attribute fuzzy order membership advantage threshold, and decision fuzzy order membership advantage threshold in the current upper approximation judgment or lower approximation judgment, respectively.

[0028] The method for determining a disadvantageous relationship is as follows:

[0029] ;

[0030] In the above equation, the condition on the right-hand side is satisfied if and only if the condition on the right-hand side is satisfied. The assessment of a disadvantageous relationship is valid. , , These are the preset thresholds for attribute fuzzy similarity disadvantage, attribute fuzzy order membership disadvantage, and decision fuzzy order membership disadvantage in the current upper or lower approximation judgment.

[0031] As a further improvement to the above-mentioned method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification, and Fuzzy similarity The calculation method is as follows:

[0032] .

[0033] As a further improvement to the above-mentioned method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification, for any two values... and Fuzzy order membership degree The calculation method is as follows:

[0034] ;

[0035] In the above formula, It is a parameter that controls the fuzzy order relation.

[0036] As a further improvement to the above-mentioned method for detecting transformer line loss anomalies based on consistency fuzzy verification, if the current anomaly detection requirement is to investigate the current archive records in the system, then:

[0037] In step S1, the reference file set T consists of typical correct file records selected by manual screening, and the file set U to be analyzed consists of the file records currently to be investigated.

[0038] In step S3, the anomaly score of each record to be analyzed in the boundary domain is calculated, and the records are sorted according to the anomaly score. The records with the highest scores are marked as potential anomaly records.

[0039] As a further improvement to the above-mentioned method for detecting abnormal line loss in transformer areas based on consistency fuzzy verification, for the boundary domain... Any file record to be analyzed in The abnormal score is calculated as follows:

[0040] ;

[0041] In the above formula, , These are the collections of text-type attribute fields and the collection of numeric attribute fields in the archive records, respectively; when hour, and These represent the files to be analyzed. Attribute fields Vectors and files to be analyzed Attribute fields The vector; when hour, and These represent the files to be analyzed. Attribute fields The numerical values ​​and the files to be analyzed Attribute fields The value; express and Fuzzy similarity, express and The fuzzy order membership degree.

[0042] As a further improvement to the above-mentioned method for detecting line loss anomalies in transformer substations based on consistency fuzzy verification, if the current anomaly detection requirement is that an anomaly has occurred in the system, and the system has been identified as having archive records that may be associated with the anomaly, then:

[0043] In step S1, the reference file set T consists of the file records that may be associated with the anomaly, and the file set U to be analyzed consists of all file records in the system.

[0044] In step S3, the file records to be analyzed in the boundary domain are correct but deviate from the reference file set T. The file records in the boundary domain are used as a reference benchmark and compared with the file records that may be associated with the anomaly. Based on the patterns shown by the file records in the boundary domain, it is determined whether the file records that may be associated with the anomaly are abnormal.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1. This invention constructs a boundary domain through fuzzy consistency analysis of archives, effectively identifying non-outlier anomalies that are difficult to detect using traditional clustering methods. Due to the fuzzy relationship judgment mechanism between attribute fields and decision fields, even if only a few fields in an archive record have logical deviations, they can be filtered out as long as their overall logical relationship with the reference archive falls within the uncertain state represented by the boundary domain. This method overcomes the dependence of traditional outlier detection on overall distance differences and is more closely aligned with the actual business scenario in power archives where line loss anomalies are caused by errors in a single field.

[0047] 2. Compared to methods relying on static rules, this invention utilizes typical samples to construct a reference set and dynamically adapts to the diversity of file structures and business changes through fuzzy relation thresholds. Since no fixed rules need to be preset, the system only needs to update the reference file set to maintain the effectiveness of its judgment when facing scenarios such as new CT / PT configurations or changes in line formulas. This avoids misjudgments or omissions caused by insufficient rule coverage and improves the system's adaptability in dynamic business environments.

[0048] 3. This invention requires only a small number of typical samples as prior knowledge and generates the boundary domain through the difference operation of the upper and lower approximate sets, without relying on large-scale labeled data or complex model training. This feature effectively overcomes the limitations of scarce and inconsistent labeled samples in power archive operations on supervised learning methods and the application of large models, significantly reducing the cost of data preparation and model deployment while ensuring the reliability of anomaly detection.

[0049] 4. By introducing fuzzy similarity calculation for textual fields and fuzzy order membership evaluation for numerical fields, this invention achieves refined comparison of multi-dimensional features of archival records. This comprehensive judgment mechanism, which integrates textual semantics and numerical logic, can more accurately capture the inherent consistency relationships between fields and improve the accuracy of identifying complex archival anomaly patterns.

[0050] 5. This method allows for flexible configuration of the reference set and the set to be analyzed according to different investigation needs. Whether it is proactive screening of current system files or retrospective analysis of related files that have experienced anomalies, targeted judgment can be achieved by adjusting the set construction strategy. It flexibly adapts to various operation and maintenance scenarios and has high practical value.

[0051] 6. This method achieves quantitative assessment and ranking of potential anomaly files by designing an anomaly score calculation model based on the similarity and order relationship between records within the boundary domain. This mechanism provides maintenance personnel with clear investigation priorities, effectively narrowing the scope of manual verification and improving the efficiency and accuracy of anomaly localization. Detailed Implementation

[0052] The technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0053] First, we will explain the common fields and their meanings in the power distribution area archives. The power distribution area archives record various information related to line loss calculation, mainly including meter number, comprehensive multiplier (CT / PT configuration), wiring direction, superior authority (e.g., user, distribution area, substation, line), accounting formula, meter reading cycle, meter installation / removal time, and meter clock. Among these, the meter number serves as the primary key for uniquely identifying the record; an incorrect comprehensive multiplier will lead to deviations in energy consumption calculation; an incorrect wiring direction will affect the use of forward / reverse active power; an incorrect superior authority archive will cause a mismatch between input and output energy; an incorrect accounting formula will also cause a mismatch between input and output line loss; abnormal energy consumption (e.g., missing or zero data) will directly lead to abnormal line loss; time errors in meter replacement records may cause energy deviations during the replacement period; meter clock deviations will cause misalignment of calculation cycles. All archive information that affects energy consumption calculation or line loss calculation falls within the scope of this method.

[0054] To facilitate subsequent analysis, all records related to the testing are first combined into a wide table based on business relationships. A wide table integrates multiple fields from different data tables into a single table according to business logic. Its fields include all attribute fields and decision fields that need to be analyzed. There are generally multiple attribute fields, while there is usually only one decision field. Attribute fields are a aggregation of record fields, such as comprehensive multiplier, superior level, and installation / removal time. Decision fields prioritize numerical fields such as electricity consumption and installation time, which are usually result data obtained from record association. There should be a causal relationship between attribute fields and decision fields; that is, attribute fields act as the "cause," and decision fields act as the "effect." For example, earlier installation time and higher multiplier usually mean a larger historical cumulative electricity consumption; a sudden decrease in electricity consumption can lead to abnormal line loss. In this case, electricity consumption can be used as a decision field, while fields affecting electricity consumption, such as installation time and multiplier, are used as attribute fields. In actual analysis, the fields involved in the analysis can be flexibly adjusted. For example, if 12 fields are selected for an analysis, 10 of them can be used as attribute fields and 2 as decision fields. The analysis wide table can then be constructed in the form of 10+1 or 11+1.

[0055] Preprocess the fields in the wide table. Numerical fields (including time fields converted to timestamps) need to be standardized to eliminate the influence of units and ensure that the data is on the same scale. Text fields need to be vectorized into vector representations that can be used for similarity calculation.

[0056] The specific implementation steps of the transformer substation line loss anomaly detection method based on consistency fuzzy verification include:

[0057] Step S1. Construct a reference file set T and a file set to be analyzed U to meet the anomaly detection requirements.

[0058] The reference archive set T contains one or more reference archive records, and the archive set U to be analyzed contains one or more archive records to be analyzed. The reference archive records and the archive records to be analyzed have the same field structure, which includes several attribute fields and one decision field. There is a causal relationship between the attribute fields and the decision field, with the attribute fields representing the "cause" and the decision field representing the "effect".

[0059] Step S2. Perform fuzzy analysis on the consistency of the reference archive set T and the archive set U to be analyzed to obtain the consistency boundary domain. The boundary domain is a subset of the archive set U to be analyzed and is used to characterize the uncertain state of the two archive records in the logical relationship between attribute fields and decision fields.

[0060] The consistency fuzzy assessment in step S2 includes:

[0061] Step S21. Traverse all the records to be analyzed, and perform the following steps S22-S23 for each record to be analyzed.

[0062] Step S22. Perform fuzzy relation judgments on the current file record to be analyzed and all reference file records respectively. If each fuzzy relation judgment meets the above approximation judgment condition, then add the current file record to be analyzed to the above approximation set.

[0063] Step S23. Perform fuzzy relation judgments on the current file record to be analyzed and all reference file records respectively. If each fuzzy relation judgment meets the next approximation judgment condition, then add the current file record to be analyzed to the next approximation set.

[0064] For any record to be analyzed and any reference file record The method for determining the fuzzy relationship between the two is as follows: calculate the records to be analyzed separately. and reference records The fuzzy relation degree between the same field values ​​is determined, and then the relationship between the fuzzy relation degree of each field and the preset threshold in the corresponding upper or lower approximation judgment condition is used to determine whether the current fuzzy relation judgment meets the upper or lower approximation judgment condition.

[0065] It should be noted that the fuzzy relationship judgment includes two types: dominant relationship judgment and suboptimal relationship judgment; the fuzzy relationship judgment types performed in steps S22 and S23 must be the same. The choice between a dominant or suboptimal relationship depends on the "positive or negative correlation" between the attribute field and the decision field in terms of business logic. A positive correlation means that when the value of the attribute field increases, the value of the decision field also tends to increase; in this case, a dominant relationship is chosen, and vice versa.

[0066] Specifically, for any file to be analyzed and any reference file The method for determining dominance relationships is as follows:

[0067] ;

[0068] In the above equation, the condition on the right-hand side is satisfied if and only if the condition on the right-hand side is satisfied. The assessment of the dominant relationship is correct; , These are the collections of text-type attribute fields and the collection of numeric attribute fields in the archive records, respectively; when hour, and These represent the files to be analyzed. Attribute fields vectors and reference files Attribute fields The vector; when hour, and These represent the files to be analyzed. Attribute fields Numerical values ​​and reference files Attribute fields The value; express and Fuzzy similarity, express and fuzzy order membership degree; and These represent the files to be analyzed. Values ​​and reference files for decision fields The value of the decision field in the middle. express and fuzzy order membership degree; , , These are the preset attribute fuzzy similarity advantage threshold, attribute fuzzy order membership advantage threshold, and decision fuzzy order membership advantage threshold in the current upper approximation judgment or lower approximation judgment, respectively.

[0069] and Fuzzy similarity The calculation method is as follows:

[0070] .

[0071] The range of fuzzy similarity is [0,1], and the closer it is to 1, the more similar it is.

[0072] For any two values and Fuzzy order membership degree The calculation method is as follows:

[0073] ;

[0074] In the above formula, It is a parameter that controls the fuzzy order relation, and is usually taken as 0.9 to 1.0.

[0075] The method for determining a disadvantageous relationship is as follows:

[0076] ;

[0077] In the above equation, the condition on the right-hand side is satisfied if and only if the condition on the right-hand side is satisfied. The assessment of a disadvantageous relationship is valid. , , These are the preset thresholds for attribute fuzzy similarity disadvantage, attribute fuzzy order membership disadvantage, and decision fuzzy order membership disadvantage in the current upper or lower approximation judgment.

[0078] The upper approximate set uses a more lenient threshold, for example, in the judgment of dominance relations. , , In the judgment of disadvantageous relationships, take , , This indicates that the records may be similar or identical; the lower approximation set uses a stricter threshold, such as in the dominance relation judgment. , , In the judgment of disadvantageous relationships, take , , This indicates that the records in the archives are similar and consistent. The boundary region is the difference between the upper approximation set and the lower approximation set, where the records are in an uncertain state and require further analysis.

[0079] Step S24. After the traversal is completed, the upper approximation set and the lower approximation set are obtained, and the difference between the upper approximation set and the lower approximation set is taken as the boundary domain.

[0080] Step S3. Perform anomaly detection analysis on the archive records to be analyzed in the boundary domain obtained in Step S2, and output the analysis results.

[0081] The first scenario involves proactive screening of the system's current archives. In this case, the reference archive set T in step S1 consists of manually selected typical correct archive records, while the archive set U to be analyzed consists of the archive records currently to be investigated. After obtaining the boundary region in step S3, the anomaly score for each archive record x to be analyzed is calculated:

[0082] ;

[0083] in, For the boundary domain, and These are sets of text and numeric attribute fields, respectively. The score reflects the degree of inconsistency between the record being analyzed and other records in the boundary region regarding attribute fields; a higher score indicates a greater likelihood of an anomaly. Finally, records are sorted according to their anomaly scores, and the top-ranked records are marked as potentially anomaly records for focused manual review.

[0084] The second scenario involves retrospective analysis of related files that have already experienced line loss anomalies. In this case, the reference file set T in step S1 consists of file records that the system has automatically locked and that may be associated with the anomaly, while the file set U to be analyzed consists of all file records in the system. After obtaining the boundary region in step S3, the file records to be analyzed within it are correct records that deviate from the reference file set T. These records are used as a reference benchmark and compared with the anomaly-related file records. The patterns exhibited by the correct records in the boundary region (e.g., the metering multiplier in a certain area is typically 0.1, while the multiplier for anomaly records is 0.01) are used to investigate whether there are any problems with the anomaly-related file records.

[0085] It should be noted that, as will be apparent to those skilled in the art, the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics thereof. The scope of the present invention is defined by the claims rather than the foregoing description.

Claims

1. A method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification, characterized in that, Includes the following steps: Step S1. Construct a reference file set T and a file set U to be analyzed to meet the anomaly detection requirements; The reference archive set T contains one or more reference archive records, and the archive set U to be analyzed contains one or more archive records to be analyzed. The reference archive records and the archive records to be analyzed have the same field structure. The field structure contains several attribute fields and one decision field. There is a causal relationship between the attribute fields and the decision field. The attribute fields are the "cause" and the decision field is the "effect". Step S2. Perform fuzzy analysis on the consistency of the reference archive set T and the archive set U to be analyzed to obtain the consistency boundary region. The boundary region is a subset of the archive set U to be analyzed and is used to characterize the uncertain state of the two archive records in the logical relationship between attribute fields and decision fields. Step S3. Perform anomaly detection analysis on the archive records to be analyzed in the boundary domain obtained in Step S2, and output the analysis results.

2. The method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification as described in claim 1, characterized in that, The consistency fuzzy assessment in step S2 includes: Step S21. Traverse all the records to be analyzed, and perform the following steps S22-S23 for each record to be analyzed; Step S22. Perform fuzzy relation judgments on the current file record to be analyzed and all reference file records respectively. If each fuzzy relation judgment meets the above approximation judgment condition, then add the current file record to be analyzed to the above approximation set. Step S23. Perform fuzzy relation judgments on the current file record to be analyzed and all reference file records respectively. If each fuzzy relation judgment meets the next approximation judgment condition, then add the current file record to be analyzed to the next approximation set. Step S24. After the traversal is completed, the upper approximation set and the lower approximation set are obtained, and the difference between the upper approximation set and the lower approximation set is taken as the boundary domain.

3. The method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification as described in claim 2, characterized in that, For any record to be analyzed and any reference file record The method for determining the fuzzy relationship between the two is as follows: calculate the records to be analyzed separately. and reference archives The fuzzy relation degree between the same field values ​​is determined, and then the relationship between the fuzzy relation degree of each field and the preset threshold in the corresponding upper or lower approximation judgment condition is used to determine whether the current fuzzy relation judgment meets the upper or lower approximation judgment condition.

4. The method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification as described in claim 2 or 3, characterized in that: The fuzzy relationship judgment includes two types: dominant relationship judgment and disadvantageous relationship judgment; the fuzzy relationship judgment types performed in steps S22 and S23 must be the same.

5. The method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification as described in claim 4, characterized in that, For any file to be analyzed and any reference file : The method for determining dominance relationships is as follows: ； In the above equation, the condition on the right-hand side is satisfied if and only if the condition on the right-hand side is satisfied. The assessment of the dominant relationship is correct; , These are the collections of text-type attribute fields and the collection of numeric attribute fields in the archive records, respectively; when hour, and These represent the files to be analyzed. Attribute fields vectors and reference files Attribute fields The vector; when hour, and These represent the files to be analyzed. Attribute fields Numerical values ​​and reference files Attribute fields The value; express and Fuzzy similarity, express and fuzzy order membership degree; and These represent the files to be analyzed. Values ​​and reference files for decision fields The value of the decision field in the middle. express and fuzzy order membership degree; , , These are the preset attribute fuzzy similarity advantage threshold, attribute fuzzy order membership advantage threshold, and decision fuzzy order membership advantage threshold in the current upper approximation judgment or lower approximation judgment, respectively. The method for determining a disadvantageous relationship is as follows: ； In the above equation, the condition on the right-hand side is satisfied if and only if the condition on the right-hand side is satisfied. The assessment of a disadvantageous relationship is valid. , , These are the preset thresholds for attribute fuzzy similarity disadvantage, attribute fuzzy order membership disadvantage, and decision fuzzy order membership disadvantage in the current upper or lower approximation judgment.

6. The method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification as described in claim 5, characterized in that, and Fuzzy similarity The calculation method is as follows: 。 7. The method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification as described in claim 5, characterized in that, For any two values and Fuzzy order membership degree The calculation method is as follows: ； In the above formula, It is a parameter that controls the fuzzy order relation.

8. The method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification as described in claim 1, characterized in that, If the current anomaly detection requirement is to check the current archive records in the system, then: In step S1, the reference file set T consists of typical correct file records selected by manual screening, and the file set U to be analyzed consists of the file records to be investigated. In step S3, the anomaly score of each record to be analyzed in the boundary domain is calculated, and the records are sorted according to the anomaly score. The records with the highest scores are marked as potential anomaly records.

9. The method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification as described in claim 8, characterized in that, For the boundary domain Any file record to be analyzed in The abnormal score is calculated as follows: ； In the above formula, , These are the collections of text-type attribute fields and the collection of numeric attribute fields in the archive records, respectively; when hour, and These represent the files to be analyzed. Attribute fields Vectors and files to be analyzed Attribute fields The vector; when hour, and These represent the files to be analyzed. Attribute fields The numerical values ​​and the files to be analyzed Attribute fields The value; express and Fuzzy similarity, express and The fuzzy order membership degree.

10. The method for detecting abnormal line loss in transformer substations based on consistency fuzzy verification as described in claim 1, characterized in that, If the current anomaly detection requirement is that a system anomaly has occurred, and based on the anomaly, file records in the system that may be associated with the anomaly have been identified, then: In step S1, the reference file set T consists of the file records that may be associated with the anomaly, and the file set U to be analyzed consists of all file records in the system. In step S3, the file records to be analyzed in the boundary domain are correct but deviate from the reference file set T. The file records in the boundary domain are used as a reference benchmark and compared with the file records that may be associated with the anomaly. Based on the patterns shown by the file records in the boundary domain, it is determined whether the file records that may be associated with the anomaly are abnormal.

Images