Power grid electric quantity abnormity work order hierarchical verification method and system

Abstract

The invention discloses a power grid electric quantity abnormity work order hierarchical verification method and system, and relates to the technical field of power grid abnormity detection, and the method comprises the steps: obtaining an electric quantity abnormity work order; extracting an abnormal time zone, a user power consumption type and data acquisition equipment information from the electric quantity abnormal work order; checking the work order with the abnormal electric quantity in different time periods, and determining an abnormal sub time zone; according to the abnormal sub-time zone, performing data acquisition and check processing to obtain an equipment measurement error; according to the abnormal sub-time zone, data post-processing flow checking is carried out, and a calculation error is obtained; calculating a total error according to the equipment measurement error and the calculation error; and according to the equipment measurement error, the calculation error and the total error, determining a main hierarchy abnormal factor and a hierarchy abnormal factor. According to the method, different levels of abnormal factors can be determined in combination with various errors, so that layered verification is realized, and the accuracy and the reliability are improved.

Description

Technical Field

[0001] This invention relates to the field of power grid anomaly detection technology, and in particular to a hierarchical verification method and system for power grid power anomaly work orders. Background Technology

[0002] In the operation of modern power systems, ensuring the stability of power supply and the accuracy of metering are crucial. With the rapid development of smart grids, the amount of operational data in power systems has increased dramatically, and user electricity consumption patterns have become more complex. In this context, the detection and handling of abnormal power consumption has become a key link in ensuring the safe and stable operation of the power grid. Traditional methods typically rely on manual inspections, simple rule-based filtering, or single calculation programs. However, when conducting detailed analysis of multi-source heterogeneous data, traditional methods struggle to analyze data characteristics at multiple levels, easily leading to missed or false anomalies, resulting in low accuracy and reliability in data verification.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main objective of this invention is to propose a hierarchical verification method and system for power grid power anomaly work orders. This method can combine multiple errors to determine the anomaly factors at different levels, thereby achieving hierarchical verification and improving accuracy and reliability.

[0005] On one hand, embodiments of the present invention provide a hierarchical verification method for power grid power anomaly work orders, including the following steps: Get abnormal power consumption work orders; Extract the abnormal time zone, user power consumption type, and data acquisition device information from the power consumption anomaly work order; Based on the abnormal time zone, the user's electricity consumption type, and the data acquisition equipment information, the abnormal electricity work order is checked in different time periods to determine the abnormal sub-time zone; Based on the abnormal sub-time zone, data acquisition and verification processing is performed to obtain the equipment measurement error; Based on the abnormal sub-time zone, the data post-processing procedure is checked to obtain the calculation error; Calculate the total error based on the equipment measurement error and the calculation error; Based on the equipment measurement error, the calculation error, and the total error, the primary level anomaly factors and the secondary level anomaly factors are determined.

[0006] On the other hand, embodiments of the present invention provide a hierarchical verification system for power grid power anomaly work orders, including: The work order acquisition module is used to acquire work orders with abnormal power consumption. The data extraction module is used to extract abnormal time zone, user electricity consumption type and data acquisition device information from the power abnormality work order; The time-segmented verification module is used to perform time-segmented verification on the abnormal power consumption work order based on the abnormal time zone, the user's power consumption type, and the data acquisition equipment information, and to determine the abnormal sub-time zone; The equipment measurement error determination module is used to perform data acquisition and verification processing based on the abnormal sub-time zone to obtain the equipment measurement error; The calculation error determination module is used to perform data post-processing verification based on the abnormal sub-time zone to obtain the calculation error; The total error calculation module is used to calculate the total error based on the measurement error of the device and the calculation error. The anomaly verification module is used to determine the main-level anomaly factors and the hierarchical anomaly factors based on the equipment measurement error, the calculation error, and the total error.

[0007] The embodiments of this application include at least the following beneficial effects: First, the embodiments of this application obtain power consumption abnormality work orders, then extract abnormal time zones, user power consumption types, and data acquisition equipment information from the power consumption abnormality work orders, and perform time-segmented verification on the power consumption abnormality work orders to determine abnormal sub-time zones. Then, based on the abnormal sub-time zones, data acquisition verification processing is performed to obtain equipment measurement errors. Based on the abnormal sub-time zones, data post-processing flow verification is performed to obtain calculation errors. Based on the equipment measurement errors and calculation errors, the total error is calculated. Finally, based on the equipment measurement errors, calculation errors, and total errors, the main-level abnormal factors and hierarchical abnormal factors are determined. This allows for the combination of multiple errors to determine abnormal factors at different levels, thereby achieving hierarchical verification and improving accuracy and reliability.

[0008] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description and the drawings. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0010] Figure 1 This is a flowchart of a hierarchical verification method for abnormal power grid work orders according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a hierarchical verification system for abnormal power grid work orders according to an embodiment of the present invention. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments.

[0012] In related technologies, ensuring the stability of power supply and the accuracy of metering are crucial in the operation of modern power systems. With the rapid development of smart grids and renewable energy, the amount of operational data in power systems has increased dramatically, and user electricity consumption patterns have become increasingly complex. Against this backdrop, the detection and handling of abnormal power consumption has become a key link in ensuring the safe and stable operation of the power grid. Traditional anomaly detection methods often rely excessively on manual inspections, simple rule-based filtering, or single calculation programs. These methods struggle to provide sufficiently accurate and in-depth insights when faced with the demanding analysis requirements of multi-source heterogeneous data environments.

[0013] Specifically, there are several significant challenges in the hierarchical verification methods for power grid electricity anomaly work orders. For example, existing technologies often use fixed comparison limits or static rule sets for the initial screening of electricity bill anomaly work orders. However, these methods fail to fully consider key factors such as the characteristics of user electricity consumption behavior over time, differences between different user categories, and the status of the data acquisition equipment itself. This results in overly coarse granularity in anomaly localization, making it difficult to effectively distinguish between merely temporary fluctuations in electricity consumption and actual equipment failures or abnormal user behavior. Furthermore, power grid data itself has strong continuous characteristics and is often accompanied by various external interferences. Traditional methods are significantly inadequate in conducting cross-time period anomaly correlation analysis and multi-level error tracing, which easily leads to missed or misjudged anomalies, increasing power grid operation and maintenance costs and introducing potential safety risks.

[0014] In power grid power anomaly detection systems, facing a variety of complex factors such as multi-source heterogeneous data environments, the temporal dynamics of user electricity consumption behavior, differences in user categories, and the status of acquisition equipment, it is necessary to overcome the limitations of existing methods that attribute anomalies to a single factor. A mechanism should be established that can collaboratively analyze multi-dimensional information such as anomaly time zones, user electricity consumption characteristics, and acquisition equipment accuracy. Furthermore, a layered and refined processing of the anomaly verification process should be implemented, breaking down the total error into sub-items such as calculation errors and equipment measurement errors, tracing the primary and secondary anomaly factors layer by layer, thereby accurately locating the root cause of the anomaly, improving the efficiency of anomaly work order processing, and enhancing the diagnostic and early warning capabilities of power grid operation status.

[0015] In light of this, this application lays the foundation for subsequent verification by obtaining power consumption anomaly work orders and extracting key information. Next, time-segmented verification is conducted to refine the initial anomaly time zones into sub-time zones, thereby focusing the analysis on the time periods with the most significant power consumption anomalies, avoiding ineffective analysis of irrelevant data, and improving the efficiency and accuracy of the verification. After determining the anomaly sub-time zones, data acquisition verification and post-processing verification are performed in parallel, assessing potential errors from both hardware and software perspectives. Data acquisition verification focuses on identifying equipment measurement errors, while post-processing verification focuses on identifying calculation errors. This dual-track parallel approach comprehensively covers the potential sources of power consumption anomalies. Subsequently, equipment measurement errors and calculation errors are combined to calculate the total error. Finally, based on the relative magnitude and characteristics of equipment measurement errors, calculation errors, and the total error, primary-level and tiered anomaly factors are determined. This tiered verification mechanism enables power grid maintenance personnel to clearly understand the root cause of the anomaly and take targeted measures. For example, if the error is due to equipment measurement, equipment repair or replacement may be necessary; if the error is due to calculation, adjustments to data processing logic or billing parameters may be required. This application enables accurate verification of power grid power anomaly work orders, improving accuracy and reliability.

[0016] The embodiments of this application will be explained in detail below with reference to the accompanying drawings: Figure 1 This is an optional flowchart of a hierarchical verification method for abnormal power grid electricity work orders provided in an embodiment of this application. Figure 1 The method may include, but is not limited to, steps S101 to S107.

[0017] Step S101: Obtain a power abnormality work order; Step S102: Extract the abnormal time zone, user power consumption type, and data acquisition device information from the power consumption anomaly work order; Step S103: Based on the abnormal time zone, user electricity consumption type, and data acquisition equipment information, conduct time-segmented verification of abnormal electricity work orders to determine the abnormal sub-time zone; Step S104: Based on the abnormal sub-time zone, perform data acquisition and verification processing to obtain the equipment measurement error; Step S105: Based on the abnormal sub-time zone, perform data post-processing flow verification to obtain the calculation error; Step S106: Calculate the total error based on the equipment measurement error and calculation error; Step S107: Based on the equipment measurement error, calculation error, and total error, determine the main-level abnormal factors and the hierarchical abnormal factors.

[0018] Steps S101 to S107 shown in the embodiments of this application can combine multiple errors to determine abnormal factors at different levels, so as to achieve hierarchical verification and improve accuracy and reliability.

[0019] In some embodiments, steps S101-S107 may first involve obtaining an abnormal electricity consumption work order. For example, this can be automatically generated by the power grid monitoring system; when the system detects that a user's electricity consumption data deviates from a preset threshold, the work order is automatically triggered. Alternatively, a user may report an abnormal electricity bill through customer service channels, and a work order may be generated after manual verification. Furthermore, potential anomalies can be identified and work orders generated through regular data quality inspections. It is understood that an abnormal electricity consumption work order refers to a report or task generated by the power grid monitoring system or user feedback, indicating an anomaly in the electricity consumption data of a specific area or user. This anomaly may manifest as a sudden increase or decrease in electricity consumption, or a continuous deviation from the normal range.

[0020] Then, the abnormal time zone, user electricity consumption type, and data acquisition device information are extracted from the electricity consumption anomaly work order. This can be achieved through automated parsing of the work order text or structured data fields. For example, the time period in which the anomaly occurred can be directly read from a specific field in the work order as the abnormal time zone, the user's electricity consumption type can be obtained from the user's profile, and detailed information about the data acquisition device associated with that user can be queried from the device management system. Alternatively, if the work order information is incomplete, the operator can be prompted to manually enter it or supplement it by querying the relevant database. It is understood that the abnormal time zone refers to the time period during which the electricity consumption data is abnormal, as specified in the electricity consumption anomaly work order, such as 2:00 AM to 4:00 AM on a certain day. The user's electricity consumption type refers to the nature of the user's electricity consumption, such as residential users, commercial users, industrial users, etc. Different types of users have different electricity consumption patterns and load characteristics. The data acquisition device information refers to the relevant information of the equipment used to collect electricity consumption data, such as the model, installation location, and communication status of the smart meter and the acquisition terminal.

[0021] Then, based on the abnormal time zone, user electricity consumption type, and data acquisition equipment information, the electricity consumption anomaly work orders are checked by time period to determine the abnormal sub-time zones, aiming to further refine the initial abnormal time zones. For example, based on user electricity consumption type (e.g., industrial users typically consume more electricity during the day on weekdays, while residential users consume more electricity at night and on weekends) and historical operating data of the data acquisition equipment, a more granular analysis of the electricity consumption data within the abnormal time zone can be performed. If the abnormal time zone is a day, it can be divided into multiple hourly or half-hourly segments, and the deviation of the electricity consumption data in each sub-segment from the normal pattern can be checked one by one. For example, the deviation rate of the electricity consumption data in each sub-segment from the historical data of the same period, or the deviation rate from the average electricity consumption of similar users, can be calculated to identify the abnormal sub-time zone with the most significant electricity consumption anomaly.

[0022] Based on the abnormal time zone, data acquisition and verification processing is performed to obtain the equipment measurement error, aiming to assess the accuracy of the data acquisition equipment itself. For example, the operating status log of the data acquisition equipment in the abnormal time zone can be checked to see if there are events such as communication interruptions, voltage fluctuations, or abnormal temperatures that may affect measurement accuracy. The raw metering data in the abnormal time zone can also be compared with auxiliary reference data from nearby areas or similar equipment to identify potential deviations in equipment measurements. For example, if a smart meter's reading in the abnormal time zone is significantly lower than the average reading of its neighboring meters or similar meters without a reasonable explanation, there may be equipment measurement error.

[0023] Based on the abnormal time zone, the data post-processing workflow is checked to obtain calculation errors. This aims to examine the entire processing chain of electricity data from collection to the final billing result or anomaly judgment flag. For example, it can check whether there are logical errors or improper parameter configurations in the process of data aggregation, data filtering, and billing rule application of the raw metering data in the abnormal time zone. For instance, if a billing rule is incorrectly applied in the abnormal time zone, leading to abnormal electricity calculation results, a calculation error will occur.

[0024] The total error is calculated based on the equipment measurement error and the calculation error. A simple weighted sum can be used to calculate these errors, or a more complex statistical model can be employed for a comprehensive assessment to reflect the overall severity of the power anomaly. For example, different weights can be assigned based on factors such as the absolute value of the error, its duration, and the scope of its impact.

[0025] Finally, based on the equipment measurement error, calculation error, and total error, primary-level and tiered anomaly factors are identified. This aims to analyze errors from multiple dimensions to pinpoint the root cause of the anomaly. For example, if equipment measurement error dominates the total error, the primary-level anomaly factor might be identified as equipment measurement error; if calculation error dominates, it might be identified as calculation error. If both contribute equally, further analysis of user behavior patterns may be needed to determine if there is a sudden change in user behavior. After identifying the primary-level anomaly factor, it can be further refined. For example, if the primary-level anomaly factor is equipment measurement error, tiered anomaly factors might be communication failures, sensor failures, or calibration deviations. It is understood that primary-level anomaly factors refer to the most important and fundamental cause of power abnormalities, such as equipment measurement errors, calculation errors, or sudden changes in user behavior. Tiered anomaly factors refer to specific anomaly causes further refined below the primary-level anomaly factors, such as communication failures or billing parameter configuration failures.

[0026] Through the above technical solutions, this embodiment effectively solves the problems of rough anomaly location, confusing error sources, and low fault diagnosis efficiency of traditional methods when facing complex power grid environments by conducting refined time-segmented verification, dual-track parallel error tracing, and multi-dimensional anomaly factor analysis. It provides strong technical support for the stable operation and precise management of the power grid.

[0027] In some embodiments, in step S104, data acquisition and verification processing is performed based on the abnormal sub-time zone to obtain the device measurement error, which may include, but is not limited to, the following steps: Extract raw metering data and auxiliary reference data from the abnormal sub-time zone from the power grid data storage center; Perform an integrity check on the original measurement data and obtain the integrity check results; The original measurement data were compared for consistency to obtain the consistency comparison results. The original measurement data and auxiliary reference data are compared to obtain the comparison results; Based on the integrity check results, consistency comparison results, and comparison results, the equipment measurement error is determined.

[0028] In some embodiments, raw metering data and auxiliary reference data for the abnormal sub-time zone can be extracted from the power grid data storage center first. For example, raw electricity metering data corresponding to a specific abnormal sub-time zone and reference data for auxiliary analysis and comparison can be obtained from the power grid's core data storage system or related databases. Raw metering data refers to real-time or historical data such as electricity, voltage, and current directly collected by front-end devices such as smart meters and sensors. Auxiliary reference data may include equipment calibration records, environmental monitoring data, power grid topology information, historical load curves, normal operating parameters of similar equipment, and other relevant business data. Its purpose is to provide comprehensive and multi-dimensional data support for subsequent equipment measurement error analysis.

[0029] Then, an integrity check is performed on the raw measurement data to obtain the integrity check results. This check aims to verify whether the extracted raw measurement data is missing, damaged, or discontinuous. For example, it can check for issues such as timestamp jumps, missing data points, incomplete data records, or incorrect data formats in the data series. The purpose of the integrity check is to ensure that the data used for subsequent analysis is complete and reliable, avoiding misjudgments due to problems with the data itself.

[0030] Next, a consistency comparison is performed on the original metering data to obtain the consistency comparison results. Cross-validation can be performed on multi-source data collected within the same time period, from the same device, or from related devices to determine if there are any contradictions or inconsistencies between the data. For example, the consistency of data transmitted from the same meter on different communication channels can be compared, or the physical logic of electricity consumption data between adjacent meters can be compared. The purpose of consistency comparison is to identify errors that may be introduced during data transmission or storage, thereby revealing the reliability of the data source.

[0031] The raw measurement data and auxiliary reference data are compared to obtain the comparison results. Raw measurement data that has undergone integrity checks and consistency comparisons can be compared with preset auxiliary reference data. For example, current measurement data can be compared with historical data from the same period, average data from similar equipment, theoretical calculations, or equipment factory calibration values. The purpose of this comparison is to identify measurement data that deviates significantly from normal patterns or expected values, thereby initially locating potential anomalies.

[0032] Finally, based on the results of the integrity check, consistency comparison, and comparison, the equipment measurement error is determined. For example, if the integrity check reveals missing data, the consistency comparison reveals contradictory data, and there is a significant deviation compared to the auxiliary reference data, then a comprehensive judgment can be made that equipment measurement error exists. The determination of equipment measurement error is based on a comprehensive judgment of multi-dimensional data verification, aiming to accurately identify errors introduced by the equipment itself or the data acquisition process.

[0033] This embodiment employs a multi-dimensional data verification mechanism to systematically identify potential errors introduced during data acquisition. First, raw metering data and auxiliary reference data within abnormal sub-time zones are extracted from the power grid data storage center, providing a necessary data foundation for subsequent verification processing. Second, the raw metering data undergoes an integrity check to eliminate problems caused by missing or corrupted data, ensuring data availability. Next, consistency comparison reveals potential logical contradictions between the same or related data sources, uncovering potential errors in the initial stages of data transmission or processing. Subsequently, comparing the raw metering data with the auxiliary reference data effectively identifies metering data that deviates from normal operating conditions or expected values. Finally, by comprehensively utilizing the integrity check results, consistency comparison results, and comparison results, equipment measurement errors can be comprehensively and accurately determined, providing a reliable basis for subsequent anomaly factor analysis.

[0034] Through the above technical solution, this embodiment enables refined and multi-dimensional verification of equipment measurement errors in power grid anomaly work orders. This embodiment can not only identify potential metering inaccuracies in the data acquisition equipment itself, but also detect errors introduced during data transmission and storage, as well as abnormal data inconsistent with normal operating modes. Therefore, it can significantly improve the accuracy and reliability of equipment measurement error identification, avoid misclassifying equipment faults as other types of anomalies, and thus lay a solid foundation for accurate hierarchical verification of power grid anomaly work orders, improving the level of refined management of power grid operation.

[0035] In some embodiments, in step S105, the data post-processing procedure is checked according to the abnormal sub-time zone to obtain the calculation error, which may include, but is not limited to, the following steps: Extract raw metering data from abnormal sub-time zones from the power grid data storage center; Perform a data aggregation logic check on the raw measurement data to obtain the data aggregation logic check results; The billing rule application is checked on the original metering data to obtain the billing rule application check results. Anomaly detection rule checks are performed on the original measurement data to obtain the anomaly detection rule check results; The calculation error is determined based on the results of the data aggregation logic check, the billing rule application check, and the anomaly judgment rule check.

[0036] In some embodiments, raw metering data within abnormal time zones can be extracted first from the power grid data storage center. This raw metering data, collected directly from metering devices in the power grid and uploaded to the power grid data storage center without any processing, typically includes information such as timestamps, metering point identifiers, instantaneous power, and cumulative electricity consumption. The power grid data storage center is a distributed database system, data warehouse, or any platform capable of storing massive amounts of power grid operation data.

[0037] Then, a data aggregation logic check is performed on the raw metering data to obtain the data aggregation logic check results. This aims to verify whether the aggregation rules followed during the process of collecting electricity data from its initial acquisition to the final generation of reports or billing data are correct. For example, daily electricity consumption and monthly electricity consumption are usually derived from the accumulation of raw metering data at smaller time granularities (such as 15-minute or 30-minute data). This check verifies whether the accumulation logic conforms to preset rules. The data aggregation logic check results can indicate whether there are aggregation errors, such as missing data, duplicate calculations, or errors in the aggregation algorithm.

[0038] The original metering data is then subjected to a billing rule application check to obtain the check results. This verifies whether the power grid system correctly applies the preset billing rules when settling electricity bills with users. These rules may include time-of-use pricing, tiered pricing, power factor adjustment, etc. This check identifies deviations or errors in the application of billing rules by comparing the original metering data with the expected results after applying the billing rules. The billing rule application check results can reveal errors in billing parameter configuration, defects in rule logic implementation, or data mapping problems.

[0039] Anomaly detection rule checks are performed on raw metering data to obtain the results. The aim is to assess whether the internal rules used by the power grid system to identify electricity anomalies are being executed correctly. These rules may be based on historical data analysis, statistical models, or expert experience to determine whether current electricity data deviates from the normal range. This check verifies the effectiveness and accuracy of the rules by simulating or reproducing the anomaly detection process. The results of the anomaly detection rule check may reflect improper anomaly threshold settings, a failed detection model, or errors in the rule execution logic.

[0040] Finally, based on the results of the data aggregation logic check, the billing rule application check, and the anomaly detection rule check, the calculation error is determined. Calculation error refers to the deviation introduced during the process of data processing from raw measurement to final data that can be used for billing or anomaly detection, due to improper data processing logic, rule application, or parameter configuration.

[0041] This embodiment systematically identifies and quantifies calculation errors by meticulously verifying key steps in the data post-processing workflow. Specifically, by extracting raw metering data from the abnormal sub-time zone from the power grid data storage center, a reliable data foundation is provided for subsequent verification work. Subsequently, data aggregation logic checks ensure the accuracy of the aggregation process of electricity data at different time granularities, avoiding calculation deviations caused by aggregation errors. Billing rule application checks focus on verifying whether the power grid system strictly processes electricity data according to the established billing strategy, thereby eliminating errors caused by improper application of billing rules. Anomaly judgment rule checks further verify the soundness and effectiveness of the system's internal logic for identifying anomalies, preventing misjudgments or omissions due to defects in anomaly judgment rules. Therefore, by checking and verifying each key step, potential calculation errors in the data post-processing workflow can be comprehensively and accurately located and quantified.

[0042] Through the above technical solution, this embodiment can delve into specific data post-processing stages such as data aggregation, billing rule application, and anomaly judgment rules, identifying which logical error or improper parameter configuration in any stage led to the abnormal power consumption. This meticulous verification process significantly improves the accuracy and specificity of calculation error identification, providing clear guidance for subsequent fault diagnosis and system optimization, thereby effectively improving the efficiency and reliability of power grid anomaly work orders.

[0043] In some embodiments, in step S105, the data post-processing procedure is checked according to the abnormal sub-time zone to obtain the calculation error, which may include, but is not limited to, the following steps: Extract raw metering data from abnormal sub-time zones from the power grid data storage center; Construct actual operation channels and benchmark simulation channels; The raw metering data is input into the actual operation channel, which is used to perform the first data processing on the raw metering data to obtain the first dataset. The first data processing includes data aggregation, first data filtering and billing rule application. The first dataset includes aggregated electricity data, key intermediate calculation results and anomaly judgment flags. The first data filtering is used to remove noise and high-frequency transient load components. The raw metering data is input into the benchmark simulation channel. The benchmark simulation channel is used to perform a second data processing on the raw metering data to obtain a second dataset. The second data processing includes data aggregation, second data filtering, and billing rule application. The second data filtering is used to remove noise and retain high-frequency transient load components. Perform time alignment on the first and second datasets; Point-by-point difference calculations are performed on the first and second datasets after time alignment to obtain the calculation error.

[0044] In some embodiments, traditional verification methods may struggle to accurately quantify computational errors caused by subtle differences in the data processing flow or specific filtering strategies, especially when it is necessary to distinguish between normal load fluctuations and abnormal data processing results. Their accuracy and robustness may be limited, potentially leading to insufficiently refined assessment of computational errors, which in turn affects the accurate identification and location of subsequent anomalies.

[0045] To this end, raw metering data within the abnormal sub-time zone can be extracted from the power grid data storage center. For example, raw electricity data, collected by metering devices within a specific abnormal sub-time zone without any advanced processing, can be obtained from the power grid's core data warehouse. Furthermore, an actual operation channel and a benchmark simulation channel are constructed to establish two independent simulation environments capable of parallel data processing. The actual operation channel aims to accurately reproduce the current actual data processing logic and parameter configuration in the power grid system, including all data aggregation, filtering, and billing rule applications. The benchmark simulation channel provides an ideal or standard data processing baseline; its processing logic is similar to the actual operation channel, but differs in key filtering strategies. For instance, the benchmark simulation channel may retain certain high-frequency transient load components that were removed in the actual operation channel for comparative analysis.

[0046] The raw metering data is then input into the actual operation channel, where it undergoes initial data processing to obtain a first dataset. Initial data filtering removes noise and high-frequency transient load components. The actual operation channel simulates the power grid system's actual processing of raw metering data. The initial data processing includes data aggregation, which summarizes the raw, fine-grained data according to preset time periods (e.g., 15 minutes, hours, or days); initial data filtering, which removes noise and high-frequency transient load components to obtain smoother, more stable power data, typically achieved through algorithms such as low-pass filters or moving averages; and billing rule application, which calculates the aggregated data based on preset electricity prices, rates, and billing cycles. Therefore, the first dataset contains aggregated power data after the actual processing, key intermediate calculation results generated during processing (e.g., power factor, demand), and anomaly flags to indicate whether the data has been marked as abnormal by the system.

[0047] The original metering data is then input into the benchmark simulation channel. This channel performs a second data processing step on the original metering data to obtain a second dataset. This second data processing includes data aggregation, second data filtering, and billing rule application. The second data filtering removes noise and retains high-frequency transient load components. The benchmark simulation channel also performs data aggregation and billing rule application on the original metering data, but its second data filtering strategy differs from the first. The second data filtering aims to remove noise but retains high-frequency transient load components. This means that the data processing results from the benchmark simulation channel will more closely approximate the complete dynamic characteristics of the original data without losing potentially useful information due to over-filtering.

[0048] The first and second datasets are time-aligned to ensure strict consistency in the time dimension between the output datasets of the two channels. Since there may be slight processing delays or differences in data granularity between the two channels, techniques such as timestamp matching, interpolation, or resampling are needed to precisely align corresponding data points in the two datasets for effective comparison.

[0049] Finally, point-by-point difference calculations are performed on the first and second datasets after time alignment to obtain the computational error. After time alignment of the two datasets, the difference between the processing results of the two channels can be obtained by subtracting each corresponding data point. These differences are considered as computational errors, reflecting the deviations between the actual running channel and the benchmark simulation channel in terms of data processing logic, parameter configuration, or filtering strategy.

[0050] To illustrate this technical solution more clearly, a specific example is used below. Suppose a power grid region experiences an abnormal electricity consumption work order within a specific abnormal sub-time zone, requiring verification of the data post-processing workflow. First, the raw metering data for that abnormal sub-time zone is extracted from the power grid data storage center. Then, an actual operation channel and a benchmark simulation channel are constructed. The actual operation channel is configured to fully simulate the data processing logic used for billing and monitoring in the current power grid system, including data aggregation, a first data filtering module designed to remove all high-frequency transient load components, and the application of current billing rules. The benchmark simulation channel is configured to perform the same data aggregation and billing rule application, but its second data filtering module only removes general noise and specifically retains high-frequency transient load components as a comparison benchmark. The same raw metering data is input into these two channels for processing, resulting in a first dataset and a second dataset. Next, time alignment is performed on these two datasets to ensure that the electricity consumption data at each time point corresponds accurately. Finally, point-by-point differential calculations are performed on the time-aligned first and second datasets. For example, if the actual operating channel experiences a lower power reading than the baseline simulated channel at a certain point in time due to filtering out a sudden large load fluctuation, this difference is precisely quantified as the calculation error introduced by the first data filtering strategy. In this way, calculation errors caused by specific logic or parameter configurations in the data post-processing flow can be clearly identified and quantified, thus providing accurate data support for root cause analysis of abnormal work orders.

[0051] Through the above technical solution, this embodiment establishes a dual-channel comparison mechanism between actual operation and benchmark simulation, which can effectively distinguish calculation deviations caused by specific filtering strategies in the data processing flow (such as removing high-frequency transient load components), thereby avoiding potential error confusion. As a result, abnormal factors related to the data post-processing flow in abnormal power supply work orders can be identified and located more accurately, significantly improving the accuracy and efficiency of abnormal work order verification, and providing solid technical support for the stable operation of the power grid and accurate billing.

[0052] In some embodiments, step S107, determining the primary-level anomaly factors and the secondary-level anomaly factors based on the equipment measurement error, calculation error, and total error, may include, but is not limited to, the following steps: Step S201: Obtain the power grid operating status and priority adjustment instructions. The power grid operating status includes the power grid load factor, reserve capacity, line congestion status and voltage stability margin. Step S202: Perform characteristic analysis on the equipment measurement error and calculation error to obtain error characteristics, including persistence, intermittency, frequency of occurrence and correlation with environmental factors; Step S203: Determine the risk weights based on the power grid operating status and priority adjustment instructions; Step S204: Based on risk weight, error characteristics, fault level and economic loss records, comprehensively evaluate the equipment measurement error and calculation error to obtain the first impact score of the equipment measurement error and the second impact score of the calculation error. Step S205: Calculate the first proportion based on the equipment measurement error and the total error; Step S206: Calculate the second proportion based on the calculation error and the total error; Step S207: If the difference between the first proportion and the second proportion is less than the preset proportion threshold, then the main level abnormal factor is determined to be a sudden change in user behavior. Step S208: If the difference between the first percentage and the second percentage is greater than the preset percentage threshold, then determine whether the first influence score is greater than the second influence score. Step S209: If the first impact score is greater than the second impact score, then the main level anomaly factor is determined to be equipment measurement error; Step S210: If the first impact score is less than the second impact score, then the main level abnormal factor is determined to be the calculation error; Step S211: Determine the hierarchical anomaly factors based on the primary level anomaly factors.

[0053] In some embodiments, relying solely on error values ​​for judgment may fail to adequately consider the real-time operating status of the power grid, the inherent characteristics of the error, and potential risk impacts, leading to inaccurate identification of anomalies or unreasonable prioritization. To address this, the power grid operating status and priority adjustment instructions can be obtained first. The power grid operating status includes load factor, reserve capacity, line congestion status, and voltage stability margin. These indicators reflect the health and carrying capacity of the power grid, providing crucial background information for assessing anomalies. Priority adjustment instructions are directives from the power grid dispatch center or management department used to adjust the priority of anomaly handling under specific circumstances, such as during power grid emergencies where certain anomalies may require priority processing.

[0054] Then, a characteristic analysis of the equipment's measurement and calculation errors is performed to obtain error characteristics, aiming to gain a deeper understanding of the nature of errors. These error characteristics include persistence, intermittency, frequency of occurrence, and correlation with environmental factors. Persistence refers to whether the error persists for a long time; intermittency refers to whether the error occurs periodically or irregularly; frequency of occurrence refers to the number of times or the probability of the error occurring; and correlation with environmental factors refers to whether the error is related to external environmental conditions such as temperature, humidity, and weather. Through these characteristic analyses, the nature of the error can be more comprehensively characterized.

[0055] Then, risk weights are determined based on the grid's operating status and priority adjustment instructions. Risk weights reflect the current grid's sensitivity and tolerance to specific anomalies. For example, when the grid load factor is high and reserve capacity is low, any potential anomaly may pose a higher risk, therefore the corresponding risk weight will be increased.

[0056] Based on risk weights, error characteristics, fault levels, and economic loss records, a comprehensive assessment of equipment measurement errors and calculation errors is conducted to obtain a first impact score for equipment measurement errors and a second impact score for calculation errors. Fault levels refer to the severity classification of abnormal events, such as general, important, and urgent. Economic loss records refer to the economic impact data caused by historical abnormal events. By comprehensively considering risk weights, error characteristics, fault levels, and economic loss records, equipment measurement errors and calculation errors can be quantitatively assessed, resulting in a first impact score for equipment measurement errors and a second impact score for calculation errors. These scores objectively reflect the potential impact of these two types of errors on power grid operation.

[0057] Based on the equipment measurement error and the total error, calculate the first proportion, and based on the calculated error and the total error, calculate the second proportion. This allows for a preliminary determination of which error type dominates in terms of total quantity. If the difference between the first and second proportions is less than a preset proportion threshold, it indicates that the two errors contribute similarly to the total error. In this case, the anomaly may not be caused by a single technical fault, but rather by a sudden change in user behavior, such as a sudden large-scale power consumption or power outage, leading to significant fluctuations in power consumption data. The primary anomaly factor can be identified as a sudden change in user behavior. If the difference between the first and second proportions is greater than the preset proportion threshold, further judgment is needed based on the impact score to determine whether the first impact score is greater than the second impact score.

[0058] If the first impact score is greater than the second impact score, it indicates that although the equipment measurement error may not account for an absolute majority in terms of total quantity, its potential impact or risk on power grid operation is higher, and the primary level anomaly factor can be identified as equipment measurement error. If the first impact score is less than the second impact score, the primary level anomaly factor is identified as calculation error.

[0059] Finally, based on the primary-level anomalies, the stratified anomalies are determined. Once the primary-level anomalies are identified, their types can be further refined to determine the stratified anomalies, thereby achieving precise localization of the causes of the anomalies.

[0060] This embodiment employs a multi-dimensional and in-depth analysis of equipment measurement and calculation errors by introducing grid operating status, priority adjustment commands, error characteristic analysis, risk weights, and a comprehensive evaluation mechanism. Specifically, acquiring grid operating status and priority adjustment commands allows for the identification of anomalies in conjunction with real-time grid operating conditions and management strategies, avoiding blind judgments detached from reality. Error characteristic analysis reveals the inherent patterns of errors; for example, persistent errors may indicate equipment aging, while intermittent errors may be related to communication interference, providing richer clues for subsequent fault diagnosis. The determination of risk weights and the comprehensive evaluation mechanism quantify the potential impact of equipment measurement and calculation errors, enabling a more objective comparison of their severity. By calculating and comparing the first and second proportions, a preliminary distinction can be made between problems caused by sudden changes in user behavior and those caused by technical errors. When technical errors dominate, further comparison of the first and second impact scores allows for precise identification of whether the problem stems from equipment measurement issues or data post-processing calculation problems, thus avoiding potential misjudgments or omissions in traditional methods and significantly improving the accuracy and relevance of anomaly identification.

[0061] Through the above technical solution, this embodiment incorporates diverse information such as power grid operating status, priority adjustment instructions, error characteristic analysis, and risk weights, making the judgment of abnormal factors more comprehensive and intelligent. It can effectively distinguish different types of primary-level abnormal factors, such as sudden changes in user behavior, equipment measurement errors, and calculation errors. This layered, multi-dimensional verification mechanism not only improves the accuracy of anomaly location but also enables more reasonable assessment and handling of anomalies based on the actual operating conditions and priority requirements of the power grid, thus providing more reliable technical support for the stable operation and refined management of the power grid.

[0062] In some embodiments, in step S211, determining the hierarchical anomaly factors based on the primary-level anomaly factors may include, but is not limited to, the following steps: If the main level anomaly is equipment measurement error, then obtain the equipment operation status log; Perform a first fault analysis on the equipment operation status log to obtain the first fault analysis results; If the first fault analysis result is an increase in the number of communication interruptions, then the hierarchical anomaly factor is determined to be a communication fault; If the main level anomaly is a calculation error, then obtain the parameter configuration information and calculation rules; A second fault analysis is performed on the parameter configuration information and calculation rules to obtain the second fault analysis results; If the second fault analysis result indicates a computational logic deployment error, then the hierarchical anomaly is determined to be a billing parameter configuration failure.

[0063] In some embodiments, the main-level anomaly factors can be determined first. If the main-level anomaly factor is a device measurement error, then the device operation status log is obtained. This device operation status log refers to detailed data recording various operating parameters, event records, alarm information, and communication status of the device within a specific time period. For example, the log may contain heartbeat packet sending records, data transmission success rate, error codes, restart events, and connection / disconnection timestamps of the communication link.

[0064] Then, a first fault analysis is performed on the equipment operation status log to obtain the first fault analysis results. Specific problems that may exist in the equipment can be diagnosed through pattern recognition, abnormal event counting, and trend analysis of the log data. For example, the first fault analysis results can be obtained by statistically analyzing the frequency and duration of communication interruption events during abnormal periods.

[0065] If the first fault analysis result shows an increase in the number of communication interruptions, then the hierarchical anomaly factor is determined to be a communication fault. Communication faults may include physical link damage, network congestion, communication module failure, or configuration errors. If the main-level anomaly factor is a calculation error, then parameter configuration information and calculation rules are obtained. Parameter configuration information refers to the various thresholds, coefficients, time intervals, and other set values ​​used in the power data post-processing workflow, while calculation rules specify the logical algorithms for power aggregation, billing, and anomaly detection. This information is typically stored in the system's configuration database or rule engine.

[0066] A second fault analysis is then performed on the parameter configuration information and calculation rules to obtain the results. This second fault analysis aims to identify potential errors in the calculation process by comparing the current configuration with the baseline configuration, checking the integrity and correctness of the calculation logic, and simulating the calculation process. For example, the second fault analysis results can be obtained by parsing the syntax of the calculation rules, testing the logical path coverage, and comparing them with historical correct calculation results.

[0067] If the second fault analysis result indicates a computational logic deployment error, then the hierarchical anomaly is determined to be a billing parameter configuration fault. If the second fault analysis result indicates a computational logic deployment error—for example, an incorrectly modified billing formula, a deviation in the data aggregation order, or improperly set anomaly judgment conditions—then the hierarchical anomaly is determined to be a billing parameter configuration fault. A billing parameter configuration fault may cause the electricity calculation result to deviate from the actual value, thereby triggering an electricity anomaly work order.

[0068] To illustrate this technical solution more clearly, a specific example is used below. Suppose a power grid area experiences abnormal electricity consumption work orders within a specific time period. After initial verification, the primary anomaly is identified as equipment measurement error. At this point, the system will automatically trigger the acquisition and analysis of the operational status logs of the relevant data acquisition equipment. For example, by analyzing the smart meter's operational logs, it is found that the number of communication interruptions with the data center during the abnormal period is significantly higher than normal, and the duration is longer. The first fault analysis result clearly indicates an increase in the number of communication interruptions; therefore, the hierarchical anomaly is identified as a communication fault. Technicians can then directly check the smart meter's communication module, network connection, or signal strength to quickly locate and resolve the problem. As another example, if the primary anomaly is identified as a calculation error, the system will acquire the parameter configuration information and calculation rules from the electricity billing system. Through a second fault analysis of this information, it is found that a newly deployed billing rule has an error in its aggregation logic when processing specific types of user electricity consumption data, resulting in a lower final electricity consumption calculation result. The second fault analysis result indicates an error in the deployment of the calculation logic; therefore, the hierarchical anomaly is identified as a billing parameter configuration fault. Technicians can immediately correct the erroneous billing rules, thereby eliminating the power consumption anomaly.

[0069] Through the above technical solution, this embodiment introduces the analysis of equipment operation status logs, parameter configuration information, and calculation rules, which can further refine the main-level anomalies into specific, actionable hierarchical anomalies, such as communication failures or billing parameter configuration failures. This refined anomaly identification capability significantly improves the accuracy and efficiency of fault diagnosis, shortens fault investigation time, and reduces operation and maintenance costs. Furthermore, clearly defined hierarchical anomalies provide a solid foundation for developing targeted repair strategies and preventative measures, thereby effectively improving the stability and reliability of power grid operation.

[0070] In some embodiments, after determining the primary-level anomaly factors and the secondary-level anomaly factors based on the device measurement error, calculation error, and total error, the method further includes the following steps: Step S301: Identify multiple primary-level anomalous factors, where the difference in the impact scores of multiple primary-level anomalous factors is less than a preset score difference threshold. Step S302: Analyze the physical first adjacency relationship of the devices associated with multiple primary-level anomalies; Step S303: Analyze the logical second adjacency relationship between the data processing links associated with multiple main-level anomalies; Step S304: Analyze the causal relationships of multiple primary-level anomaly factors in historical failures; Step S305: Based on the first adjacency relationship, the second adjacency relationship, and the causal relationship, cluster multiple primary-level abnormal factors to obtain composite abnormal events; Step S306: Determine the target processing strategy for handling complex abnormal events based on the collaborative processing rules; Step S307: Assign a comprehensive priority to complex abnormal events according to the target processing strategy.

[0071] In some embodiments, since anomalies are often not caused by a single factor, multiple primary-level anomalies may coexist, and there may be complex relationships between them, such as physical adjacency, logical dependence, or historical causal relationships. If these anomalies are identified and processed independently, it may not be possible to comprehensively and effectively solve the problem, and may even lead to inefficient processing or the omission of potential complex risks.

[0072] To address this, multiple primary-level anomalies can be identified, where the difference in impact scores among these factors is less than a preset score difference threshold. This implies that these anomalies are similar in degree of influence, and it is inappropriate to simply consider one as the sole dominant factor; rather, their potential synergistic effects should be considered. The preset score difference threshold can be set based on actual power grid operation experience and risk management strategies; for example, it can be set to 5% or 10% of the total error.

[0073] Then, the physical first adjacency relationships of the devices associated with multiple primary-level anomalies are analyzed to explore whether these anomalies originate from physically connected or adjacent devices. For example, if two anomalies are associated with different devices within the same substation or different nodes on the same transmission line, they may have a physical first adjacency relationship. This analysis helps to understand the propagation and association of anomalies from a spatial perspective.

[0074] Analyzing the logical second-order adjacency relationships among multiple primary-level anomalies aims to identify whether these anomalies have dependency or sequential relationships in data flow or business processes. For example, one anomaly may be related to the data acquisition stage, while another anomaly may be related to the aggregation calculation stage based on the acquired data, forming a logical upstream-downstream relationship. This analysis helps reveal the inherent connections between anomalies within the data processing chain.

[0075] Analyze the causal relationships among multiple primary-level anomalies in historical failures. This can be done by reviewing historical failure records and expert experience databases to assess whether these anomalies have co-occurred in the past, or whether the occurrence of one anomaly frequently leads to the occurrence of another. For example, historical data shows that communication failures often lead to increased measurement errors in data acquisition equipment. This analysis helps predict the evolution and interaction of anomalies from both temporal and empirical perspectives.

[0076] Then, based on the first adjacency relationship, the second adjacency relationship, and causal association, multiple primary-level anomalous factors are clustered to obtain composite anomalous events. Various algorithms can be used, such as rule-based expert systems, machine learning algorithms (e.g., K-means, hierarchical clustering), or graph theory methods, to group strongly correlated primary-level anomalous factors into a single composite anomalous event. Composite anomalous events represent more complex and multi-factor-intertwined anomalous situations in the power grid.

[0077] Finally, based on the collaborative processing rules, a target handling strategy for dealing with complex anomalies is determined. The collaborative processing rules are a set of predefined processing procedures and resource allocation schemes designed for different types of complex anomalies. These rules consider the complexity, potential impact, and synergy of required resources for complex anomalies, aiming to provide a more comprehensive and efficient solution than single-factor handling. Furthermore, based on the target handling strategy, complex anomalies are assigned a comprehensive priority. This comprehensive priority not only considers the impact of individual anomalies but also comprehensively assesses the overall risk, urgency, and potential threat to power grid stability of complex anomalies, ensuring that resources are prioritized for the most critical complex anomalies.

[0078] Through the above technical solution, this embodiment significantly improves the accuracy and depth of diagnosis for complex anomalies by performing multi-dimensional correlation analysis and clustering on multiple primary-level anomaly factors. This enables the development of more targeted and collaborative processing strategies, optimizes the processing flow of power grid anomaly work orders, improves fault response efficiency and resource utilization, and helps enhance the stability and reliability of power grid operation, reducing potential risks and economic losses caused by complex anomalies.

[0079] In some embodiments, in step S306, the target processing strategy for handling complex abnormal events is determined according to the collaborative processing rules, which may include, but is not limited to, the following steps: Step S401: Obtain information on sudden changes in the power grid operating environment. This information includes extreme weather warnings, reports of large-scale equipment failures, network security attack alerts, and emergency declarations in power grid dispatch instructions. Step S402: Based on the information on sudden changes, assess the current risk level and resource availability of the power grid. The risk level includes power grid stability risk, economic loss risk and user experience risk. Resource availability includes the number of repair personnel, backup equipment inventory and communication bandwidth. Step S403: Adjust the priority weights and resource allocation strategies in the collaborative processing rules according to the risk level and resource availability; Step S404: Determine the target processing strategy based on the adjusted collaborative processing rules.

[0080] In some embodiments, since the power grid environment may face sudden changes, such as extreme weather or large-scale equipment failures, if the collaborative processing rules fail to fully consider these dynamic factors, the determined target processing strategy may not be flexible enough or able to respond in a timely manner, thereby affecting the processing efficiency and effectiveness of abnormal events.

[0081] To this end, information on sudden changes in the power grid operating environment can be obtained first. This information includes extreme weather warnings issued by meteorological departments, such as typhoons, blizzards, and high temperatures, which indicate natural disasters that may damage the power grid's physical infrastructure; large-scale equipment failure reports generated by monitoring systems or manual reports, such as substation trips and line interruptions, which reflect abnormal conditions at the power grid hardware level; network security attack alerts, which indicate potential threats to the power grid control system or communication network; and emergency declarations in power grid dispatch instructions, which are issued by the dispatch center based on real-time operating conditions and require emergency measures.

[0082] Then, based on information about sudden changes, the current risk level and resource availability of the power grid are assessed. The risk level is a quantitative assessment of the potential harm the power grid may face under the impact of the current emergency. This can include grid stability risk (the risk of impairment to the grid's ability to maintain normal voltage and frequency); economic loss risk (the economic costs such as equipment damage and power outage compensation caused by the abnormal event); and user experience risk (the degree of dissatisfaction among users due to power outages or deterioration in power quality). Resource availability refers to the status of various resources that the power grid can utilize in response to abnormal events. This can include the number of repair personnel (the availability of on-site maintenance and troubleshooting personnel); spare parts inventory (the reserve of spare parts to replace faulty equipment); and communication bandwidth (the communication network capacity used for data transmission and command issuance). By assessing these risks and resources, a comprehensive understanding of the power grid's vulnerability and response capabilities can be obtained.

[0083] Then, based on the risk level and resource availability, the priority weights and resource allocation strategies in the collaborative processing rules are adjusted. Priority weights refer to the ranking of the importance of different anomalies or processing tasks when handling complex anomalies; for example, in extreme weather, the priority of ensuring critical loads may be increased. Resource allocation strategies refer to how to efficiently allocate limited resources such as repair personnel, backup equipment, and communication bandwidth to different anomaly handling tasks. By combining real-time risk levels and resource availability, collaborative processing rules can transform from static presets to dynamic adaptations, ensuring that the most critical issues are addressed first in emergencies and that existing resources are used rationally.

[0084] Finally, based on the adjusted collaborative processing rules, the target processing strategy is determined. This target processing strategy is a specific action plan optimized for the current power grid operating environment and resource conditions, aiming to resolve complex anomalies in the most effective way.

[0085] Through the above technical solution, this embodiment can significantly improve the adaptability and robustness of the hierarchical verification method for power grid power anomaly work orders when handling complex anomaly events. This solution makes the collaborative processing rules no longer fixed, but dynamically adjustable according to sudden changes in the power grid operating environment, thereby ensuring that the determined target processing strategy can respond to actual needs more accurately and promptly. Especially in the face of emergencies such as extreme weather, large-scale failures, or cyberattacks, real-time assessment of risk levels and resource availability can optimize priority weights and resource allocation, effectively reducing power grid stability risks, economic loss risks, and user experience risks, thereby improving the efficiency of anomaly event handling and the overall resilience of the power grid.

[0086] In some embodiments, step S402, assessing the current risk level of the power grid based on the sudden change information, may include, but is not limited to, the following steps: Cross-validate information on sudden changes; Based on historical power grid operation data and power grid expert experience database, historical experience analysis is conducted on the information of sudden changes after cross-validation to obtain historical experience analysis results; Identify the power grid operation mode, which includes normal operation mode, emergency operation mode, or maintenance operation mode. Based on the power grid operation mode, a risk integration strategy is selected from a pre-set risk preference database; The risk level is assessed based on the risk integration strategy and the results of historical experience analysis.

[0087] In some embodiments, due to the diverse and uncertain sources of information regarding sudden changes, and the complex and ever-changing nature of power grid operation, risk assessments based solely on preliminary information may result in inaccurate assessments or fail to adequately consider the actual operating conditions of the power grid, thus affecting the effectiveness of subsequent processing strategies. Therefore, cross-validation of sudden change information can be performed first. Received sudden change information (such as extreme weather warnings, reports of large-scale equipment failures, cybersecurity attack alerts, and emergency declarations in power grid dispatch instructions) can be compared and checked for consistency across multiple sources. For example, an extreme weather warning can be compared with data from multiple meteorological departments, satellite imagery, and on-site inspection reports to confirm its authenticity and accuracy. The aim is to eliminate false information, reduce information bias, and ensure that the data foundation upon which subsequent risk assessments are based is reliable.

[0088] Then, based on historical power grid operation data and a power grid expert experience database, historical experience analysis is performed on the cross-validated information regarding sudden changes, yielding historical experience analysis results. This allows for the prediction and assessment of the potential impact of current sudden changes by leveraging past power grid operation records and accumulated expert knowledge. Specifically, information such as power grid response, fault propagation paths, and recovery times during similar events can be extracted from historical power grid operation data. Simultaneously, by combining this with knowledge from the power grid expert experience database regarding specific fault modes, risk transmission mechanisms, and emergency response plans, an in-depth analysis of the potential impact of the current event can be conducted. The aim is to learn from history and provide valuable references and predictive basis for current risk assessment.

[0089] Re-identifying the power grid's operating mode involves identifying whether it is a normal operation mode, an emergency operation mode, or a maintenance operation mode. The current operating state of the power grid can be determined based on factors such as the current load level, equipment status, dispatch instructions, and the external environment. For example, when the load is stable, there are no major faults, and no planned maintenance, the power grid is in a normal operation mode; when a large-scale power outage occurs, equipment malfunctions severely, or a cyberattack occurs, the power grid may enter an emergency operation mode; and when there is planned maintenance or upgrades to some equipment, it is in a maintenance operation mode. The purpose is to provide important context for risk assessment, as the vulnerability and resilience of the power grid differ significantly under different operating modes.

[0090] Based on the power grid operating mode, risk integration strategies are selected from a pre-defined risk appetite library. This library is a collection of various risk management strategies, predefined according to different power grid operating modes and risk tolerance. For example, under normal operating mode, a conservative risk integration strategy might be preferred to minimize any potential disturbances; while under emergency operating mode, a more aggressive strategy might be needed to quickly restore power supply or prevent system collapse. Risk integration strategies can include risk avoidance, risk reduction, risk transfer, or risk acceptance. The aim is to ensure that risk assessment and subsequent decisions are aligned with the power grid's current operating objectives and risk tolerance.

[0091] Finally, the risk level is assessed based on the risk integration strategy and historical experience analysis. The risk level assessment is conducted after comprehensively considering both the risk integration strategy and the results of historical experience analysis. For example, if historical experience analysis indicates that a particular incident has led to serious consequences under a specific operational mode, and the currently selected risk integration strategy is risk avoidance, then the risk level of the incident will be assessed as high. The assessment result can be a quantitative value or a qualitative classification (e.g., low, medium, high risk). Its purpose is to provide a clear basis for subsequent priority adjustments and resource allocation.

[0092] Through the above technical solution, this embodiment effectively eliminates interference from false or inaccurate information by cross-validating information on sudden changes, ensuring the input quality of risk assessment. By combining historical power grid operation data and expert experience for analysis, risk assessment is no longer an isolated judgment, but rather based on rich practical experience and professional knowledge, thereby improving the accuracy of risk prediction. Furthermore, by identifying power grid operation modes and selecting corresponding risk integration strategies, the risk assessment results can better adapt to the actual operating context and management objectives of the power grid, avoiding assessment biases based on fixed thresholds. Therefore, this embodiment provides a more solid and reliable basis for subsequent target processing strategy formulation and comprehensive prioritization, thereby improving the response efficiency and handling effectiveness of abnormal power grid events, effectively ensuring the safe and stable operation and economic benefits of the power grid.

[0093] The beneficial effects of implementing the embodiments of the present invention include: First, the embodiments of this application obtain power consumption abnormality work orders, then extract abnormal time zones, user power consumption types, and data acquisition equipment information from the power consumption abnormality work orders, and perform time-segmented verification of the power consumption abnormality work orders to determine abnormal sub-time zones. Then, based on the abnormal sub-time zones, data acquisition verification processing is performed to obtain equipment measurement errors. Based on the abnormal sub-time zones, data post-processing flow verification is performed to obtain calculation errors. Based on the equipment measurement errors and calculation errors, the total error is calculated. Finally, based on the equipment measurement errors, calculation errors, and total errors, the main-level abnormal factors and hierarchical abnormal factors are determined. This allows for the combination of multiple errors to determine abnormal factors at different levels, thereby achieving hierarchical verification and improving accuracy and reliability.

[0094] like Figure 2 As shown, this embodiment of the invention also provides a hierarchical verification system for abnormal power grid work orders, including: The work order acquisition module 501 is used to acquire work orders with abnormal power consumption. Data extraction module 502 is used to extract abnormal time zone, user electricity consumption type and data acquisition device information from power abnormality work orders; The time-segmented verification module 503 is used to perform time-segmented verification on power abnormality work orders based on abnormal time zones, user power consumption types, and data acquisition equipment information, and to determine abnormal sub-time zones. The equipment measurement error determination module 504 is used to perform data acquisition and verification processing based on abnormal sub-time zones to obtain the equipment measurement error; The calculation error determination module 505 is used to perform data post-processing verification based on abnormal sub-time zones to obtain the calculation error; The total error calculation module 506 is used to calculate the total error based on the equipment measurement error and calculation error; The anomaly verification module 507 is used to determine the main-level anomaly factors and the hierarchical anomaly factors based on the equipment measurement error, calculation error and total error.

[0095] The content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0096] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

Claims

1. A hierarchical verification method for power grid power anomaly work orders, characterized in that, Includes the following steps: Get abnormal power consumption work orders; Extract the abnormal time zone, user power consumption type, and data acquisition device information from the power consumption anomaly work order; Based on the abnormal time zone, the user's electricity consumption type, and the data acquisition equipment information, the abnormal electricity work order is checked in different time periods to determine the abnormal sub-time zone; Based on the abnormal sub-time zone, data acquisition and verification processing is performed to obtain the equipment measurement error; Based on the abnormal sub-time zone, the data post-processing procedure is checked to obtain the calculation error; Calculate the total error based on the equipment measurement error and the calculation error; Based on the equipment measurement error, the calculation error, and the total error, the primary level anomaly factors and the secondary level anomaly factors are determined.

2. The method according to claim 1, characterized in that, The step of performing data acquisition and verification processing based on the abnormal sub-time zone to obtain the equipment measurement error includes: Extract the original metering data and auxiliary reference data within the abnormal sub-time zone from the power grid data storage center; The original measurement data is subjected to an integrity check to obtain the integrity check results; The original measurement data are compared for consistency to obtain the consistency comparison results; The original measurement data and the auxiliary reference data are compared to obtain the comparison results; The measurement error of the device is determined based on the integrity check result, the consistency comparison result, and the comparison result.

3. The method according to claim 1, characterized in that, The step of performing data post-processing verification based on the abnormal sub-time zone to obtain the calculation error includes: Extract the original metering data within the abnormal sub-time zone from the power grid data storage center; Perform a data aggregation logic check on the original measurement data to obtain the data aggregation logic check result; The original metering data is subjected to a billing rule application check to obtain the billing rule application check result; The original measurement data is subjected to anomaly detection rule checks to obtain anomaly detection rule check results; The calculation error is determined based on the data aggregation logic check results, the billing rule application check results, and the anomaly judgment rule check results.

4. The method according to claim 1, characterized in that, The step of performing data post-processing verification based on the abnormal sub-time zone to obtain the calculation error includes: Extract the original metering data within the abnormal sub-time zone from the power grid data storage center; Construct actual operation channels and benchmark simulation channels; The raw metering data is input into the actual operation channel, which is used to perform a first data processing on the raw metering data to obtain a first dataset. The first data processing includes data aggregation, first data filtering, and billing rule application. The first dataset includes aggregated electricity data, key intermediate calculation results, and anomaly judgment flags. The first data filtering is used to remove noise and high-frequency transient load components. The original metering data is input into the benchmark simulation channel, which is used to perform a second data processing on the original metering data to obtain a second dataset. The second data processing includes data aggregation, second data filtering, and billing rule application. The second data filtering is used to remove noise and retain high-frequency transient load components. Perform time alignment processing on the first dataset and the second dataset; The calculation error is obtained by performing point-by-point difference calculation on the first and second datasets after time alignment.

5. The method according to claim 1, characterized in that, The process of determining primary-level and secondary-level anomaly factors based on the equipment measurement error, the calculation error, and the total error includes: Obtain power grid operating status and priority adjustment instructions, wherein the power grid operating status includes power grid load factor, reserve capacity, line congestion status and voltage stability margin; The measurement error and calculation error of the equipment are analyzed to obtain error characteristics, which include persistence, intermittency, frequency of occurrence and correlation with environmental factors. Risk weights are determined based on the power grid operating status and the priority adjustment instructions; Based on the risk weight, error characteristics, fault level, and economic loss records, the equipment measurement error and the calculation error are comprehensively evaluated to obtain a first impact score for the equipment measurement error and a second impact score for the calculation error. Calculate the first percentage based on the device measurement error and the total error; Calculate the second proportion based on the calculation error and the total error; If the difference between the first percentage and the second percentage is less than a preset percentage threshold, then the main-level abnormal factor is determined to be a sudden change in user behavior. If the difference between the first percentage and the second percentage is greater than a preset percentage threshold, then determine whether the first impact score is greater than the second impact score. If the first impact score is greater than the second impact score, then the main-level anomaly factor is determined to be equipment measurement error; If the first impact score is less than the second impact score, then the main-level anomaly factor is determined to be a calculation error; Based on the primary-level anomaly factors, the hierarchical anomaly factors are determined.

6. The method according to claim 5, characterized in that, The step of determining the hierarchical anomaly factors based on the primary-level anomaly factors includes: If the main-level anomaly is a device measurement error, then obtain the device operation status log; A first fault analysis is performed on the device's operating status log to obtain the first fault analysis result; If the first fault analysis result is an increase in the number of communication interruptions, then the hierarchical anomaly factor is determined to be a communication fault; If the main-level anomaly is a calculation error, then obtain the parameter configuration information and calculation rules; A second fault analysis is performed on the parameter configuration information and the calculation rules to obtain the second fault analysis result; If the second fault analysis result is a computational logic deployment error, then the hierarchical anomaly is determined to be a billing parameter configuration fault.

7. The method according to claim 1, characterized in that, After determining the primary-level anomaly factors and secondary-level anomaly factors based on the equipment measurement error, the calculation error, and the total error, the method further includes: Identify multiple primary-level anomalous factors, wherein the difference in the impact scores of the multiple primary-level anomalous factors is less than a preset score difference threshold; Analyze the physical first adjacency relationship of the devices associated with the multiple primary-level anomalies; Analyze the logical second adjacency relationship between the data processing links associated with the multiple primary-level anomalies; Analyze the causal relationships among the multiple primary-level anomalies in historical failures; Based on the first adjacency relationship, the second adjacency relationship, and the causal relationship, the multiple primary-level abnormal factors are clustered to obtain composite abnormal events; Based on the collaborative processing rules, a target processing strategy for handling the complex abnormal events is determined; Based on the target processing strategy, the composite abnormal events are assigned a comprehensive priority.

8. The method according to claim 7, characterized in that, The step of determining the target processing strategy for handling the complex abnormal events based on the collaborative processing rules includes: Acquire information on sudden changes in the power grid operating environment, including extreme weather warnings, reports of large-scale equipment failures, network security attack alerts, and emergency declarations in power grid dispatch instructions; Based on the information on sudden changes, assess the current risk level and resource availability of the power grid. The risk level includes power grid stability risk, economic loss risk, and user experience risk. The resource availability includes the number of emergency repair personnel, backup equipment inventory, and communication bandwidth. Based on the risk level and resource availability, adjust the priority weights and resource allocation strategies in the collaborative processing rules; The target processing strategy is determined based on the adjusted collaborative processing rules.

9. The method according to claim 8, characterized in that, The assessment of the current power grid risk level based on the sudden change information includes: Cross-validate the information on the sudden changes; Based on historical power grid operation data and power grid expert experience database, historical experience analysis is conducted on the information of sudden changes after cross-validation to obtain historical experience analysis results; Identify the power grid operation mode, which includes normal operation mode, emergency operation mode, or maintenance operation mode; Based on the power grid operation mode, a risk integration strategy is selected from a preset risk preference database; The risk level is assessed based on the risk integration strategy and the results of the historical experience analysis.

10. A hierarchical verification system for power grid power anomaly work orders, characterized in that, include: The work order acquisition module is used to acquire work orders with abnormal power consumption. The data extraction module is used to extract abnormal time zone, user electricity consumption type and data acquisition device information from the power abnormality work order; The time-segmented verification module is used to perform time-segmented verification on the abnormal power consumption work order based on the abnormal time zone, the user's power consumption type, and the data acquisition equipment information, and to determine the abnormal sub-time zone; The equipment measurement error determination module is used to perform data acquisition and verification processing based on the abnormal sub-time zone to obtain the equipment measurement error; The calculation error determination module is used to perform data post-processing verification based on the abnormal sub-time zone to obtain the calculation error; The total error calculation module is used to calculate the total error based on the measurement error of the device and the calculation error. The anomaly verification module is used to determine the main-level anomaly factors and the hierarchical anomaly factors based on the equipment measurement error, the calculation error, and the total error.

Images