A power metering box early warning method, system, terminal and medium

Abstract

This invention belongs to the technical field of power metering boxes, specifically disclosing a method, system, terminal, and medium for early warning of power metering boxes. The method involves collecting multi-source operational data from the power metering box within a preset sampling period, performing time alignment processing on the multi-source operational data, and calculating data reliability weights based on data quality. An individualized operational baseline model for the power metering box is constructed based on the historical statistical distribution of the multi-source operational data, and an adaptive early warning threshold is generated. The operational data within the current sampling period is compared with the adaptive early warning threshold to construct multi-dimensional risk characteristics and calculate a comprehensive risk assessment value. The early warning level of the power metering box is determined based on the comprehensive risk assessment value, and the corresponding early warning information is output. Through the above technical solution, a comprehensive assessment and graded early warning of the operational risks of power metering boxes are achieved, improving the accuracy and adaptability of the early warning results.

Description

Technical Field

[0001] This invention belongs to the field of power metering box technology, specifically relating to a power metering box early warning method, system, terminal and medium. Background Technology

[0002] As a key device in power distribution systems for electricity metering, monitoring, and management, power metering boxes are typically installed outdoors for extended periods. Their operational status directly affects the accuracy of electricity metering, the safety of power system operation, and the reliability of electricity management. With the expansion of power system scale and the improvement of intelligence levels, outdoor power metering boxes are widely used in various complex environments such as urban roads, residential areas, industrial parks, and remote areas. These environments are characterized by large fluctuations in temperature and humidity, complex external electromagnetic interference, frequent mechanical vibrations, and variable climatic conditions.

[0003] In existing technologies, the monitoring and early warning of the operating status of power metering boxes usually adopts the method of threshold judgment based on single or a few environmental parameters such as internal temperature, humidity or gas concentration. Some solutions also combine tilt, vibration or external meteorological data to alert for abnormal situations.

[0004] While these solutions can monitor obvious abnormal operating conditions to some extent, they mostly rely on fixed thresholds or empirical parameters, making it difficult to fully consider the differences in installation location, operating history, environmental conditions, and load characteristics of different metering boxes. Therefore, the accuracy and adaptability of the early warning results remain limited. Furthermore, existing early warning solutions generally focus on instantaneous judgment of data at the current moment. When short-term disturbances occur in the operating environment, or when sensor data is missing or fluctuates, false alarms or missed alarms are easily generated. Summary of the Invention

[0005] This invention addresses the problems in the prior art by providing a method, system, terminal, and medium for early warning of power metering boxes. It solves the problem that relying on fixed thresholds or empirical parameters makes it difficult to fully consider the differences in installation location, operating history, environmental conditions, and load characteristics of different metering boxes, resulting in limited accuracy and adaptability of early warning results. At the same time, it solves the problem that existing early warning schemes generally focus on instantaneous judgment of data at the current moment, which can easily lead to false alarms or missed alarms when there are short-term disturbances in the operating environment or when there are missing or fluctuating sensor data.

[0006] The technical solution adopted in this invention is as follows: Firstly, this application provides a method for early warning of electricity metering boxes, the method comprising the following steps: Within a preset sampling period, multi-source operational data of the power metering box is collected. The multi-source operational data includes internal environmental data of the box, external environmental data of the box, and metering operation status data. Time alignment processing is performed on multi-source operational data, and data trust weights are calculated for various types of operational data based on data integrity, data stability, and historical consistency. Based on the historical statistical distribution of multi-source operation data, an individualized operation baseline model is constructed for each power metering box, and an adaptive early warning threshold is generated on the operation baseline model according to the operation data of the current period. Within the current sampling period, the weighted multi-source operating data is compared with the corresponding adaptive early warning threshold to calculate a multi-dimensional risk feature vector characterizing the degree of deviation of the container's operating status. Based on a multidimensional risk feature vector and combined with the time lag effect relationship between operational data, a comprehensive risk assessment value is calculated. Based on the correspondence between the comprehensive risk assessment value and the preset risk level range, the early warning level of the power metering box is determined, and the early warning information corresponding to the early warning level is output.

[0007] Furthermore, performing time alignment processing on multi-source runtime data and calculating data trust weights includes: Operational data from different sensor sources are resampled and interpolated along a unified time axis to form a multi-source data vector aligned at the same sampling time t. For the i-th type of operational data, calculate its data integrity index within a preset time window. Data stability indicators and historical consistency indicators ,in:

[0008] in, This represents the number of missing sample points for the i-th type of runtime data within the time window. Indicates the total number of sampling points;

[0009] in, The standard deviation of the running data within the time window. The mean, To prevent constants with a denominator of zero;

[0010] in, This represents the running data value at the current sampling time. This represents the average value of the corresponding running data within the historical reference window; Data credibility weights satisfy:

[0011] in, Data credibility weight.

[0012] Furthermore, constructing an individualized operating baseline model and generating adaptive early warning thresholds includes: For various types of operational data after being weighted by data credibility weights, a corresponding dynamic operational baseline is constructed within a preset historical time window; For the i-th type of runtime data, calculate its weighted historical mean within the historical time window. Weighted historical dispersion ,in:

[0013]

[0014] in, Indicates a historical time window, This represents the value of the i-th type of running data at time t. This indicates the data reliability weight at the corresponding time point; Based on the weighted historical mean and weighted historical dispersion, an adaptive early warning threshold is generated for the i-th type of operational data. The adaptive warning threshold satisfies:

[0015] in, The threshold sensitivity coefficient is adaptively adjusted according to the operating phase, environment type, or time period. When changes occur in the operational phase, environment type, or time period, the historical time window is adjusted. and threshold sensitivity coefficient Update.

[0016] Furthermore, the calculation of the comprehensive risk assessment value based on multidimensional risk feature vectors includes: Based on the adaptive early warning threshold, the deviation of the running data at the current sampling time and its historical lag time is calculated to obtain multidimensional lag risk characteristics; Wherein, the normalized deviation of the i-th type of running data at lag k time is... Determine as follows:

[0017] in, This represents the value of the i-th type of running data lagging forward k sampling periods from the current time t; Based on the differences in the degree of risk impact of different lag orders, a lag attenuation factor is introduced. Based on this, a comprehensive risk assessment value R is constructed, which satisfies the following:

[0018] in, The lag order k decreases monotonically and is used to characterize the weight of the impact of historical operating status on current risk. Based on the trend of changes in the comprehensive risk assessment value over time, the cumulative or sudden characteristics of the risk are determined.

[0019] Furthermore, the multidimensional hysteresis risk characteristics further include mechanistic features constructed based on equipment failure mechanisms, and the mechanistic features include at least one of the following or a combination thereof: Condensation risk characteristics are constructed based on the air state parameters inside the enclosure; water ingress or sealing degradation characteristics are constructed based on the humidity change characteristics inside the enclosure over time; and insulation degradation characteristics are constructed based on the electrical state parameters of the metering circuit. Among them, the risk characteristics of condensation , This indicates the dew point temperature calculated based on the internal temperature and humidity of the enclosure. Indicates the temperature of the inner wall of the chamber; Humidity change characteristic quantity , This indicates the change in relative humidity inside the enclosure within a preset time window. Indicates the corresponding time length; Insulation degradation characteristic quantity , This indicates the insulation resistance value at the current moment.

[0020] Furthermore, before issuing the warning information, a warning confirmation and escalation step is included. The warning confirmation and escalation step includes: When the calculated comprehensive risk assessment value reaches the preset early warning triggering condition, the current operating status of the power metering box will be switched to the early warning confirmation status. When the early warning is confirmed, increase the sampling frequency of operational data related to the comprehensive risk assessment value, and continuously collect operational data within the preset confirmation time window; The comprehensive risk assessment value is recalculated based on the operational data collected within the confirmation time window, and the recalculated comprehensive risk assessment value is compared with the early warning triggering conditions. When the recalculated comprehensive risk assessment value continues to meet the warning triggering conditions, the current warning is confirmed as a valid warning, and the corresponding warning level is determined according to the magnitude or trend of the comprehensive risk assessment value. For different warning levels, output the corresponding warning information or implement different handling strategies.

[0021] Furthermore, the metering operation status data also includes structural health data to characterize the structural stability of the power metering box. The method also includes a closed-loop learning step to update the early warning model based on the operation and maintenance results, wherein: Structural health data includes enclosure vibration data, attitude change data, or a combination thereof. By extracting features from the structural health data, the installation status of the enclosure, the loosening status of the structure, or the trend of changes in foundation stability can be characterized. Based on changes in structural health data, the method of monitoring the installation status of the enclosure based on three-dimensional geometric scanning can be replaced or supplemented to achieve continuous online assessment of the structural health status of the enclosure. After the early warning information is output and the corresponding operation and maintenance handling is completed, the actual operation and maintenance results corresponding to the early warning information are obtained, and the operation and maintenance results are associated and stored with the operation data and comprehensive risk assessment results when the early warning was triggered. Based on historical data stored in the association, the operating baseline model, early warning thresholds, or risk assessment strategies are updated.

[0022] Secondly, this application provides an early warning system for electricity metering boxes, used to implement the early warning method for electricity metering boxes as described in the first aspect, the system comprising: The data acquisition module is used to collect multi-source operating data of the power metering box within a preset sampling period. The multi-source operating data includes internal environmental data of the box, external environmental data of the box, and metering operation status data. The data processing module communicates with the data acquisition module and is used to perform time alignment processing on multi-source running data and process various types of running data to obtain running datasets for early warning analysis. The baseline modeling and threshold generation module communicates with the data processing module and is used to build an operating baseline model of the power metering box based on the operating dataset and generate corresponding early warning thresholds. The risk assessment module communicates with the baseline modeling and threshold generation module and is used to calculate the comprehensive risk assessment value based on the operational data and early warning thresholds within the current sampling period. The early warning determination and output module is connected in communication with the risk assessment module. It is used to determine the early warning level of the power metering box based on the correspondence between the comprehensive risk assessment value and the preset risk level range, and output the early warning information corresponding to the early warning level.

[0023] Thirdly, this application provides a terminal, including: The memory is used to store the early warning program of the power metering box; A processor is configured to implement the steps of the power meter box early warning method as described in the first aspect when executing the power meter box early warning device.

[0024] Fourthly, this application provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the power metering box early warning method as described in the first aspect.

[0025] As can be seen from the above technical solutions, the advantages of the present invention are: By uniformly collecting and processing multi-source operating data from power metering boxes, a comprehensive analysis of the internal and external environments of the boxes and their metering operation status can be achieved. This enables a more comprehensive characterization of the actual operating status of power metering boxes under complex outdoor operating conditions, avoiding the problem of insufficient information caused by relying solely on a single environmental parameter for judgment.

[0026] By introducing a data credibility weighting calculation mechanism into the multi-source operational data processing process, and comprehensively considering factors such as data integrity, stability, and historical consistency, operational data from different sources and of different qualities are processed differently. This reduces the interference of abnormal sampling, data loss, or transient fluctuations on the early warning results and improves the stability and reliability of the early warning analysis process.

[0027] By constructing an individualized operating baseline model for each power metering box and generating an adaptive early warning threshold based on the operating baseline model, the early warning judgment can be dynamically adjusted in combination with the operating history and environmental characteristics of the power metering box itself. This avoids the problem of insufficient adaptability of fixed thresholds in different scenarios, thereby improving the pertinence and accuracy of the early warning results.

[0028] By introducing an analysis mechanism that considers the historical lag effect of operational data, the risk assessment process comprehensively considers the cumulative or evolving effects of the current and historical states on operational risks. This enables the comprehensive risk assessment results to reflect the persistence and development trend of risks, helping to identify potential operational hazards in advance, rather than being limited to responses to momentary anomalies.

[0029] By constructing mechanistic characteristic quantities based on the failure mechanism of power metering boxes, physical or electrical changes closely related to equipment reliability, such as condensation, water ingress, and insulation degradation, are incorporated into the early warning analysis process. This enables the early warning results to more accurately point to potential failure modes and provide clear references for subsequent operation and maintenance.

[0030] By setting up an early warning confirmation and tiered escalation mechanism, abnormal risks are reconfirmed before outputting early warning information, and different levels of early warning results are output according to the degree of risk. This effectively suppresses false alarms caused by short-term environmental disturbances or occasional data anomalies, and improves the credibility and practicality of early warning results.

[0031] By introducing a box stability assessment method based on structural health data, and feeding the actual operation and maintenance results back to the early warning model after the operation and maintenance is completed, the system can continuously update the operating baseline, early warning threshold, or risk assessment strategy. This enables the early warning system to continuously optimize its judgment capabilities as the operating time increases, thereby improving the long-term effectiveness and intelligence level of power metering box operation risk management. Attached Figure Description

[0032] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This is a flowchart of the power metering box early warning method of the present invention; Figure 2 This is an architectural diagram of the power metering box early warning system of the present invention. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] Please see Figure 1 As shown, this application provides an early warning method for electricity metering boxes, including the following steps: Step S1: Within the preset sampling period, collect multi-source operating data of the power metering box. The multi-source operating data includes internal environmental data of the box, external environmental data of the box, and metering operation status data. In some embodiments, the internal environmental data of the power metering box may include temperature, humidity, gas concentration, or a combination thereof, to characterize the environmental changes inside the box. The external environmental data may include ambient temperature, ambient humidity, rainfall, wind speed, external electromagnetic field strength, or a combination thereof, to reflect the external operating environment of the power metering box. The metering operation status data may include the operating status parameters, electrical operating parameters, or metering-related status information of the metering device. In specific implementation, various types of operational data can be collected by sensors deployed inside and outside the power metering box, and periodically acquired according to a preset sampling period to form a multi-source operational data set.

[0036] Step S2: Perform time alignment processing on multi-source running data, and calculate the corresponding data trust weights for various types of running data based on data integrity, data stability and historical consistency; In some embodiments, since different types of operational data may originate from different sensing devices, their sampling frequencies, sampling times, or transmission delays may differ. Therefore, before analyzing multi-source operational data, time alignment processing can be performed on various types of operational data. This can be achieved, for example, through resampling, interpolation, or time window matching, so that operational data from different data sources form corresponding datasets at the same sampling time. After time alignment, the credibility of various types of operational data can be assessed based on factors such as data missingness, short-term fluctuations, and deviations from historical operational states. This allows for the assignment of different data credibility weights to different operational data, reducing the impact of abnormal or low-quality data on subsequent early warning analysis results.

[0037] Step S3: Based on the historical statistical distribution of multi-source operation data, construct an individualized operation baseline model for each power metering box, and generate an adaptive early warning threshold on the operation baseline model according to the operation data of the current time period. In some embodiments, multi-source operational data from historical operation can be stored for each power metering box, and an operational baseline model reflecting the normal operating status of the power metering box can be constructed based on the historical operational data. The operational baseline model can be used to describe the normal operating level of the power metering box under different time periods and environmental conditions. In specific applications, the operational baseline model can be invoked in conjunction with the operational data of the current time period to generate an adaptive early warning threshold that matches the current operating status of the power metering box, enabling the early warning threshold to be dynamically adjusted as the operating environment or operating status changes.

[0038] Step S4: Within the current sampling period, compare the weighted multi-source operating data with the corresponding adaptive early warning threshold, and calculate the multi-dimensional risk feature vector characterizing the degree of deviation of the container's operating status. In some embodiments, the data credibility weights obtained in step S2 can be used to weight the multi-source operational data, giving higher-credibility operational data a greater weight in the analysis process. Based on this, the weighted operational data is compared item by item with the adaptive warning thresholds generated in step S3 to determine the deviation of each type of operational data from its normal operating range. By comprehensively describing the degree of deviation of different types of operational data, a multi-dimensional risk feature vector can be constructed to characterize the current operational risk level of the power metering box, thus providing a foundation for subsequent risk assessment.

[0039] Step S5: Based on the multidimensional risk feature vector and combined with the time lag effect relationship between operational data, calculate the comprehensive risk assessment value; In some embodiments, the risk assessment process can consider not only the multidimensional risk feature vector within the current sampling period, but also the impact of historical sampling period operational data on the current operational risk, to conduct a comprehensive analysis of the multidimensional risk characteristics. For example, the cumulative or delayed effects of certain operational anomalies over time can be considered, thus introducing the time lag relationship between operational data in the comprehensive risk assessment. Through a comprehensive analysis of the current risk characteristics and historical risk trends, a comprehensive risk assessment value reflecting the overall operational risk level of the power metering box can be obtained.

[0040] Step S6: Based on the correspondence between the comprehensive risk assessment value and the preset risk level range, determine the early warning level of the power metering box and output the early warning information corresponding to the early warning level.

[0041] In some embodiments, multiple risk level ranges can be pre-set to distinguish different levels of operational risk. When the comprehensive risk assessment value falls into different risk level ranges, the system can determine that the power metering box is at the corresponding warning level and output corresponding warning information. The warning information can be used to prompt maintenance personnel to pay attention to the operating status of the power metering box, or as a reference for subsequent maintenance and handling, thereby realizing hierarchical management of the operational risks of the power metering box.

[0042] In some embodiments, performing time alignment processing on multi-source runtime data and calculating data trust weights includes: Operational data from different sensor sources are resampled and interpolated along a unified time axis to form a multi-source data vector aligned at the same sampling time t. In the specific implementation process, different types of operational data may be collected by different sensors, and their sampling periods, sampling start times or communication delays may differ. By resampling and interpolating various types of operational data, the originally asynchronous data can be mapped onto a unified time axis, so that a complete set of multi-source operational data can be obtained at the same sampling time, thus providing a basis for subsequent joint analysis.

[0043] For the i-th type of operational data, calculate its data integrity index within a preset time window. Data stability indicators and historical consistency indicators ,in:

[0044] in, This represents the number of missing sample points for the i-th type of runtime data within the time window. Indicates the total number of sampling points;

[0045] in, The standard deviation of the running data within the time window. The mean, To prevent constants with a denominator of zero;

[0046] in, This represents the running data value at the current sampling time. This represents the average value of the corresponding running data within the historical reference window; Data credibility weights satisfy:

[0047] in, Data credibility weight.

[0048] In some embodiments, a separate time window can be set for each type of operational data. Within this time window, statistical analysis can be performed on the data collection and change characteristics to evaluate the data quality from multiple dimensions, so as to distinguish the credibility of different operational data in subsequent analysis.

[0049] By analyzing the missing data within a statistical time window, we can reflect the completeness of this type of operational data during the collection process. When there are many missing sampling points, it indicates that the reliability of this type of operational data is low, thus reducing its influence weight in subsequent analysis.

[0050] By analyzing the fluctuation of operational data within a time window, the stability of this type of operational data can be reflected. When the operational data fluctuates significantly in a short period of time, its stability is poor and it may be affected by external interference or sensor malfunctions.

[0051] By comparing the current sampling data with the historical reference state, it can be determined whether there is a significant deviation in this type of operational data, thus reflecting its consistency with the historical operational state.

[0052] In some embodiments, by comprehensively calculating data integrity, stability, and historical consistency, normalized data credibility weights can be assigned to different operational data, so that data with higher credibility has a greater influence in subsequent risk analysis, while the impact of data with lower credibility on the early warning results is correspondingly weakened.

[0053] In some embodiments, constructing an individualized operating baseline model and generating adaptive warning thresholds includes: For various types of operational data after being weighted by data credibility weights, a corresponding dynamic operational baseline is constructed within a preset historical time window; In the specific implementation process, historical operating data can be stored for each power meter box, and a dynamic operating baseline reflecting the normal operating status of the power meter box can be formed on this basis. This allows the operating baseline to reflect the operating characteristics of the meter box itself, rather than using a uniform fixed standard.

[0054] For the i-th type of runtime data, calculate its weighted historical mean within the historical time window. Weighted historical dispersion ,in:

[0055]

[0056] in, Indicates a historical time window, This represents the value of the i-th type of running data at time t. This indicates the data reliability weight at the corresponding time point; Based on the weighted historical mean and weighted historical dispersion, an adaptive early warning threshold is generated for the i-th type of operational data. The adaptive warning threshold satisfies:

[0057] in, The threshold sensitivity coefficient is adaptively adjusted according to the operating phase, environment type, or time period. When changes occur in the operational phase, environment type, or time period, the historical time window is adjusted. and threshold sensitivity coefficient Update.

[0058] By introducing data credibility weights and weighting historical operating data, the impact of abnormal or low-quality data can be reduced when calculating operating baseline parameters, making the operating baseline more stable and reliable.

[0059] In some embodiments, the historical time window can be set according to the actual application scenario, such as being updated by day, week or month, to ensure that the operating baseline can reflect the recent real operating status.

[0060] By introducing a fluctuation range based on the operating baseline, early warning thresholds that adjust with changes in operating status can be generated for different operating data, thereby avoiding the problem of insufficient adaptability of fixed thresholds in complex environments.

[0061] In some embodiments, different threshold sensitivity levels can be set for different operating stages or different environmental conditions, so that the early warning mechanism can avoid frequent false alarms while ensuring sensitivity.

[0062] By dynamically updating the parameters for establishing the operational baseline, the warning threshold can be continuously adapted to changes in operational conditions.

[0063] In some embodiments, calculating the comprehensive risk assessment value based on a multidimensional risk feature vector includes: Based on the adaptive early warning threshold, the deviation of the running data at the current sampling time and its historical lag time is calculated to obtain multidimensional lag risk characteristics; In practice, it is possible not only to analyze the operational status within the current sampling period, but also to combine the operational data from historical sampling periods to reflect the evolution of operational risks from a time perspective.

[0064] Wherein, the normalized deviation of the i-th type of running data at lag k time is... Determine as follows:

[0065] in, This represents the value of the i-th type of running data lagging forward k sampling periods from the current time t; By performing normalized deviation analysis on the operational data at historical lag times, the persistence or cumulative nature of abnormal states over time can be reflected.

[0066] Based on the differences in the degree of risk impact of different lag orders, a lag attenuation factor is introduced. Based on this, a comprehensive risk assessment value R is constructed, which satisfies the following:

[0067] in, The lag order k decreases monotonically and is used to characterize the weight of the impact of historical operating status on current risk. By weighting and integrating risk characteristics with different time lags, a comprehensive risk assessment result reflecting the overall operational risk level can be obtained.

[0068] Based on the trend of changes in the comprehensive risk assessment value over time, the cumulative or sudden characteristics of the risk are determined.

[0069] By analyzing the changing trends of the comprehensive risk assessment value, we can distinguish between long-term cumulative risks and short-term sudden risks.

[0070] In some embodiments, the multidimensional hysteresis risk characteristics further include mechanistic features constructed based on the equipment failure mechanism, wherein the mechanistic features include at least one of the following or a combination thereof: Condensation risk characteristics are constructed based on the air state parameters inside the enclosure; water ingress or sealing degradation characteristics are constructed based on the humidity change characteristics inside the enclosure over time; and insulation degradation characteristics are constructed based on the electrical state parameters of the metering circuit. In the specific implementation process, physical or electrical mechanisms closely related to equipment reliability can be introduced into risk analysis to enhance the directional accuracy of early warning results towards actual failure modes.

[0071] Among them, the risk characteristics of condensation , This indicates the dew point temperature calculated based on the internal temperature and humidity of the enclosure. Indicates the temperature of the inner wall of the chamber; By analyzing the relationship between dew point temperature and inner wall temperature, it is possible to determine whether there is a risk of condensation inside the enclosure.

[0072] Humidity change characteristic quantity , This indicates the change in relative humidity inside the enclosure within a preset time window. Indicates the corresponding time length; By analyzing the rate of humidity change, abnormal situations such as water ingress and decreased sealing performance can be identified.

[0073] Insulation degradation characteristic quantity , This indicates the insulation resistance value at the current moment.

[0074] By monitoring changes in insulation resistance, the changing trend of the insulation status of the metering circuit can be reflected.

[0075] In some embodiments, a warning confirmation and escalation step is included before outputting the warning information. The warning confirmation and escalation step includes: When the calculated comprehensive risk assessment value reaches the preset early warning triggering condition, the current operating status of the power metering box will be switched to the early warning confirmation status. In some embodiments, when an abnormal risk is initially determined to exist, the warning information may not be output directly, but the warning confirmation process may be initiated instead.

[0076] When the early warning is confirmed, increase the sampling frequency of operational data related to the comprehensive risk assessment value, and continuously collect operational data within the preset confirmation time window; By increasing the sampling frequency, more intensive operational data can be obtained to verify whether the abnormal state is persistent.

[0077] The comprehensive risk assessment value is recalculated based on the operational data collected within the confirmation time window, and the recalculated comprehensive risk assessment value is compared with the early warning triggering conditions. In some embodiments, false risks caused by short-term disturbances or occasional anomalies can be effectively filtered out by recalculating the risk assessment value.

[0078] When the recalculated comprehensive risk assessment value continues to meet the warning triggering conditions, the current warning is confirmed as a valid warning, and the corresponding warning level is determined according to the magnitude or trend of the comprehensive risk assessment value. This confirmation mechanism can improve the credibility of early warning results.

[0079] For different warning levels, output the corresponding warning information or implement different handling strategies.

[0080] In some embodiments, different warning levels may correspond to different operation and maintenance response measures.

[0081] In some embodiments, the metering operation status data further includes structural health data characterizing the structural stability of the power metering box, and the method further includes a closed-loop learning step for updating the early warning model based on the operation and maintenance results, wherein: Structural health data includes enclosure vibration data, attitude change data, or a combination thereof. By extracting features from the structural health data, the installation status of the enclosure, the loosening status of the structure, or the trend of changes in foundation stability can be characterized. In the specific implementation process, the structural status information of the box can be continuously obtained through vibration or attitude monitoring, thereby realizing online assessment of the structural health status.

[0082] Based on changes in structural health data, the method of monitoring the installation status of the enclosure based on three-dimensional geometric scanning can be replaced or supplemented to achieve continuous online assessment of the structural health status of the enclosure. By employing structural health data, reliance on complex field scanning equipment can be reduced, increasing the flexibility of system deployment.

[0083] After the early warning information is output and the corresponding operation and maintenance handling is completed, the actual operation and maintenance results corresponding to the early warning information are obtained, and the operation and maintenance results are associated and stored with the operation data and comprehensive risk assessment results when the early warning was triggered. By linking the early warning results with the actual handling situation, a traceable operational record can be formed.

[0084] Based on historical data stored in the association, the operating baseline model, early warning thresholds, or risk assessment strategies are updated.

[0085] By continuously incorporating operation and maintenance feedback information, the early warning method can be continuously optimized during long-term operation, thereby improving the accuracy and adaptability of early warning of operational risks of power metering boxes.

[0086] Please see Figure 2 As shown, in some embodiments, this application provides an early warning system for power metering boxes, used to implement a method for early warning of power metering boxes, the system comprising: The data acquisition module is used to collect multi-source operating data of the power metering box within a preset sampling period. The multi-source operating data includes internal environmental data of the box, external environmental data of the box, and metering operation status data. The data processing module communicates with the data acquisition module and is used to perform time alignment processing on multi-source running data and process various types of running data to obtain running datasets for early warning analysis. The baseline modeling and threshold generation module communicates with the data processing module and is used to build an operating baseline model of the power metering box based on the operating dataset and generate corresponding early warning thresholds. The risk assessment module communicates with the baseline modeling and threshold generation module and is used to calculate the comprehensive risk assessment value based on the operational data and early warning thresholds within the current sampling period. The early warning determination and output module is connected in communication with the risk assessment module. It is used to determine the early warning level of the power metering box based on the correspondence between the comprehensive risk assessment value and the preset risk level range, and output the early warning information corresponding to the early warning level.

[0087] In some embodiments, this application provides a terminal, including: The memory is used to store the early warning program of the power metering box; A processor is used to execute the steps of the power metering box early warning method when the power metering box early warning system is implemented.

[0088] In some embodiments, this application provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the power metering box early warning method.

[0089] The above description is merely a preferred embodiment of one or more embodiments of this specification and is not intended to limit the scope of one or more embodiments of this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of this specification should be included within the protection scope of one or more embodiments of this specification.

Claims

1. A power metering box early warning method, characterized in that, Includes the following steps: Within a preset sampling period, multi-source operational data of the power metering box is collected. The multi-source operational data includes internal environmental data of the box, external environmental data of the box, and metering operation status data. Time alignment processing is performed on multi-source operational data, and data trust weights are calculated for various types of operational data based on data integrity, data stability, and historical consistency. Based on the historical statistical distribution of multi-source operation data, an individualized operation baseline model is constructed for each power metering box, and an adaptive early warning threshold is generated on the operation baseline model according to the operation data of the current period. Within the current sampling period, the weighted multi-source operating data is compared with the corresponding adaptive early warning threshold to calculate a multi-dimensional risk feature vector characterizing the degree of deviation of the container's operating status. Based on a multidimensional risk feature vector and combined with the time lag effect relationship between operational data, a comprehensive risk assessment value is calculated. Based on the correspondence between the comprehensive risk assessment value and the preset risk level range, the early warning level of the power metering box is determined, and the early warning information corresponding to the early warning level is output.

2. The power metering box early warning method according to claim 1, characterized in that, Performing time alignment processing on multi-source runtime data and calculating data confidence weights includes: Operational data from different sensor sources are resampled and interpolated along a unified time axis to form a multi-source data vector aligned at the same sampling time t. For the i-th type of operation data, its data integrity index, data stability index and history consistency index are calculated respectively within a preset time window wherein:​​ wherein, represents the number of missing sampling points of the i-th type of running data within the time window, represents the total number of sampling points; wherein, is the standard deviation of the operational data within the time window, is the mean value, is a constant to prevent the denominator from being zero; in, This represents the running data value at the current sampling time. This represents the average value of the corresponding running data within the historical reference window; Data credibility weights satisfy: in, Data credibility weight.

3. The power metering box early warning method according to claim 2, characterized in that, Building an individualized operational baseline model and generating adaptive early warning thresholds includes: For various types of operational data after being weighted by data credibility weights, a corresponding dynamic operational baseline is constructed within a preset historical time window; For the i-th type of runtime data, calculate its weighted historical mean within the historical time window. Weighted historical dispersion ,in: in, Indicates a historical time window, This represents the value of the i-th type of running data at time t. This indicates the data reliability weight at the corresponding time point; Based on the weighted historical mean and weighted historical dispersion, an adaptive early warning threshold is generated for the i-th type of operational data. The adaptive warning threshold satisfies: in, The threshold sensitivity coefficient is adaptively adjusted according to the operating phase, environment type, or time period. When changes occur in the operational phase, environment type, or time period, the historical time window is adjusted. and threshold sensitivity coefficient Update.

4. The power metering box early warning method according to claim 3, characterized in that, The comprehensive risk assessment value calculated based on multidimensional risk feature vectors includes: Based on the adaptive early warning threshold, the deviation of the running data at the current sampling time and its historical lag time is calculated to obtain multidimensional lag risk characteristics; Wherein, the normalized deviation of the i-th type of running data at lag k time is... Determine as follows: in, This represents the value of the i-th type of running data lagging forward k sampling periods from the current time t; Based on the differences in the degree of risk impact of different lag orders, a lag attenuation factor is introduced. Based on this, a comprehensive risk assessment value R is constructed, which satisfies the following: in, The lag order k decreases monotonically and is used to characterize the weight of the impact of historical operating status on current risk. Based on the trend of changes in the comprehensive risk assessment value over time, the cumulative or sudden characteristics of the risk are determined.

5. The power metering box early warning method according to claim 4, characterized in that, The multidimensional hysteresis risk characteristics further include mechanistic features constructed based on equipment failure mechanisms, and the mechanistic features include at least one of the following or a combination thereof: Condensation risk characteristics are constructed based on the air state parameters inside the enclosure; water ingress or sealing degradation characteristics are constructed based on the humidity change characteristics inside the enclosure over time; and insulation degradation characteristics are constructed based on the electrical state parameters of the metering circuit. Among them, the risk characteristics of condensation , This indicates the dew point temperature calculated based on the internal temperature and humidity of the enclosure. Indicates the temperature of the inner wall of the chamber; Humidity change characteristic quantity , This indicates the change in relative humidity inside the enclosure within a preset time window. Indicates the corresponding time length; Insulation degradation characteristic quantity , This indicates the insulation resistance value at the current moment.

6. The power metering box early warning method according to claim 4 or 5, characterized in that, Before issuing the warning information, a warning confirmation and escalation process is also included. The warning confirmation and escalation process includes: When the calculated comprehensive risk assessment value reaches the preset early warning triggering condition, the current operating status of the power metering box will be switched to the early warning confirmation status. When the early warning is confirmed, increase the sampling frequency of operational data related to the comprehensive risk assessment value, and continuously collect operational data within the preset confirmation time window; The comprehensive risk assessment value is recalculated based on the operational data collected within the confirmation time window, and the recalculated comprehensive risk assessment value is compared with the early warning triggering conditions. When the recalculated comprehensive risk assessment value continues to meet the warning triggering conditions, the current warning is confirmed as a valid warning, and the corresponding warning level is determined according to the magnitude or trend of the comprehensive risk assessment value. For different warning levels, output the corresponding warning information or implement different handling strategies.

7. The power metering box early warning method according to claim 1, characterized in that, The metering operation status data also includes structural health data to characterize the structural stability of the power metering box. The method also includes a closed-loop learning step to update the early warning model based on operation and maintenance results, wherein: Structural health data includes enclosure vibration data, attitude change data, or a combination thereof. By extracting features from the structural health data, the installation status of the enclosure, the loosening status of the structure, or the trend of changes in foundation stability can be characterized. Based on changes in structural health data, the method of monitoring the installation status of the enclosure based on three-dimensional geometric scanning can be replaced or supplemented to achieve continuous online assessment of the structural health status of the enclosure. After the early warning information is output and the corresponding operation and maintenance handling is completed, the actual operation and maintenance results corresponding to the early warning information are obtained, and the operation and maintenance results are associated and stored with the operation data and comprehensive risk assessment results when the early warning was triggered. Based on historical data stored in the association, the operating baseline model, early warning thresholds, or risk assessment strategies are updated.

8. A power metering box early warning system, used to implement the power metering box early warning method as described in claim 1, characterized in that, The system includes: The data acquisition module is used to collect multi-source operating data of the power metering box within a preset sampling period. The multi-source operating data includes internal environmental data of the box, external environmental data of the box, and metering operation status data. The data processing module communicates with the data acquisition module and is used to perform time alignment processing on multi-source running data and process various types of running data to obtain running datasets for early warning analysis. The baseline modeling and threshold generation module communicates with the data processing module and is used to build an operating baseline model of the power metering box based on the operating dataset and generate corresponding early warning thresholds. The risk assessment module communicates with the baseline modeling and threshold generation module and is used to calculate the comprehensive risk assessment value based on the operational data and early warning thresholds within the current sampling period. The early warning determination and output module is connected in communication with the risk assessment module. It is used to determine the early warning level of the power metering box based on the correspondence between the comprehensive risk assessment value and the preset risk level range, and output the early warning information corresponding to the early warning level.

9. A terminal, characterized in that, include: The memory is used to store the early warning program of the power metering box; A processor is configured to implement the steps of the power meter box early warning method as described in claim 1 when executing the power meter box early warning device.

10. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions. When the computer reads the computer instructions from the storage medium, the computer executes the power metering box early warning method as described in claim 1.

Images