Electric energy meter error state evaluation method and device, electronic equipment and storage medium

Abstract

The application provides an electric energy meter error state evaluation method and device, electronic equipment and storage medium, and belongs to the technical field of electric power metering monitoring. The method comprises the following steps: acquiring a current electric energy value of a master meter and a current electric energy value of a slave meter, combining a historical electric energy sequence of the master meter and the slave meter, and determining a trend consistency state of the master meter and the slave meter; fusing the current electric energy value of the master meter and the current electric energy value of the slave meter according to the trend consistency state to obtain reference electric energy data; acquiring an electric energy display value of each sub-meter, and adopting a trained time sequence analysis model to analyze and process the reference electric energy data of a continuous time sequence and the electric energy display value of each sub-meter to obtain an electric energy prediction value corresponding to each sub-meter; and comparing the electric energy prediction value with the electric energy display value for each sub-meter to obtain an error evaluation result of the sub-meter. The application improves the timeliness and accuracy of electric energy meter error state evaluation.

Description

Technical Field

[0001] This invention relates to the field of power metering and monitoring technology, specifically to a method, device, electronic equipment, and storage medium for assessing the error status of an electricity meter. Background Technology

[0002] High-voltage energy meters are core metering devices in power systems for electricity trade settlement. The accuracy of their metering results and the reliability of their operation directly affect the fairness and economic interests of electricity bill settlement among power grid companies, power generation companies, and large users. According to relevant national metrology regulations and power industry standards, high-voltage energy meters are mandatory verification instruments and must undergo regular on-site calibration and periodic verification to ensure that their metering performance meets the required specifications.

[0003] In current operation and maintenance, monitoring the error status of high-voltage electricity meters mainly relies on periodic on-site manual verification and offline data analysis. This approach has significant limitations: First, it is essentially a discrete sampling inspection, unable to capture and warn of metering error drift caused by factors such as changes in environmental temperature and humidity, electromagnetic interference, mechanical vibration, and aging of internal components during long-term continuous operation. This increases the risk of metering inaccuracies and affects the fairness and reliability of electricity market transactions. Second, when assessing the status of electricity meters, traditional statistical methods or simple classification models are insufficient to effectively extract deep patterns from electrical energy data.

[0004] In summary, the existing method of periodic on-site manual verification and offline data analysis has the drawbacks of inaccurate power metering and untimely error correction. Summary of the Invention

[0005] In view of this, it is necessary to provide a method, device, electronic device and storage medium for assessing the error status of electricity meters, in order to solve the technical problems of inaccurate electricity metering and untimely error correction caused by the existing methods of periodic on-site manual verification and offline data analysis.

[0006] To address the aforementioned technical problems, in a first aspect, the present invention provides a method for evaluating the error status of an electricity meter, comprising:

[0007] Obtain the current energy value of the main table and the current energy value of the secondary table, and combine them with the historical energy sequences of the main table and the secondary table to determine the trend consistency status of the main table and the secondary table;

[0008] Based on the trend consistency status, the current energy value of the main table and the current energy value of the secondary table are merged to obtain reference energy data;

[0009] The power consumption display value of each sub-meter is obtained, and the reference power consumption data and the power consumption display value of each sub-meter are analyzed and processed using a trained time series analysis model to obtain the power consumption prediction value corresponding to each sub-meter.

[0010] For each sub-meter, the predicted power consumption value is compared with the displayed power consumption value to obtain the error assessment result of the sub-meter.

[0011] In one possible implementation, before obtaining the current energy value of the main meter and the current energy value of the secondary meter, the energy meter error status assessment method further includes:

[0012] Acquire synchronous monitoring data of the master table and each sub-table within the target area, wherein the master table includes a main table and sub-tables;

[0013] A metering topology analysis was performed on the synchronous monitoring data to obtain topology verification results;

[0014] When the topology verification result is successful, the step of obtaining the current power value of the main table and the current power value of the secondary table is executed.

[0015] In one possible implementation, the synchronous monitoring data includes three-phase voltage amplitude, three-phase voltage phase angle, three-phase current amplitude, three-phase current phase angle, and power factor. The step of performing metering topology analysis on the synchronous monitoring data to obtain topology verification results includes:

[0016] Based on the three-phase voltage amplitude, three-phase voltage phase angle, three-phase current amplitude, and three-phase current phase angle in the synchronous monitoring data, calculate the instantaneous active power and average phase difference of the main meter and each sub-meter;

[0017] For any master table or sub-table, the power flow direction is determined based on the instantaneous active power, and the power flow direction is matched with a preset initial direction to obtain a first matching result;

[0018] For any master table or sub-table, the average phase difference and power factor are matched to obtain a second matching result;

[0019] Based on the first matching result and the second matching result, the topology verification result is determined.

[0020] In one possible implementation, obtaining the current energy value of the main table and the current energy value of the secondary table, and combining the historical energy sequences of the main table and the secondary table to determine the trend consistency status of the main table and the secondary table, includes:

[0021] Collect the T adjacent to the current time period Y The historical energy value sequence of the main meter and the historical energy value sequence of the secondary meter within a time period;

[0022] Based on the historical energy value sequence of the main meter, the historical energy value sequence of the secondary meter, the current energy value of the main meter and the current energy value of the secondary meter, the sliding standard deviation of the energy difference between the main meter and the secondary meter is determined, and the sliding standard deviation is used as the dynamic consistency threshold.

[0023] The absolute value of the difference between the current power value of the main table and the current power value of the secondary table is compared with the dynamic consistency threshold to obtain the trend consistency status of the main table and the secondary table.

[0024] In one possible implementation, the step of fusing the current energy value of the main table and the current energy value of the secondary table according to the trend consistency state to obtain reference energy data includes:

[0025] When the trend consistency status is consistent, the current energy value of the main table and the current energy value of the secondary table are merged by taking the arithmetic average to obtain the reference energy data;

[0026] When the consistency status is inconsistent, based on the energy values ​​of the main table and the secondary table within the time period, the time smoothness index and the load change trend conformity index of the main table and the secondary table are calculated respectively. The time smoothness index represents the smoothness of the jump during the time period, and the load change trend conformity index represents the degree of consistency between the load change and the recent trend.

[0027] The time smoothness index and the load change trend conformity index are used to generate the fusion weights;

[0028] Based on the fusion weight, the current energy value of the main table and the current energy value of the secondary table are weighted and fused to obtain the reference energy data.

[0029] In one possible implementation, the time series analysis model is a neural network model containing a Transformer encoder, the Transformer encoder including a self-attention mechanism, and the composite loss function used by the time series analysis model introduces a physical constraint of energy conservation so that the sum of the sub-meter energy data equals the total meter energy data.

[0030] In one possible implementation, after comparing the predicted energy value with the displayed energy value for each sub-meter to obtain the error assessment result for that sub-meter, the method further includes:

[0031] The method of comparing the error assessment result with a preset error threshold is used to determine whether there is an abnormal energy meter;

[0032] When an abnormal energy meter is detected, an alarm message is generated to issue a status alarm. The alarm message includes the identifier of the abnormal energy meter, the error assessment result, and a timestamp.

[0033] On the other hand, the present invention also provides an energy meter error status assessment device, comprising:

[0034] The status determination module is used to obtain the current energy value of the main table and the current energy value of the secondary table, and combine them with the historical energy sequence of the main table and the secondary table to determine the trend consistency status of the main table and the secondary table;

[0035] The data fusion module is used to fuse the current energy value of the main table and the current energy value of the secondary table according to the trend consistency status to obtain reference energy data;

[0036] The power prediction module is used to obtain the power display value of each sub-meter, and to use a trained time series analysis model to analyze and process the reference power data of the continuous time series and the power display value of each sub-meter to obtain the power prediction value corresponding to each sub-meter.

[0037] The result determination module is used to compare the predicted power consumption value with the displayed power consumption value for each sub-meter to obtain the error evaluation result of the sub-meter.

[0038] Thirdly, the present invention also provides an electronic device, including a memory and a processor, wherein,

[0039] The memory is used to store programs;

[0040] The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the energy meter error state assessment method described in any of the above implementations.

[0041] Fourthly, the present invention also provides a computer-readable storage medium for storing a computer-readable program or instruction, which, when executed by a processor, can implement the steps in the energy meter error state assessment method described in any of the above implementations.

[0042] The beneficial effects of this invention are as follows: The energy meter error status assessment method provided by this invention obtains the current energy value of the main meter and the current energy value of the sub-meter, and combines them with the historical energy sequences of the main and sub-meters to determine the trend consistency status of the main and sub-meters; based on the trend consistency status, the current energy values ​​of the main meter and the sub-meter are fused to obtain reference energy data, thereby making the reference energy data more objective and accurate, avoiding the impact of a single meter failure on the subsequent overall error assessment; simultaneously, by obtaining the energy display value of each sub-meter, and using a trained time series analysis model to analyze and process the continuous time series reference energy data and the energy display value of each sub-meter, the predicted energy value corresponding to each sub-meter is obtained; for each sub-meter, the predicted energy value is compared with the displayed energy value to obtain the sub-meter's error assessment result. This achieves real-time online assessment of energy errors using model prediction, improving the accuracy and timeliness of error assessment. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments 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.

[0044] Figure 1 A schematic diagram of an application scenario for the energy meter error status assessment method provided by the present invention;

[0045] Figure 2 A schematic flowchart of an embodiment of the energy meter error status assessment method provided by the present invention;

[0046] Figure 3 A schematic flowchart of another embodiment of the energy meter error status assessment method provided by the present invention;

[0047] Figure 4 For the present invention Figure 3 A schematic diagram of an embodiment of S302;

[0048] Figure 5 For the present invention Figure 2 A schematic diagram of an embodiment of S201;

[0049] Figure 6 For the present invention Figure 2 A schematic diagram of an embodiment of S202;

[0050] Figure 7 A schematic flowchart of another embodiment of the energy meter error status assessment method provided by the present invention;

[0051] Figure 8 A schematic diagram of an embodiment of the energy meter error status assessment device provided by the present invention;

[0052] Figure 9 A schematic diagram of an embodiment of the electronic device provided by the present invention. Detailed Implementation

[0053] 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 a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0054] In the description of the embodiments of the present invention, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.

[0055] The terms "first," "second," etc., used in the embodiments of this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a technical feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature.

[0056] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0057] Before demonstrating the embodiments, the following terms will be explained.

[0058] A high-voltage electricity meter is a metering device installed on the high-voltage side of a power system (usually 10kV and above) to measure the consumption or transmission of electrical energy.

[0059] Master Meter and Branch Meter are terms used to describe metering topology. The master meter is installed at the entrance or bus of the power supply area to measure the total power consumption of that area. Branch meters are installed at each branch circuit or user location to measure the power consumption of that branch. Under ideal, lossless conditions, the sum of the readings of all branch meters should equal the reading of the master meter; this reflects the law of conservation of energy in this scenario.

[0060] A primary meter and a secondary / redundant meter refer to two electricity meters installed in parallel at the same critical metering point (usually the location of the main meter), forming a redundant backup structure. The primary meter typically serves as the main metering and billing basis, while the secondary meter serves as a real-time comparison and backup. Both measure the same electrical circuit, and theoretically, their readings should be highly consistent.

[0061] Synchronized monitoring data refers to electrical quantity data collected from electricity meters located at different locations at the same absolute timestamp (or the same sampling period). Its synchronicity is fundamental for accurate power flow calculations and topology analysis, and typically includes voltage, current, phase, power factor, etc.

[0062] Instantaneous active power refers to the rate at which electrical energy is actually consumed or converted into other forms of energy (such as heat, light, and mechanical energy) per unit time, measured in watts (W) or kilowatts (kW). The sign (positive / negative) of its value is directly used to determine the actual direction of power flow and is one of the core criteria for topology verification.

[0063] Power flow direction refers to the direction of net transmission of electrical energy in the power grid. In this embodiment, based on the symbol definition of instantaneous active power: forward means that electrical energy flows from the power grid to the user (electricity consumption); reverse means that electrical energy flows from the user side to the power grid (such as distributed generation power transmission back).

[0064] Power factor (PF) is the ratio of active power to apparent power, numerically equal to the cosine of the phase difference angle between voltage and current. It reflects the efficiency of electrical energy utilization. In this embodiment, it is used to verify the inherent consistency of measured values ​​to identify hidden wiring or measurement faults.

[0065] Topology verification / topology status refers to the physical connection between the electricity meter and the electrical circuit. Topology verification involves analyzing electrical data (such as power flow direction and power factor) to determine whether the actual physical wiring conforms to the preset and correct connection relationship. Topology status is the verification result, which is divided into "normal" or "abnormal".

[0066] In this embodiment, the law of conservation of energy specifically refers to the physical constraint in electricity metering that "the total energy consumption of the main meter equals the sum of the energy consumption of each sub-meter." This law is incorporated into the loss function to guide the training of the deep learning model, ensuring that the model's output conforms to physical laws.

[0067] The Dynamic Consistency Threshold is a threshold that is automatically calculated and changes over time based on the fluctuations in historical data differences between the primary and secondary tables (usually expressed as sliding standard deviation). It is used to dynamically determine whether the primary and secondary table data are in a "consistent" state at any given moment, avoiding the limitations of fixed thresholds.

[0068] A time series analysis model is a mathematical model specifically designed to process data sequences arranged chronologically over time, extract patterns, make predictions, or detect anomalies. In this embodiment, it specifically refers to the core algorithm module used to learn the time series patterns of electrical energy data and predict the theoretical values ​​of sub-meters.

[0069] The Transformer Encoder is a deep learning network architecture based on the self-attention mechanism. It excels at capturing long-range dependencies in sequential data, and in this embodiment, it is used to efficiently model complex correlation patterns in electrical energy data over time.

[0070] Reconstruction Loss is the first term of the composite loss function in this embodiment. It measures the model's accuracy in reconstructing or predicting the input historical data of each sub-table, forcing the model to learn the temporal variation patterns of each sub-table itself.

[0071] The Physical Constraint Loss is the second term in the composite loss function of this embodiment. It measures whether the model's predictions violate the physical law of energy conservation, i.e., the deviation between the sum of the predicted values ​​of each sub-table and the reference value of the overall table. This term guides the model to respect physical laws while fitting the data.

[0072] Sliding standard deviation is a statistical method used to calculate the dispersion of a data sequence within a time window. In this embodiment, it is used to calculate the volatility of the historical difference sequence between the primary and secondary tables, thereby obtaining a dynamic consistency threshold.

[0073] The time smoothness index is used to quantify the degree of change in a data series between adjacent time points. The smoother the change, the higher the index value, which generally indicates that the data is more reliable.

[0074] The load change trend consistency index is used to determine whether the direction of change (increase / decrease) of the current data point is consistent with the overall trend formed by recent historical data. A higher score indicates that the data change follows inertia and is more reliable.

[0075] This invention provides a method, apparatus, electronic device, and storage medium for assessing the error status of an electricity meter. The technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0076] Figure 1 This is a schematic diagram illustrating an application scenario of the electricity meter error state assessment method provided by the present invention, such as... Figure 1 As shown, the application scenario includes a master table and X sub-tables. The energy direction of the master table and sub-tables is opposite. The data in the master table and sub-tables are both timestamped and the data is synchronized in time. At any fixed moment, the sum of the energy recorded in the sub-tables should be approximately equal to the energy recorded in the master table.

[0077] It should be noted that, as a preferred embodiment, the master table adopts a master table and a slave table to form a master-slave structure.

[0078] Figure 2 This is a schematic flowchart of an embodiment of the energy meter error status assessment method provided by the present invention. The method is applied to... Figure 1 The application scenarios shown are as follows: Figure 2 As shown, the method for assessing the error status of the electricity meter includes:

[0079] S201. Obtain the current energy value of the main table and the current energy value of the secondary table, and combine them with the historical energy sequences of the main table and the secondary table to determine the trend consistency status of the main table and the secondary table.

[0080] As a preferred approach, two tables—a master table and a slave table—are set up in the same location at the site, forming a primary and backup system. In an energy metering system, the values ​​of the master and slave tables should theoretically be highly consistent. Under normal circumstances, the average value can be directly taken as the reference energy data, assuming that "both tables are equally reliable." However, in reality, the reliability of the two tables changes dynamically. When the difference between the two values ​​is too large, at least one table will inevitably have poor reliability due to various reasons. Forcibly averaging in this case would introduce erroneous information into the reference benchmark, causing energy calculation, anomaly detection, and error assessment to all fail. In this embodiment, the reliability of the master and slave tables is determined by the trend consistency status, thereby deciding the subsequent fusion method.

[0081] The trend consistency status is used to determine how closely the main table and the sub-table approximate the theoretical trend. In this embodiment, data from historical energy series is used for analysis and reference determination.

[0082] S202. Based on the trend consistency status, merge the current energy value of the main table and the current energy value of the secondary table to obtain reference energy data.

[0083] Specifically, based on the trend consistency status of the main table and the secondary table, the corresponding fusion strategy is selected to fuse the current energy value of the main table and the current energy value of the secondary table to obtain reference energy data.

[0084] In one specific optional implementation of this embodiment, the trend consistency state is consistent, indicating that the main table and the sub-table are equally reliable. In this case, the average value is used for fusion. In another specific optional implementation of this embodiment, the trend consistency state is inconsistent, indicating that the main table and the sub-table are not equally reliable. In this case, the weights of the main table and the sub-table are calculated based on the electricity data collected within the time period, and then weighted fusion is performed.

[0085] S203. Obtain the energy display value of each sub-meter, and use the trained time series analysis model to analyze and process the reference energy data of the continuous time series and the energy display value of each sub-meter to obtain the energy prediction value corresponding to each sub-meter.

[0086] In some embodiments of the present invention, the time series analysis model is a neural network model containing a Transformer encoder. The Transformer encoder contains a self-attention mechanism. The composite loss function used in the time series analysis model introduces the physical constraint of energy conservation so that the sum of the sub-meter energy data equals the total meter energy data.

[0087] The core architecture of the time series analysis model used in this embodiment is the Transformer encoder. Through its core self-attention mechanism, especially causal self-attention, this model can effectively capture the time-series dependencies and inherent patterns in the input electrical energy sequence.

[0088] During implementation, the first step is to integrate the reference energy data from the master meter over T consecutive time periods. E 0 and the energy readings of each sub-meter E 1 to E X This forms an original sequence with T rows and (X+1) columns. Z :

[0089] .

[0090] Original sequence Z The mapping is performed as a high-dimensional vector representation to facilitate subsequent neural network processing. Specifically, formula (1) is used for each row. z t ( t Calculate the embedding vector for the number of rows:

[0091] (1)

[0092] W e The linear projection matrix maps the original multidimensional data to the latent space. p t The positional encoding vector uses fixed sine coding.

[0093] Then each row is embedded into a vector h t Composition of embedded sequences H , H =[ h 1, h 2, ..., h t The input is then fed into an L-layer Transformer encoder. Each layer of this encoder contains a multi-head self-attention sublayer and a feedforward neural network sublayer, and is optimized through residual connections and layer normalization to finally obtain an output representation containing complex temporal features. H out Based on output representation H out Each sub-table at any time can be reconstructed through a linear output layer. t Theoretical value of electrical energy Ê x (t) .

[0094] The training objective function of this time series analysis model consists of two parts. The first term is the reconstruction loss, which is the actual value of all sub-tables at each time step. E x (t) Compared with model predictions Ê x(t) The first term is the sum of the squares and mean of the differences between the sub-meters, intended to drive the model to accurately learn the historical variation patterns of each sub-meter. The second term is the physical relationship constraint term, which is specifically the sum of the predicted values ​​of each sub-meter at the same time and the reference energy value of the total meter. E 0 (t) The mean of the sum of squares of the differences. This implements the physical law of energy conservation, which states that "total electrical energy equals the sum of the electrical energy of each branch," as a soft constraint to guide model training.

[0095] As a preferred approach, this embodiment imports historical data from the sub-tables and the master table into the model for backpropagation during idle periods, updates the parameters using the Adam optimizer until the loss converges, and completes one round of model training, obtaining the model parameters after training. This allows for direct use of the trained model in real-time applications, improving prediction efficiency. During the prediction phase, the master table and sub-table data for a period prior to the predicted time are converted into matrices, imported into the trained model, and a forward propagation is performed to quickly obtain the model prediction values ​​for each sub-table at each time step.

[0096] This embodiment employs a Transformer encoder structure, which effectively models long-term temporal dependencies in electrical energy data, enhancing feature extraction capabilities. More importantly, it explicitly introduces an energy conservation physical constraint term into the objective function, forcing the time-series analysis model to conform to fundamental physical laws while learning statistical patterns in the data. This combination of data-driven and physical modeling significantly improves the physical rationality and overall accuracy of the time-series analysis model's predictions of theoretical values ​​for sub-metered electrical energy, thereby increasing the accuracy of subsequent error assessment results based on comparisons between predicted and actual values.

[0097] S204. For each sub-meter, compare the predicted energy value with the displayed energy value to obtain the sub-meter error assessment result.

[0098] In this embodiment, the current energy values ​​of the main table and the secondary table are obtained, and combined with the historical energy sequences of the main and secondary tables, the trend consistency status of the main and secondary tables is determined. Based on the trend consistency status, the current energy values ​​of the main table and the secondary table are fused to obtain reference energy data, making the reference energy data more objective and accurate, and avoiding the impact of a single table failure on the subsequent overall error assessment. Simultaneously, the energy display value of each sub-table is obtained, and a trained time-series analysis model is used to analyze and process the continuous time-series reference energy data and the energy display value of each sub-table to obtain the predicted energy value for each sub-table. For each sub-table, the predicted energy value is compared with the displayed energy value to obtain the sub-table's error assessment result. This achieves real-time online assessment of energy errors using model prediction, improving the accuracy and timeliness of error assessment.

[0099] In some embodiments of the present invention, such as Figure 3 As shown, before obtaining the current energy value of the main meter and the current energy value of the secondary meter, the energy meter error status assessment method also includes:

[0100] S301. Obtain the synchronous monitoring data of the master table and each sub-table within the target area. The master table includes the main table and the sub-table.

[0101] S302. Perform metrological topology analysis on the synchronous monitoring data to obtain topology verification results;

[0102] S303. When the topology verification result is successful, execute the step of obtaining the current power value of the main table and the current power value of the secondary table.

[0103] The synchronous monitoring data includes, but is not limited to, the voltage, current, and power data of each electricity meter.

[0104] Specifically, if the direction of electricity flow from the meter is incorrect due to unknown changes (e.g., construction errors, equipment replacements, etc.), and this is directly imported into the evaluation method, it will lead to implicit logical contradictions in the input data, causing the model to fail to converge or produce misleading results. Therefore, if the identification result is inconsistent with the initial topology, evaluation will be impossible. In this embodiment, before proceeding to error evaluation, the current topology status of the collection point is first confirmed through metering topology analysis to ensure the reliability of the data and whether the power flow direction has changed, thus ensuring the validity of the data in subsequent steps.

[0105] In some embodiments of the present invention, such as Figure 4 As shown, the synchronous monitoring data includes three-phase voltage amplitude, three-phase voltage phase angle, three-phase current amplitude, three-phase current phase angle, and power factor. A metering topology analysis was performed on the synchronous monitoring data to obtain topology verification results, including:

[0106] S401. Based on the three-phase voltage amplitude, three-phase voltage phase angle, three-phase current amplitude, and three-phase current phase angle in the synchronous monitoring data, calculate the instantaneous active power and average phase difference of the main meter and each sub-meter.

[0107] S402. For any master table or sub-table, determine the power flow direction based on the instantaneous active power, and match the power flow direction with the preset initial direction to obtain the first matching result;

[0108] S403. For any master meter or sub-meter, match the average phase difference and power factor to obtain the second matching result;

[0109] S404. Based on the first matching result and the second matching result, determine the topology verification result.

[0110] This step aims to ensure that the data used for subsequent error assessments has the correct physical logic, and to perform online verification of the real-time connection relationship and energy flow of the metering system by synchronously collecting multiple electrical quantities.

[0111] Specifically, during implementation, it is necessary to obtain the three-phase voltage amplitude and phase angle, as well as the three-phase current amplitude and phase angle of each energy meter at the same timestamp.

[0112] First, the instantaneous power of each phase is calculated based on the three-phase voltage amplitude, three-phase voltage phase angle, three-phase current amplitude, and three-phase current phase angle, and then summed to obtain the instantaneous active power of the energy meter.

[0113] The power flow direction is then determined based on the instantaneous active power. If the instantaneous active power is greater than zero, the power flow direction is determined to be forward power consumption; if the instantaneous active power is less than zero, the power flow direction is determined to be reverse power supply. The power flow direction is then matched with a preset initial direction to obtain the first matching result.

[0114] Specifically, when the current direction is consistent with the preset initial direction, the corresponding first matching result is a successful match; otherwise, the first matching result is a failed match.

[0115] To further identify hidden wiring errors (such as incorrect phase sequence causing the power factor value to still be within a reasonable range), this embodiment introduces a power factor consistency check.

[0116] Specifically, the average phase difference between voltage and current is calculated based on the three-phase voltage amplitude, three-phase voltage phase angle, three-phase current amplitude, and three-phase current phase angle. φ Thus, the theoretical power factor is obtained. cos(φ) At the same time, the actual power factor reported by the electricity meter is directly read from the monitoring data. PF By judging | PF - cos(φ) The second matching result is obtained by checking whether the measured values ​​exceed a preset tolerance threshold. The second matching result is considered a successful match only if the power flow direction is consistent with the initial record and the power factor passes the consistency check; otherwise, the second matching result is considered a failed match.

[0117] If both the first and second matching results are successful, the process is allowed to continue; otherwise, a topology anomaly message is output and the evaluation is paused.

[0118] Through the comprehensive verification of the above-mentioned multi-source data, this embodiment can proactively identify abnormal power flow direction and hidden wiring errors caused by construction errors, equipment replacement or failure at the front end of the evaluation process. It effectively filters out data sources with logical contradictions and avoids evaluation failure or misjudgment caused by inputting erroneous data into subsequent models, thereby improving the reliability and accuracy of the overall error evaluation results.

[0119] In some embodiments of the present invention, such as Figure 5 As shown, the current energy values ​​of the main table and the slave table are obtained, and combined with the historical energy sequences of the main and slave tables, the trend consistency status of the main and slave tables is determined, including:

[0120] S501, Collect the T data adjacent to the current time period. Y The historical energy value sequence of the main meter and the historical energy value sequence of the secondary meter within a time period;

[0121] S502. Based on the historical energy value sequence of the main meter, the historical energy value sequence of the secondary meter, the current energy value of the main meter and the current energy value of the secondary meter, determine the sliding standard deviation of the energy difference between the main meter and the secondary meter, and use the sliding standard deviation as the dynamic consistency threshold.

[0122] S503. Compare the absolute value of the difference between the current power value of the main table and the current power value of the secondary table with the dynamic consistency threshold to obtain the trend consistency status of the main table and the secondary table.

[0123] It should be noted that in this embodiment... T Y The time period is the nearest T-period before the current time period. Y There are several time periods, and the length of each time period can be set according to the actual application needs. No specific limit is set here.

[0124] Specifically, in this embodiment, formulas (2) and (3) are used to determine the sliding standard deviation:

[0125] (2)

[0126] (3)

[0127] in, For T Y The average of the energy difference between the main meter and the slave meter over a given time period. For the sliding standard deviation, t For the current time period, For the first The energy value of the main meter over a time period. For the first The electrical energy value of the secondary meter during each time period.

[0128] It should be understood that the moving standard deviation reflects the degree of recent typical fluctuation and dispersion of the difference in electricity consumption between the main meter and the auxiliary meter. A smaller moving standard deviation indicates a high degree of consistency between the two meters recently; a larger moving standard deviation indicates... A large discrepancy between the two tables indicates a significant recent difference. Therefore, this embodiment uses the moving standard deviation as a dynamic consistency threshold to assess the consistency of future trends.

[0129] Furthermore, after determining the dynamic consistency threshold, the absolute value of the difference between the current energy value of the main table and the current energy value of the secondary table is compared with the dynamic consistency threshold to obtain the trend consistency status of the main table and the secondary table. This trend consistency status includes consistency and inconsistency.

[0130] In some embodiments of the present invention, such as Figure 6 As shown, based on the trend consistency status, the current energy values ​​of the main table and the secondary table are merged to obtain reference energy data, including:

[0131] S601. When the trend consistency status is consistent, the current energy value of the main meter and the current energy value of the secondary meter are merged by taking the arithmetic average to obtain reference energy data.

[0132] S602. When the consistency status is inconsistent, based on the energy values ​​of the main table and the secondary table within the time period, calculate the time smoothness index and load change trend conformity index of the main table and the secondary table respectively. The time smoothness index represents the smoothness of the jump during the time period, and the load change trend conformity index represents the degree of consistency between the load change and the recent trend.

[0133] S603. Use time smoothness index and load change trend conformity index to generate fusion weights;

[0134] S604. Based on the fusion weight, the current energy value of the main table and the current energy value of the secondary table are weighted and fused to obtain reference energy data.

[0135] This embodiment aims to select and execute an appropriate fusion method based on the consistency status of the main table and the secondary table data to generate high-quality master table reference energy data. If the consistency status is consistent, it indicates that the recent fluctuations of both are small and the current difference is within the historical normal fluctuation range. In this case, the arithmetic mean of the values ​​of the two tables can be directly used as the fusion result, i.e. E 0 (t) =[ E 主 (t) + E 副 (t) ] / 2.

[0136] If the consistency status is "consistent," it indicates that the current difference significantly exceeds historical norms, and forcibly using the average value would introduce errors. In this case, this embodiment employs a weighted fusion method based on dynamic reliability assessment.

[0137] Specifically, two evaluation indicators need to be calculated for the main table and the sub-table respectively: one is the time smoothness indicator ( S 1) This indicator quantifies the degree of drastic change in the current electricity consumption relative to its recent average change. It is calculated using an exponential decay function; if the ratio of the current increment to the recent average increment exceeds a preset threshold... θ 1. If the time smoothness index value decreases, it indicates that there may be anomalies in the data at that moment. Secondly, the load change trend conformity index ( S 2) This is used to determine whether the current direction of the meter's electricity consumption change (increase or decrease) is consistent with its recent trend (e.g., the previous 7 cycles). If the direction is consistent, a high score (e.g., 1.0) is assigned; otherwise, a low score (e.g., 0.1) is assigned. Subsequently, the smoothness index of the same meter is... S 1. Trend Conformity Index S Adding the two together yields the overall reliability score for the table. S(t) Finally, by normalizing the credibility scores of the main and sub-tables, their fusion weights were obtained. w 主 (t) and w 副 (t) And perform a weighted summation accordingly: E 0 (t) = w 主 (t) E 主 (t) + w 副 (t) E 副 (t) .

[0138] Preferably, in this embodiment, formula (4) is used to calculate the time smoothness of the main table and the secondary table respectively:

[0139] (4)

[0140] in, This is the decay rate parameter; The acceptable range of variation parameter; M To select MThe past time period. The numerator represents the current change, and the denominator represents the average change over the recent period. t The less smooth the timeline, the better the table. S The smaller 1 is, the smoother the transition at time t. S The larger 1 is, the better. In one specific alternative implementation, Set it to 1.5. Set to 2, M Set it to 7.

[0141] Preferably, in this embodiment, formulas (5), (6), and (7) are used to calculate the trend conformity index for the main table and the sub-table respectively:

[0142] (5)

[0143] (6)

[0144] (7)

[0145] in, Sign To determine the sign, positive numbers are output as 1, and negative numbers are output as -1. An indicator used to determine the direction of instantaneous change in the electricity consumption of a single energy meter in the current cycle. This represents the recent trend for a single electricity meter.

[0146] This embodiment introduces a dynamic weight allocation mechanism based on data smoothness and trend consistency. When there is a significant divergence between the main and secondary table data, the weights can be dynamically adjusted. This effectively reduces the interference caused by instantaneous data anomalies or jumps in a single table on the reference benchmark. Compared with simple fixed strategies or the mean method, it can generate more robust and reliable total table reference energy data, which is conducive to improving the accuracy of subsequent error assessment.

[0147] In some embodiments of the present invention, such as Figure 7 As shown, after comparing the predicted energy value with the displayed energy value for each sub-meter to obtain the sub-meter error assessment result, the energy meter error status assessment method further includes:

[0148] S701. The method of comparing the error assessment result with the preset error threshold is used to determine whether there is an abnormal energy meter;

[0149] S702. When an abnormal energy meter is detected, an alarm message is generated to issue a status alarm. The alarm message includes the identifier of the abnormal energy meter, the error assessment result, and a timestamp.

[0150] The preset error threshold can be set according to the actual application needs, and there are no restrictions here.

[0151] In this embodiment, by comparing the error assessment result with a preset error threshold, it is quickly determined whether there is an abnormal energy meter. When it is determined that there is an abnormal energy meter, an alarm message is generated to issue a status alarm, which is conducive to timely elimination of abnormalities and improves the efficiency and timeliness of fault handling.

[0152] To better implement the electricity meter error state assessment method in this embodiment of the invention, based on the electricity meter error state assessment method, correspondingly, as follows: Figure 8 As shown, this embodiment of the invention also provides an energy meter error status assessment device, the energy meter error status assessment device 800 comprising:

[0153] The status determination module 801 is used to obtain the current energy value of the main table and the current energy value of the secondary table, and combine the historical energy sequences of the main table and the secondary table to determine the trend consistency status of the main table and the secondary table;

[0154] The data fusion module 802 is used to fuse the current energy value of the main table and the current energy value of the slave table according to the trend consistency status to obtain reference energy data;

[0155] The power prediction module 803 is used to obtain the power display value of each sub-meter, and to use a trained time series analysis model to analyze and process the reference power data of the continuous time series and the power display value of each sub-meter to obtain the power prediction value corresponding to each sub-meter.

[0156] The result determination module 804 is used to compare the predicted power consumption value with the displayed power consumption value for each sub-meter to obtain the error assessment result of the sub-meter.

[0157] The electricity meter error status assessment device 800 provided in the above embodiments can realize the technical solutions described in the above electricity meter error status assessment method embodiments. The specific implementation principles of each module or unit can be found in the corresponding content in the above electricity meter error status assessment method embodiments, which will not be repeated here.

[0158] like Figure 9 As shown, the present invention also provides an electronic device 900. The electronic device 900 includes a processor 901, a memory 902, and a display 903. Figure 9 Only some components of the electronic device 900 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0159] In some embodiments, processor 901 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 902 or process data, such as the energy meter error status assessment method of the present invention.

[0160] In some embodiments, processor 901 may be a single server or a group of servers. The server group may be centralized or distributed. In some embodiments, processor 901 may be local or remote. In some embodiments, processor 901 may be implemented on a cloud platform. In one embodiment, the cloud platform may include a private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, intranet, multi-cloud, etc., or any combination thereof.

[0161] In some embodiments, memory 902 may be an internal storage unit of electronic device 900, such as a hard disk or memory of electronic device 900. In other embodiments, memory 902 may also be an external storage device of electronic device 900, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on electronic device 900.

[0162] Furthermore, the memory 902 may include both internal storage units of the electronic device 900 and external storage devices. The memory 902 is used to store application software and various types of data installed on the electronic device 900.

[0163] In some embodiments, display 903 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 903 is used to display information from electronic device 900 and to display a visual user interface. Components 901-903 of electronic device 900 communicate with each other via a system bus.

[0164] In one embodiment, when the processor 901 executes the energy meter error status assessment program in the memory 902, the following steps can be implemented:

[0165] Obtain the current energy value of the main table and the current energy value of the secondary table, and combine them with the historical energy sequences of the main table and the secondary table to determine the trend consistency status of the main table and the secondary table;

[0166] Based on the trend consistency status, the current energy value of the main table and the current energy value of the secondary table are merged to obtain reference energy data;

[0167] The power consumption display value of each sub-meter is obtained, and the reference power consumption data and the power consumption display value of each sub-meter are analyzed and processed using a trained time series analysis model to obtain the power consumption prediction value corresponding to each sub-meter.

[0168] For each sub-meter, the predicted power consumption value is compared with the displayed power consumption value to obtain the error assessment result of the sub-meter.

[0169] It should be understood that when the processor 901 executes the energy meter error status assessment program in the memory 902, in addition to the functions mentioned above, it can also perform other functions, as can be found in the description of the corresponding method embodiments above.

[0170] Furthermore, the embodiments of the present invention do not specifically limit the type of electronic device 900 mentioned. Electronic device 900 can be a mobile phone, tablet computer, personal digital assistant (PDA), wearable device, laptop computer, or other portable electronic device. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable electronic device can also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, electronic device 900 may not be a portable electronic device, but rather a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0171] Accordingly, this application also provides a computer-readable storage medium for storing computer-readable programs or instructions. When the programs or instructions are executed by a processor, they can implement the steps or functions in the energy meter error status assessment methods provided in the above-described method embodiments.

[0172] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.), and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0173] The above provides a detailed description of the method, apparatus, electronic device, and storage medium for evaluating the error status of electricity meters provided by this invention. Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for assessing the error status of an electricity meter, characterized in that, In a power metering scenario with one master meter and X sub-meters, the power flow of the master meter and sub-meters is opposite. The data of the master meter and sub-meters are timestamped and synchronized in time. At any fixed moment, the sum of the power consumption recorded by the sub-meters should approximately equal the power consumption recorded by the master meter. The master meter uses a master table and sub-tables to form a master-slave structure. The method includes: Obtain the current energy value of the main table and the current energy value of the secondary table, and combine them with the historical energy sequences of the main table and the secondary table to determine the trend consistency status of the main table and the secondary table; Based on the trend consistency status, the current energy value of the main table and the current energy value of the secondary table are merged to obtain reference energy data; The power consumption display value of each sub-meter is obtained, and the reference power consumption data and the power consumption display value of each sub-meter are analyzed and processed by the trained time series analysis model to obtain the power consumption prediction value corresponding to each sub-meter. The trained time series analysis model uses a composite loss function to introduce the physical constraint of energy conservation so that the sum of the sub-meter power consumption data is equal to the total meter power consumption data. For each sub-meter, the predicted power consumption value is compared with the displayed power consumption value to obtain the error assessment result of the sub-meter.

2. The method for assessing the error status of an electricity meter according to claim 1, characterized in that, Before obtaining the current energy value of the main meter and the current energy value of the secondary meter, the energy meter error status assessment method includes: Acquire synchronous monitoring data of the master table and each sub-table within the target area, wherein the master table includes a main table and sub-tables; A metering topology analysis was performed on the synchronous monitoring data to obtain topology verification results; When the topology verification result is successful, the step of obtaining the current power value of the main table and the current power value of the secondary table is executed.

3. The method for evaluating the error status of an electricity meter according to claim 2, characterized in that, The synchronous monitoring data includes three-phase voltage amplitude, three-phase voltage phase angle, three-phase current amplitude, three-phase current phase angle, and power factor. The metering topology analysis performed on the synchronous monitoring data to obtain topology verification results includes: Based on the three-phase voltage amplitude, three-phase voltage phase angle, three-phase current amplitude, and three-phase current phase angle in the synchronous monitoring data, calculate the instantaneous active power and average phase difference of the main meter and each sub-meter; For any master table or sub-table, the power flow direction is determined based on the instantaneous active power, and the power flow direction is matched with a preset initial direction to obtain a first matching result; For any master table or sub-table, the average phase difference and power factor are matched to obtain a second matching result; Based on the first matching result and the second matching result, the topology verification result is determined.

4. The method for assessing the error status of an electricity meter according to claim 1, characterized in that, The process of obtaining the current energy values ​​of the main table and the secondary table, and combining them with the historical energy sequences of the main and secondary tables to determine the trend consistency status of the main and secondary tables, includes: Collect data adjacent to the current time period. T Y The historical energy value sequence of the main meter and the historical energy value sequence of the secondary meter within a time period; Based on the historical energy value sequence of the main meter, the historical energy value sequence of the secondary meter, the current energy value of the main meter and the current energy value of the secondary meter, the sliding standard deviation of the energy difference between the main meter and the secondary meter is determined, and the sliding standard deviation is used as the dynamic consistency threshold. The absolute value of the difference between the current power value of the main table and the current power value of the secondary table is compared with the dynamic consistency threshold to obtain the trend consistency status of the main table and the secondary table.

5. The method for evaluating the error status of an electricity meter according to claim 1, characterized in that, The step of fusing the current energy value of the main table and the current energy value of the secondary table according to the trend consistency status to obtain reference energy data includes: When the trend consistency status is consistent, the current energy value of the main table and the current energy value of the secondary table are merged by taking the arithmetic average to obtain the reference energy data; When the trend consistency status is inconsistent, based on the energy values ​​of the main table and the sub-table within the time period, the time smoothness index and load change trend conformity index of the main table and the sub-table are calculated respectively. The time smoothness index represents the smoothness of the jump during the time period, and the load change trend conformity index represents the degree of consistency between the load change and the recent trend. The time smoothness index and the load change trend conformity index are used to generate the fusion weights; Based on the fusion weight, the current energy value of the main table and the current energy value of the secondary table are weighted and fused to obtain the reference energy data.

6. The method for evaluating the error status of an electricity meter according to claim 1, characterized in that, The time series analysis model is a neural network model containing a Transformer encoder. The Transformer encoder includes a self-attention mechanism. The composite loss function used in the time series analysis model introduces the physical constraint of energy conservation so that the sum of the sub-meter energy data equals the total meter energy data.

7. The method for assessing the error status of an electricity meter according to any one of claims 1 to 6, characterized in that, After comparing the predicted energy value with the displayed energy value for each sub-meter to obtain the error assessment result for that sub-meter, the energy meter error status assessment method includes: The method of comparing the error assessment result with a preset error threshold is used to determine whether there is an abnormal energy meter; When an abnormal energy meter is detected, an alarm message is generated to issue a status alarm. The alarm message includes the identifier of the abnormal energy meter, the error assessment result, and a timestamp.

8. A device for assessing the error status of an electricity meter, characterized in that, In a power metering scenario with one master meter and X sub-meters, the power flow of the master meter and sub-meters is opposite. The data from the master meter and sub-meters are timestamped and synchronized in time. At any fixed moment, the sum of the power consumption recorded by the sub-meters should approximately equal the power consumption recorded by the master meter. The master meter uses a master meter and sub-meters to form a master-slave structure. The device includes: The status determination module is used to obtain the current energy value of the main table and the current energy value of the secondary table, and combine them with the historical energy sequence of the main table and the secondary table to determine the trend consistency status of the main table and the secondary table; The data fusion module is used to fuse the current energy value of the main table and the current energy value of the secondary table according to the trend consistency status to obtain reference energy data; The power prediction module is used to obtain the power display value of each sub-meter, and to use a trained time series analysis model to analyze and process the reference power data and the power display value of each sub-meter in the continuous time series to obtain the power prediction value corresponding to each sub-meter. The trained time series analysis model uses a composite loss function to introduce the physical constraint of energy conservation so that the sum of the sub-meter power data is equal to the total meter power data. The result determination module is used to compare the predicted power consumption value with the displayed power consumption value for each sub-meter to obtain the error evaluation result of the sub-meter.

9. An electronic device, characterized in that, Including memory and processor, among which, The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the energy meter error status assessment method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the energy meter error state assessment method according to any one of claims 1 to 7.

Images