Online Deviation Calibration Method and System for Electricity Meters

Abstract

This invention discloses an online deviation calibration method and system for electricity meters, relating to the field of electricity detection technology, including the following steps: S1: Constructing a hierarchical topology measurement relationship for the distribution area and collecting metering data from the main meter and user electricity meters in the distribution area, while simultaneously collecting hierarchical node data from the meter box metering module and the branch box metering module, triggering control for the collection of frozen electricity, and completing meter synchronization before collecting frozen electricity, recording the clock deviation during the synchronization process and forming a unified time reference; S2: Performing time series information compression and anomaly detection based on the multi-dimensional dynamic and static data distribution changes of the distribution area and electricity meters. Based on the energy conservation of hierarchical topology measurement and time-frozen electricity, this invention introduces a decodeable calibration watermark beacon excitation and, through related decoding inversion path loss constraints, combines non-convex sparse multi-objective solution with uncertainty threshold constraints to achieve quantitative calibration output of online deviation of electricity meters.

Description

Technical Field

[0001] This invention relates to the field of electrical energy detection technology, specifically to an online deviation calibration method and system for electricity meters. Background Technology

[0002] Electricity metering devices play a fundamental role in the electricity metering system, undertaking the basic functions of measuring electrical energy, storing metering data, and applying metering results. Based on the management requirements of monitoring the operational status, predicting the lifespan, and evaluating the condition of smart meters, the condition evaluation and replacement of electricity meters are gradually forming a management and application model based on the distribution area. In a distribution area power supply scenario, one distribution area corresponds to the power supply area of ​​one transformer, and the distribution area's main meter corresponds to multiple user electricity meters. The electricity relationship within the distribution area can be established based on the law of conservation of energy, establishing a correspondence between the power supply of the main meter and the actual electricity consumption of each sub-meter, the fixed losses of the distribution area, and line losses. This provides an interpretable physical mechanism basis for electricity meter condition evaluation and online calibration.

[0003] In a transformer substation power supply scenario, one substation corresponds to the power supply area of ​​one transformer, and the main meter of the substation corresponds to multiple user electricity meters. The electricity relationship within the substation can be established based on the law of conservation of energy, establishing a correspondence between the power supply of the main meter of the substation and the actual electricity consumption of each sub-meter, the fixed losses of the substation, and the line losses. This provides an interpretable physical mechanism basis for electricity meter status evaluation and online calibration. To support online evaluation and calibration applications, the system can adopt a data interaction link from the main data acquisition station to the substation concentrator, the meter box metering module, and the user electricity meter to output the calibration results, quality status evaluation results, quality status prediction results, and operation and maintenance handling measures for each electricity meter.

[0004] In practical applications, due to the limited power load of users, the variation range is insufficient when collecting power consumption data at a low frequency, making it difficult to construct and stably solve the model equations. If the assumption of constant line loss rate is adopted in the model solution within the iteration window, it is inconsistent with the actual operating characteristics, and the calculation of line loss rate is difficult in engineering. Therefore, it is necessary to construct a model structure that reduces the impact of line loss rate on the calculation of operating error. Summary of the Invention

[0005] The purpose of this invention is to provide an online deviation calibration method and system for electricity meters to solve the problems mentioned in the background art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: an online deviation calibration method for electricity meters, comprising the following steps:

[0007] S1: Construct the hierarchical topology measurement relationship of the distribution area and collect the metering data of the main meter and user electricity meters in the distribution area. At the same time, collect the hierarchical node data of the meter box metering module and the branch box metering module. Trigger control is performed on the frozen power collection operation and the meter time synchronization is completed before the frozen power collection. The clock deviation of the time synchronization process is recorded and a unified time reference is formed. The unified time reference is used to align the frozen power collection time with each sampling time.

[0008] S2: Based on the distribution changes of multi-dimensional dynamic and static data of transformer substations and electricity meters, time series information compression and anomaly detection are performed. Time series information compression refers to the aggregation and resampling of the collected sequence according to a preset time window to reduce data redundancy. The identified abnormal disturbance factors are processed and a high-quality balanced dataset is output to form a usable sample set for subsequent modeling and solving.

[0009] S3: In the selected hierarchical nodes of the layered topology of the substation, the controlled energy injection scheduling of the calibration watermark beacon is executed, the corresponding data on the total table side of the substation is collected synchronously and the relevant decoding is performed, the path loss constraint is inverted based on the decoding result and associated with the layered topology, and the calibration watermark beacon is the correlated decoded power disturbance contribution formed by the controlled calibration watermark injection on the total table side of the substation.

[0010] S4: Based on the physical mechanism model of energy conservation in the transformer area, a deviation solution model is established, and the path loss constraint is introduced into the solution process to weaken the influence of time-varying line loss on deviation calculation. Non-convex sparse multi-objective solution is performed to obtain the online deviation estimate of each user's electricity meter. The uncertainty of the solution result is evaluated and the standard uncertainty component is redistributed to meet the threshold constraint. Non-convex sparse multi-objective solution refers to the optimization solution that applies sparse constraints to the deviation vector under the objective function of simultaneously minimizing the line loss term and the deviation term.

[0011] S5: Output the calibration results, quality status evaluation results, and quality status prediction results of each electricity meter, and generate corresponding operation and maintenance measures. For cases where the uncertainty threshold constraint is not met, perform closed-loop adjustment of the watermark injection strategy and data interaction strategy. Verify the output results through the scenario simulation and digital twin mixed reality evaluation and verification platform and form a traceable record. Perform scenario simulation verification of the interaction between the digital twin model and the measured data.

[0012] According to the above technical solution, step S1 includes:

[0013] S1-1: Establish a hierarchical mapping relationship between the main meter of the distribution area and the user's electricity meter, and establish a physical mechanism model of energy conservation in the distribution area under the unified time base, satisfying the following equation: ,in The total power supply of the transformer substations is the amount supplied to the transformer substations under the unified time base. This refers to the number of user electricity meters corresponding to the total meter for the aforementioned distribution area. For the first The electricity consumption characteristic of each user's electricity meter under the unified time base. For the first Online deviation of individual user's electricity meter For fixed losses in the transformer area, For the line loss item, this step is used to establish a consistent physical constraint between the total power supply of the transformer area meter and the power consumption of each user's electricity meter, the fixed loss item of the transformer area and the line loss item, so as to quantitatively solve the online deviation in the subsequent process.

[0014] S1-2: Before triggering the freeze on electricity collection, perform meter time synchronization and record the clock deviation to align the collected data from the main meter of the distribution area, the meter box metering module, the branch box metering module, and the user's electricity meter to the unified time reference, so that the energy conservation physical mechanism model of the input distribution area is... With each To ensure that the same time base condition is met, so as to avoid time inconsistencies. Amplified by unbiased factors.

[0015] According to the above technical solution, S2 includes:

[0016] S2-1: Based on the unified time base dataset formed in step S1, obtain the power supply sequence of the transformer substation master meter, the power consumption sequence of each user's energy meter, the hierarchical mapping relationship between the transformer substation master meter and user energy meters, the frozen power collection trigger time, and the number of clock deviation records. Index and align the above data according to the unified time base to form a sample set to be processed. ;

[0017] S2-2: For the sample set The power supply sequence and power consumption sequence of each sampling time are subjected to integrity verification to obtain a set of missing markers. For samples with missing markers, missing value imputation is performed on the same time base. The missing value imputation includes: generating surrogate values ​​for short-term missing values ​​using forward imputation rules, and generating surrogate values ​​for consecutive missing values ​​using local interpolation rules, thereby obtaining a complete sample set. ;

[0018] S2-3: Calculate the time consistency index for each sampling moment based on the clock deviation record. If the time consistency index exceeds a preset time consistency threshold, mark the corresponding sampling moment as a time-inconsistent sample and perform time window correction on the sampling window to which the sample belongs. The time window correction is as follows: shift the boundary of the sampling window corresponding to the frozen electricity according to the clock deviation record, so that the power supply sequence of the transformer area and the electricity consumption sequence within the sampling window meet the same frozen electricity triggering benchmark, thereby obtaining a time-consistent sample set. ;

[0019] S2-4: Construct the energy conservation residual sequence for each sampling time based on the physical mechanism model of energy conservation in the transformer area. The energy conservation residual sequence is used to characterize the deviation between the power supply sequence and the power consumption sequence of the transformer area under physical constraints. A mutation detection is performed on the energy conservation residual sequence to obtain a mutation sample set. A persistent offset detection is performed on the energy-conserving residual sequence to obtain an offset sample set. , will belong to and The samples are labeled as anomalous perturbation candidate samples;

[0020] S2-5: Perform a topology consistency check on the hierarchical mapping relationship at each sampling time. The topology consistency check includes: checking the number of user energy meters corresponding to the total meter of the distribution area. The system checks whether changes at adjacent sampling times satisfy preset topology stability constraints and verifies whether the user's electricity meter's affiliation in the hierarchical mapping relationship deviates from the topology configuration record maintained by the data acquisition master station. Samples that do not satisfy the topology stability constraints are marked as a set of topology-inconsistent samples. ;

[0021] S2-6: Process the abnormal perturbation candidate samples and topologically inconsistent samples separately to obtain a high-quality balanced dataset. The treatment includes: processing the set of mutant samples. Perform a culling process on the continuously offset sample set. Perform segmentation processing and retain stable segments within the segment boundaries for topologically inconsistent sample sets. Perform weight reduction processing and decrease its constraint weights in subsequent solutions to improve the quality of the balanced dataset. This serves as the basis for the input data of the total surface observation power of the transformer area in step S3, and as the basis for the input data of the energy conservation modeling and online deviation solution in step S4.

[0022] According to the above technical solution, step S3 includes:

[0023] S3-1: Perform controlled calibration watermark injection on selected hierarchical nodes in the tiered topology of the transformer substation. The selected hierarchical nodes are those that satisfy topology consistency verification and whose corresponding branches have controllable load modulation capabilities. Controllable load modulation capability refers to the ability to change the active power of the branch according to a preset watermark sequence within the watermark injection window. The controlled calibration watermark injection is as follows: under a unified time reference, within the preset watermark injection window, apply controlled power modulation to the power consumption branches corresponding to the hierarchical nodes, so that a calibration watermark contribution with decodeable characteristics is superimposed on the overall meter side observation measurement of the transformer substation. The watermark injection window is derived from the high-quality balanced dataset output in step S2. Select from the corresponding stable segments, and avoid the mutation samples, continuous offset samples and topologically inconsistent samples marked in step S2. The controlled power modulation is achieved by executing a power setting command on the controllable load of the corresponding branch of the hierarchical node. The power setting command causes the branch power to undergo a power perturbation of a predetermined amplitude according to the calibration watermark sequence within the watermark injection window, thereby ensuring that the observations within the watermark injection window meet the conditions of temporal consistency and topological consistency, so that the subsequent related decoding can stably extract the calibration watermark contribution.

[0024] S3-3: Within the watermark injection window, observe the total surface power of the station area. With the calibration watermark sequence Correlation decoding is performed to obtain an estimate of the path transmission coefficients, wherein the correlation decoding satisfies... ,in The path transmission coefficient The estimated value, The number of sampling points within the watermark injection window is used to extract the calibration watermark contribution and form the path transmission coefficient estimate under the condition of background load power.

[0025] S3-4: Based on Calculate the estimated watermark contribution of the watermark injection power on the total surface of the distribution area. The watermark contribution estimate was used to estimate the total surface-side observed power of the station area. Watermark contribution deduction is performed to obtain a background load power estimate, which is then used in subsequent step S4 to reduce the line loss term. The difference term, which is limited to the background load power estimate, is used to constrain the time-varying effect of the line loss rate by the path transmission coefficient estimate, thereby reducing the coupling interference of the line loss rate on the online deviation solution result.

[0026] According to the above technical solution, S4 includes:

[0027] S4-1. Using the physical mechanism model of energy conservation in the transformer area as the physical constraint and the estimated background load power as the basis for line loss constraint, a coupled solution model of line loss term and online deviation is constructed so that a solvable set of equations can still be formed under the condition that the user's power load variation range is limited and the acquisition frequency is low.

[0028] S4-2. In the coupled solution process, the online deviation of each user's electricity meter is used as the basis. As the solution object, a sparsity constraint is introduced to ensure that only a finite number of user electricity meter deviations appear as significant deviations in the solution results. This is to match the operational characteristics of a relatively limited number of misaligned meters in a batch misalignment replacement scenario. Simultaneously, measurement uncertainty is assessed and controlled in the solution results to minimize the relative expanded uncertainty. Meets preset threshold parameters The combined standard uncertainty With relative expanded uncertainty satisfy: , ,in For the first The standard uncertainty of the uncertainty component For the number of uncertainty components, As a coverage factor, the uncertainty components include at least the uncertainty components introduced by line loss estimation, clock skew, transformer area meter error, and reading error, and are satisfied by redistributing the standard uncertainty components. This ensures that the reliability of the online deviation solution results can be quantitatively controlled under the threshold parameter constraint, whereby... The parameters are determined based on the allowable error limit parameters corresponding to the metering level of the electricity meter and the preset proportional coefficient parameters.

[0029] According to the above technical solution, S5 includes:

[0030] S5-1, Output the online deviation for each user's electricity meter. and with corresponding It outputs the energy meter quality status evaluation results and quality status prediction results, and generates operation and maintenance measures corresponding to the quality status evaluation results, so that the operation and maintenance measures and the quantitative results of online deviation are established in a one-to-one correspondence.

[0031] S5-2, when Not satisfied At that time, a closed-loop adjustment is performed to control the duration of the watermark injection window. and watermark injection power amplitude Make adjustments and re-execute steps S3 and S4 until the output result meets the requirements. This closed-loop adjustment enables the link of enhanced observability—path constraint formation—deviation solution—uncertainty control to be repeatedly executed and converge to a calibration output that satisfies the threshold parameters.

[0032] An online deviation calibration system for electricity meters includes:

[0033] The measurement freeze module is used to complete the acquisition of multi-source metering data, topology association, and time reference unification before freezing electricity collection within the transformer area topology level.

[0034] The excitation inversion module is used to generate a readable controlled energy watermark within the transformer area and perform relevant decoding on the transformer area's overall table to invert path loss constraints, thereby improving the observability of the calibration equation set and reducing the impact of time-varying line loss rate on deviation calculation.

[0035] The solution output module is used to construct a non-convex sparse multi-objective optimization solution process based on the energy conservation physical mechanism model and path loss constraints, and output the deviation calibration results, quality status evaluation results, prediction results, and operation and maintenance measures of each energy meter under uncertainty threshold constraints.

[0036] According to the above technical solution, the measurement freezing module includes a data acquisition module, a topology association module, a frozen power triggering module, and a clock deviation recording module. The data acquisition module is used to acquire metering data from the main station, the distribution area concentrator, the meter box metering module, and the user's electricity meter. The topology association module is used to establish a hierarchical mapping relationship between the distribution area main meter and the user's electricity meter. The frozen power triggering module is used to complete the meter time synchronization before acquiring the frozen power. The clock deviation recording module is used to record the clock deviation introduced by the time synchronization.

[0037] The excitation inversion module includes a calibration watermark beacon module, a watermark scheduling module, a correlation decoding module, and a path loss inversion module. The calibration watermark beacon module is used to generate controlled watermark energy at the hierarchical nodes corresponding to the meter box metering module and the branch box metering module. The watermark scheduling module is used to schedule the watermark injection window and injection amplitude. The correlation decoding module is used to perform correlation decoding on the watermark sequence at the main meter side of the transformer area. The path loss inversion module is used to form a path loss constraint based on the decoding result.

[0038] The solution output module includes an anomaly handling module, an energy conservation modeling module, a non-convex sparse multi-objective solution module, an uncertainty assessment module, a testing and verification module, and an operation and maintenance strategy module. The anomaly handling module uses time-series information compression and anomaly detection to process anomaly factors. The energy conservation modeling module establishes a physical mechanism model of energy conservation in the transformer area. The non-convex sparse multi-objective solution module couples the solution of transformer area line loss and electricity meter operating error. The uncertainty assessment module redistributes the standard uncertainty components and satisfies uncertainty threshold constraints. The testing and verification module verifies the accuracy of the model output and uncertainty assessment based on scene simulation and digital twin mixed reality technology. The operation and maintenance strategy module outputs the calibration results, quality status evaluation results, prediction results, and operation and maintenance measures for each electricity meter.

[0039] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: The present invention is based on the physical mechanism of energy conservation in the transformer area, and constructs an interpretable online evaluation and calibration framework by precisely decoupling measurement and acquisition in a hierarchical and graded manner and by multi-level node collaborative calculation; furthermore, by injecting decodeable controlled energy disturbances into the transformer area through calibration watermark beacons and performing relevant decoding on the transformer area's overall table side, an observable constraint on path loss is formed, thereby improving the observability of the equation system from the source and reducing the impact of time-varying coupling of line loss on deviation calculation;

[0040] Simultaneously, by collecting data on time synchronization and clock deviation before freezing power, the time consistency of multi-source data is ensured; and by assessing uncertainty, synthesizing and redistributing components, the calibration reliability is constrained in the form of threshold parameters, thereby outputting single meter deviation, reliability index, quality status evaluation results, prediction results and operation and maintenance measures, realizing quantifiable, verifiable and deployable online deviation calibration for transformer areas and single meters. Attached Figure Description

[0041] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0042] Figure 1 This is a schematic diagram of the structure of an online deviation calibration system for electricity meters. Detailed Implementation

[0043] 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.

[0044] Please see Figure 1 This invention provides a technical solution: an online deviation calibration method for electricity meters, comprising the following steps:

[0045] S1: Construct the hierarchical topology measurement relationship of the distribution area and collect the metering data of the main meter and user electricity meters in the distribution area. At the same time, collect the hierarchical node data of the meter box metering module and the branch box metering module. Trigger control is performed on the frozen power collection operation and the meter time synchronization is completed before the frozen power collection. The clock deviation of the time synchronization process is recorded and a unified time reference is formed. The unified time reference is used to align the frozen power collection time with each sampling time.

[0046] S2: Based on the distribution changes of multi-dimensional dynamic and static data of transformer substations and electricity meters, time series information compression and anomaly detection are performed. Time series information compression refers to the aggregation and resampling of the collected sequence according to a preset time window to reduce data redundancy. The identified abnormal disturbance factors are processed and a high-quality balanced dataset is output to form a usable sample set for subsequent modeling and solving.

[0047] S3: Perform controlled energy injection scheduling of calibration watermark beacons at selected hierarchical nodes in the layered topology of the substation, synchronously collect corresponding data from the substation's overall table and perform related decoding, invert the path loss constraint based on the decoding results and associate it with the layered topology, and the calibration watermark beacon contributes to the reproducible decoded power disturbance formed by the controlled calibration watermark injection at the substation's overall table.

[0048] S4: Based on the physical mechanism model of energy conservation in the transformer area, a deviation solution model is established, and the path loss constraint is introduced into the solution process to weaken the influence of time-varying line loss on deviation calculation. Non-convex sparse multi-objective solution is performed to obtain the online deviation estimate of each user's electricity meter. The uncertainty of the solution result is evaluated and the standard uncertainty component is redistributed to meet the threshold constraint. Non-convex sparse multi-objective solution refers to the optimization solution that applies sparse constraints to the deviation vector under the objective function of simultaneously minimizing the line loss term and the deviation term.

[0049] S5: Output the calibration results, quality status evaluation results, and quality status prediction results of each energy meter, and generate corresponding operation and maintenance measures. For cases where the uncertainty threshold constraint is not met, perform closed-loop adjustment of the watermark injection strategy and data interaction strategy. Verify the output results through the scenario simulation and digital twin mixed reality evaluation and verification platform and form a traceable record. Perform scenario simulation verification of the interaction between the digital twin model and the measured data.

[0050] Step S1 includes:

[0051] S1-1: Establish the hierarchical mapping relationship between the main meter of the distribution area and the user's electricity meter, and establish a physical mechanism model of energy conservation in the distribution area under a unified time base, satisfying the following equation: ,in This represents the power supply of the corresponding transformer area under a unified time base, as indicated by the transformer area master meter. This represents the number of user electricity meters corresponding to the main meter for the distribution area. For the first The electricity consumption characteristics of each user's electricity meter under a unified time base. For the first Online deviation of individual user's electricity meter For fixed losses in the transformer area, For the line loss term, this step establishes a consistent physical constraint between the total power supply of the transformer substation, the electricity consumption of each user's meter, the fixed loss term of the transformer substation, and the line loss term. This allows for subsequent quantitative solutions to online deviations. The relationship between the total power supply of the transformer substation and the electricity consumption of each user's meter is anchored by a verifiable physical constraint, providing a unique mechanistic basis for subsequent deviation calculations, rather than pure statistical fitting. At the transformer substation scale, the power supply recorded by the total meter should equal the sum of the actual electricity consumption of each sub-meter, plus fixed losses and line losses. If there are metering deviations in the sub-meters, there is a systematic deviation between the apparent electricity consumption of the sub-meters and the actual electricity consumption; this deviation is addressed through this formula.

[0052] S1-2: Before triggering the freeze on electricity collection, perform meter time synchronization and record clock deviation to align the collected data from the main meter of the distribution area, the meter box metering module, the branch box metering module, and the user's electricity meter to a unified time reference, so that the energy conservation physical mechanism model of the distribution area is input. With each To ensure that the same time base condition is met, so as to avoid time inconsistencies. Amplified by non-biased factors;

[0053] S2 includes:

[0054] S2-1: Based on the unified time base dataset formed in step S1, obtain the power supply sequence of the transformer substation master meter, the power consumption sequence of each user's energy meter, the hierarchical mapping relationship between the transformer substation master meter and user energy meters, the frozen power collection trigger time, and the number of clock deviation records. Index and align the above data according to the unified time base to form a sample set to be processed. ;

[0055] S2-2: For the sample set The power supply sequence and power consumption sequence of each sampling time are subjected to integrity verification to obtain a set of missing markers. For samples with missing labels, missing value imputation is performed on the same time base. Missing value imputation includes: generating surrogate values ​​for short-term missing values ​​using forward imputation rules, and generating surrogate values ​​for consecutive missing values ​​using local interpolation rules, thereby obtaining a complete sample set. ;

[0056] S2-3: Calculate the time consistency index for each sampling moment based on the clock deviation record. If the time consistency index exceeds the preset time consistency threshold, the corresponding sampling moment is marked as a time-inconsistent sample, and time window correction is performed on the sampling window to which the sample belongs. The time window correction is as follows: the boundary of the sampling window corresponding to the frozen electricity is shifted according to the clock deviation record, so that the power supply sequence of the transformer area and the electricity consumption sequence within the sampling window meet the same frozen electricity triggering benchmark, thereby obtaining a time-consistent sample set. ;

[0057] S2-4: Construct the energy conservation residual sequence for each sampling time based on the physical mechanism model of energy conservation in the transformer area. The energy conservation residual sequence is used to characterize the deviation between the power supply sequence and the power consumption sequence of a transformer substation under physical constraints. A mutation detection is performed on the energy conservation residual sequence to obtain a mutation sample set. A persistent migration detection is performed on the energy-conserving residual sequence to obtain a set of migrated samples. , will belong to and The samples are labeled as anomalous perturbation candidate samples;

[0058] S2-5: Perform topology consistency verification on the hierarchical mapping relationship at each sampling time. Topology consistency verification includes: verifying the number of user energy meters corresponding to the total meter of the transformer area. The system checks whether changes at adjacent sampling times satisfy preset topology stability constraints and verifies whether the user's electricity meter's affiliation in the hierarchical mapping relationship deviates from the topology configuration record maintained by the data acquisition master station. Samples that do not satisfy the topology stability constraints are marked as a set of topology-inconsistent samples. ;

[0059] S2-6: Process the abnormal perturbation candidate samples and topologically inconsistent samples separately to obtain a high-quality balanced dataset. The treatment includes: processing the set of mutant samples. Perform a culling process on the continuously offset sample set. Perform segmentation processing and retain stable segments within the segment boundaries for topologically inconsistent sample sets. Perform weight reduction processing and decrease the constraint weights in subsequent solutions to achieve a high-quality balanced dataset. This serves as the basis for the input data of the total surface observation power of the transformer area in step S3, and as the basis for the input data of the energy conservation modeling and online deviation solution in step S4.

[0060] Conventional data cleaning methods often employ simple threshold removal, moving averages, or empirical rule segmentation, neglecting the decisive impact of transformer area energy relationships and topological changes on data usability. This results in mixing genuine deviation clues with anomalous disturbances, leading to instability in subsequent calibration models. This step first constructs a consistency check using energy relationships, then combines this with the stability of topological mapping to perform a double screening of samples. Abrupt disturbances, persistent offset disturbances, and topologically inconsistent samples are separately marked and processed through removal, segmentation to retain stable segments, and reduction of constraint weights. This ensures that the data entering subsequent watermark decoding and deviation calculation possesses temporal consistency, topological consistency, and controllable disturbances, providing clean and interpretable input for watermark decoding and deviation calculation, and preventing related decoding from being contaminated by abnormal loads or topological changes. Anomaly handling is elevated from statistical cleaning to a joint screening and processing of physical and topological consistency, and the processing results are explicitly used as the constraint basis for subsequent solutions.

[0061] Step S3 includes:

[0062] S3-1: Perform controlled calibration watermark injection on selected hierarchical nodes in the tiered topology of the transformer substation. Selected hierarchical nodes are those that meet topology consistency verification and whose corresponding branches have controllable load modulation capabilities. Controllable load modulation capability refers to the ability to change the active power of the branch according to a preset watermark sequence within the watermark injection window. Controlled calibration watermark injection involves applying controlled power modulation to the power consumption branches corresponding to the hierarchical nodes within the preset watermark injection window under a unified time reference, so that a readable calibration watermark contribution is superimposed on the overall meter readings of the transformer substation. The watermark injection window is derived from the high-quality balanced dataset output in step S2. Select from the corresponding stable segments, and avoid the mutation samples, continuous offset samples and topologically inconsistent samples marked in step S2. Controlled power modulation is achieved by executing a power setting command on the controllable load of the corresponding branch of the hierarchical node. The power setting command causes the branch power to undergo a power perturbation of a predetermined amplitude according to the calibration watermark sequence within the watermark injection window, thereby ensuring that the observations within the watermark injection window meet the conditions of temporal consistency and topological consistency, so that subsequent related decoding can stably extract the calibration watermark contribution.

[0063] S3-2: Within the watermark injection window, the total meter-side observed power of the station area satisfies: ,in For the total table of the station area at the sampling time The observation power, High-quality balanced dataset taken from the output of step S2 The observed power sequence of the central distribution area is obtained by dividing the distribution area power supply sequence by sampling interval. The conversion yields, For background load power that does not include the contribution of calibration watermark, This refers to the number of hierarchical nodes participating in controlled calibration watermark injection within the watermark injection window. For the first The watermark injection power amplitude of each level node For the first The calibration watermark sequence corresponding to each hierarchical node is a preset decodable sequence, used to extract the watermark contribution on the master table side through relevant decoding. For the first The path transmission coefficient from each hierarchical node to the overall table of the distribution area. To observe the noise term, this step is used to distinguish the background load power and the calibration watermark contribution in the total table side observation, thereby providing a quantitative basis for subsequent extraction of path transmission coefficients and constraint of line loss terms;

[0064] This formula employs the concept of linear superposition, meaning that background load and watermark disturbance coexist within the same observation channel. The total apparent measurement can be expressed as a background term + several watermark terms + noise term. The path transmission coefficient reflects the effective transmission / attenuation of the watermark from the injection point to the total surface, converging complex factors such as line loss and topological coupling into an estimable coefficient.

[0065] S3-3: Within the watermark injection window, observe the total surface power of the station area. With calibration watermark sequence Correlation decoding is performed to obtain estimates of the path transmission coefficients, and the correlation decoding satisfies... ,in Path transmission coefficient The estimated value, The number of sampling points within the watermark injection window is used to extract the calibration watermark contribution and form a path transmission coefficient estimate under the condition of background load power. The contribution corresponding to each watermark sequence is extracted from the total apparent measurements, and the estimated path transmission coefficient is output for subsequent constraint line loss and stability deviation calculations. This formula correlates and accumulates the observed sequence with a certain watermark sequence. The watermark portion accumulates in the same direction, while the background load and noise are statistically uncorrelated with the watermark sequence and will cancel each other out or not accumulate, thus achieving separation.

[0066] Conventional approaches typically rely passively on natural load fluctuations, using long-term statistical regression or fixed line loss assumptions to guess line impacts and deduce meter deviations. In low-load, low-load distribution areas, these methods often result in unobservable equations, drifting results, and insufficient reliability. This step selects a branch within the distribution area with controllable modulation conditions and injects a small, decodeable power disturbance within a short time window. This disturbance manifests as a separable watermark contribution at the main meter level, and through relevant decoding, this contribution is extracted from the background load, thus obtaining observable constraints related to the power supply path. Actively enhancing observability and transforming the line loss impact from uncontrollable interference into a constrained quantity provides a stable and repeatable input basis for subsequent deviation calculations. Its originality lies in transforming the metering calibration problem into an active experimental design and decodeable signal extraction problem, no longer relying on natural load for excitation.

[0067] S3-4: Based on Calculate the estimated watermark contribution of the watermark injection power on the total surface of the distribution area. The watermark contribution estimate was used to estimate the total surface-side observed power of the station area. Watermark contribution is subtracted to obtain an estimate of background load power, which is then used in subsequent step S4 to reduce line loss. By limiting the difference term to be consistent with the background load power estimate, the time-varying effect of the line loss rate is constrained by the path transmission coefficient estimate, thereby reducing the coupling interference of the line loss rate on the online deviation solution result.

[0068] S4 includes:

[0069] S4-1. Using the physical mechanism model of energy conservation in the transformer area as the physical constraint and the estimated background load power as the basis for line loss constraint, a coupled solution model of line loss term and online deviation is constructed so that a solvable set of equations can still be formed under the condition that the user's power load variation range is limited and the acquisition frequency is low.

[0070] S4-2. In the coupled solution process, the online deviation of each user's electricity meter is used as the basis. As the solution object, a sparsity constraint is introduced to ensure that only a finite number of user electricity meter deviations appear as significant deviations in the solution results. This is to match the operational characteristics of a relatively limited number of misaligned meters in a batch misalignment replacement scenario. Simultaneously, measurement uncertainty is assessed and controlled in the solution results to minimize the relative expanded uncertainty. Meets preset threshold parameters The combined standard uncertainty With relative expanded uncertainty satisfy: , ,in For the first The standard uncertainty of the uncertainty component For the number of uncertainty components, As a coverage factor, the uncertainty components must include at least the uncertainty components introduced by line loss estimation, clock skew, transformer area meter error, and reading error, and must be redistributed to meet the standard uncertainty components. This ensures that the reliability of the online deviation solution results can be quantitatively controlled under the threshold parameter constraint, whereby... The calibration results are determined based on the permissible error limit parameters corresponding to the metering level of the electricity meter and the preset proportional coefficient parameters. The reliability of the calibration results is quantified in the form of expanded uncertainty and used as a threshold constraint for closed-loop control. Multiple independent uncertainty components are synthesized using the root mean square method to obtain the combined standard uncertainty, which is then multiplied by the coverage factor to obtain the expanded uncertainty, used to express the confidence interval of the result at a given confidence level.

[0071] Conventional methods often directly minimize errors or perform linear fitting, estimating line loss and deviation together in a single model. When the load variation in a transformer area is insufficient or line loss fluctuates constantly, conventional solutions are prone to multiple solutions, drift, or overfitting, ultimately providing only qualitative judgments and failing to generate reliable, verifiable, and implementable outputs. This step explicitly introduces path constraints obtained from previous steps when calculating deviations, weakening the coupling of line influences to deviations. Simultaneously, leveraging the operational characteristic of a few abnormal meter replacements, it guides deviation calculations to a result format where only a few meters are significantly inaccurate, and controls the reliability of the results as a hard constraint. When the reliability fails to meet the standard, it automatically returns to the previous steps to adjust the watermark injection strategy and re-solves. This upgrades the ability to calculate deviations to one that is reliably and repeatably calculated, thereby supporting the output of operation and maintenance strategies.

[0072] S5 includes:

[0073] S5-1, Output the online deviation for each user's electricity meter. and with corresponding It outputs the energy meter quality status evaluation results and quality status prediction results, and generates operation and maintenance measures corresponding to the quality status evaluation results, so that the operation and maintenance measures and the quantitative results of online deviation are established in a one-to-one correspondence.

[0074] S5-2, when Not satisfied At that time, a closed-loop adjustment is performed to control the duration of the watermark injection window. and watermark injection power amplitude Make adjustments and re-execute steps S3 and S4 until the output result meets the requirements. This closed-loop adjustment enables the link of enhanced observability—path constraint formation—deviation solution—uncertainty control to be repeatedly executed and converge to a calibration output that meets the threshold parameters.

[0075] An online deviation calibration system for electricity meters includes:

[0076] The measurement freeze module is used to complete the acquisition of multi-source metering data, topology association, and time reference unification before freezing electricity collection within the transformer area topology level.

[0077] The excitation inversion module is used to generate a readable controlled energy watermark within the transformer area and perform relevant decoding on the transformer area's overall table to invert path loss constraints, thereby improving the observability of the calibration equation set and reducing the impact of time-varying line loss rate on deviation calculation.

[0078] The solution output module is used to construct a non-convex sparse multi-objective optimization solution process based on the energy conservation physical mechanism model and path loss constraints, and output the deviation calibration results, quality status evaluation results, prediction results and operation and maintenance measures of each energy meter under the uncertainty threshold constraint.

[0079] The measurement freeze module includes a data acquisition module, a topology association module, a freeze power trigger module, and a clock deviation recording module. The data acquisition module is used to acquire metering data from the main station, the distribution area concentrator, the meter box metering module, and the user's electricity meter. The topology association module is used to establish the hierarchical mapping relationship between the distribution area master meter and the user's electricity meter. The freeze power trigger module is used to complete the meter time synchronization before acquiring the freeze power. The clock deviation recording module is used to record the clock deviation introduced by the time synchronization.

[0080] The excitation inversion module includes a calibration watermark beacon module, a watermark scheduling module, a correlation decoding module, and a path loss inversion module. The calibration watermark beacon module is used to generate controlled watermark energy at the hierarchical nodes corresponding to the meter box metering module and the branch box metering module. The watermark scheduling module is used to schedule the watermark injection window and injection amplitude. The correlation decoding module is used to perform correlation decoding on the watermark sequence at the main meter side of the transformer area. The path loss inversion module is used to form path loss constraints based on the decoding results.

[0081] The solution output module includes an anomaly handling module, an energy conservation modeling module, a non-convex sparse multi-objective solution module, an uncertainty assessment module, a testing and verification module, and an operation and maintenance strategy module. Among them, the anomaly handling module is used to process anomaly factors by using time series information compression and anomaly detection; the energy conservation modeling module is used to establish a physical mechanism model of energy conservation in the transformer area; the non-convex sparse multi-objective solution module is used to couple the solution of transformer area line loss and electricity meter operating error; the uncertainty assessment module is used to redistribute the standard uncertainty components and satisfy the uncertainty threshold constraint; the testing and verification module is used to verify the accuracy of model output and uncertainty assessment based on scene simulation and digital twin mixed reality technology; and the operation and maintenance strategy module is used to output the calibration results, quality status evaluation results, prediction results, and operation and maintenance measures for each electricity meter.

[0082] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0083] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An online deviation calibration method for electricity meters, characterized in that: Includes the following steps: S1: Construct the hierarchical topology measurement relationship of the distribution area and collect the metering data of the main meter and user electricity meters in the distribution area. At the same time, collect the hierarchical node data of the meter box metering module and the branch box metering module. Trigger control is performed on the frozen power collection operation and the meter time synchronization is completed before the frozen power collection. The clock deviation of the time synchronization process is recorded and a unified time reference is formed. The unified time reference is used to align the frozen power collection time with each sampling time. S2: Based on the distribution changes of multi-dimensional dynamic and static data of transformer substations and electricity meters, time series information compression and anomaly detection are performed. Time series information compression refers to the aggregation and resampling of the collected sequence according to a preset time window to reduce data redundancy. The identified abnormal disturbance factors are processed and a high-quality balanced dataset is output to form a usable sample set for subsequent modeling and solving. S3: In the selected hierarchical nodes of the layered topology of the substation, the controlled energy injection scheduling of the calibration watermark beacon is executed, the corresponding data on the substation's total table side is collected synchronously and the relevant decoding is performed, the path loss constraint is inverted based on the decoding result and associated with the layered topology, and the calibration watermark beacon is the correlated decoded power disturbance contribution formed by the controlled calibration watermark injection on the substation's total table side. S4: Based on the physical mechanism model of energy conservation in the transformer area, a deviation solution model is established, and the path loss constraint is introduced into the solution process to weaken the influence of time-varying line loss on deviation calculation. Non-convex sparse multi-objective solution is performed to obtain the online deviation estimate of each user's electricity meter. The uncertainty of the solution result is evaluated and the standard uncertainty component is redistributed to meet the threshold constraint. Non-convex sparse multi-objective solution refers to the optimization solution that applies sparse constraints to the deviation vector under the objective function of simultaneously minimizing the line loss term and the deviation term. S5: Output the calibration results, quality status evaluation results, and quality status prediction results of each electricity meter, and generate corresponding operation and maintenance measures. For cases where the uncertainty threshold constraint is not met, perform closed-loop adjustment of the watermark injection strategy and data interaction strategy. Verify the output results through the scenario simulation and digital twin mixed reality evaluation and verification platform and form a traceable record. Perform scenario simulation verification of the interaction between the digital twin model and the measured data.

2. The online deviation calibration method for electricity meters according to claim 1, characterized in that: Step S1 includes: S1-1: Establish a hierarchical mapping relationship between the main meter of the distribution area and the user's electricity meter, and establish a physical mechanism model of energy conservation in the distribution area under the unified time base, satisfying the following equation: ,in The total power supply of the transformer substations is the amount supplied to the transformer substations under the unified time base. This refers to the number of user electricity meters corresponding to the total meter for the aforementioned distribution area. For the first The electricity consumption characteristic of each user's electricity meter under the unified time base. For the first Online deviation of individual user's electricity meter For fixed losses in the transformer area, This refers to line loss. S1-2: Before triggering the freeze on electricity collection, perform meter time synchronization and record the clock deviation to align the collected data from the main meter of the distribution area, the meter box metering module, the branch box metering module, and the user's electricity meter to the unified time reference, so that the energy conservation physical mechanism model of the input distribution area is... With each To ensure that the same time base condition is met, so as to avoid time inconsistencies. Amplified by unbiased factors.

3. The online deviation calibration method for electricity meters according to claim 2, characterized in that: S2 includes: S2-1: Based on the unified time base dataset formed in step S1, obtain the power supply sequence of the transformer substation master meter, the power consumption sequence of each user's energy meter, the hierarchical mapping relationship between the transformer substation master meter and user energy meters, the frozen power collection trigger time, and the number of clock deviation records. Index and align the above data according to the unified time base to form a sample set to be processed. ; S2-2: For the sample set The power supply sequence and power consumption sequence of each sampling time are subjected to integrity verification to obtain a set of missing markers. For samples with missing markers, missing value imputation is performed at the same time reference. The missing value imputation includes: generating surrogate values ​​for short-term missing values ​​using forward imputation rules, and generating surrogate values ​​for consecutive missing values ​​using local interpolation rules, thereby obtaining a complete sample set. ; S2-3: Calculate the time consistency index for each sampling moment based on the clock deviation record. If the time consistency index exceeds a preset time consistency threshold, mark the corresponding sampling moment as a time-inconsistent sample and perform time window correction on the sampling window to which the sample belongs. The time window correction is as follows: shift the boundary of the sampling window corresponding to the frozen electricity according to the clock deviation record, so that the power supply sequence of the transformer area and the electricity consumption sequence within the sampling window meet the same frozen electricity triggering benchmark, thereby obtaining a time-consistent sample set. ; S2-4: Construct the energy conservation residual sequence for each sampling time based on the physical mechanism model of energy conservation in the transformer area. The energy conservation residual sequence is used to characterize the deviation between the power supply sequence and the power consumption sequence of the transformer area under physical constraints. A mutation detection is performed on the energy conservation residual sequence to obtain a mutation sample set. A persistent offset detection is performed on the energy-conserving residual sequence to obtain an offset sample set. , will belong to and The samples are labeled as anomalous perturbation candidate samples; S2-5: Perform a topology consistency check on the hierarchical mapping relationship at each sampling time. The topology consistency check includes: checking the number of user energy meters corresponding to the total meter of the distribution area. The system checks whether changes at adjacent sampling times satisfy preset topology stability constraints and verifies whether the user's electricity meter's affiliation in the hierarchical mapping relationship deviates from the topology configuration record maintained by the data acquisition master station. Samples that do not satisfy the topology stability constraints are marked as a set of topology-inconsistent samples. ; S2-6: Process the abnormal perturbation candidate samples and topologically inconsistent samples separately to obtain a high-quality balanced dataset. The treatment includes: processing the set of mutant samples. Perform a culling process on the continuously offset sample set. Perform segmentation processing and retain stable segments within the segment boundaries for topologically inconsistent sample sets. Perform weight reduction processing and decrease its constraint weights in subsequent solutions to improve the quality of the balanced dataset. This serves as the basis for the input data of the total surface observation power of the transformer area in step S3, and as the basis for the input data of the energy conservation modeling and online deviation solution in step S4.

4. The online deviation calibration method for electricity meters according to claim 3, characterized in that: Step S3 includes: S3-1: Perform controlled calibration watermark injection on selected hierarchical nodes in the tiered topology of the transformer substation. The selected hierarchical nodes are those that satisfy topology consistency verification and whose corresponding branches have controllable load modulation capabilities. Controllable load modulation capability refers to the ability to change the active power of the branch according to a preset watermark sequence within the watermark injection window. The controlled calibration watermark injection is as follows: under a unified time reference, within the preset watermark injection window, apply controlled power modulation to the power consumption branches corresponding to the hierarchical nodes, so that a calibration watermark contribution with decodeable characteristics is superimposed on the overall meter side observation measurement of the transformer substation. The watermark injection window is derived from the high-quality balanced dataset output in step S2. Select from the corresponding stable segments, and avoid the mutation samples, continuous offset samples and topologically inconsistent samples marked in step S2. The controlled power modulation is achieved by executing a power setting command on the controllable load of the corresponding branch of the hierarchical node. The power setting command causes the branch power to undergo a power perturbation of a predetermined amplitude according to the calibration watermark sequence within the watermark injection window. S3-2: Within the watermark injection window, the total surface-side observation power of the station area satisfies: ,in The total meter reading of the transformer area at the sampling time The observation power, High-quality balanced dataset taken from the output of step S2 The observed power sequence of the central distribution area, wherein the observed power sequence is composed of the distribution area power supply sequence at sampling intervals. The conversion yields, For background load power that does not include the contribution of calibration watermark, The number of hierarchical nodes participating in controlled calibration watermark injection within the watermark injection window. For the first The watermark injection power amplitude of each level node For the first The calibration watermark sequence corresponding to each hierarchical node is a preset decodable sequence, used to extract the watermark contribution on the master table side through relevant decoding. For the first The path transmission coefficient from each hierarchical node to the overall table of the distribution area. For the observed noise term; S3-3: Within the watermark injection window, observe the total surface power of the station area. With the calibration watermark sequence Correlation decoding is performed to obtain an estimate of the path transmission coefficients, wherein the correlation decoding satisfies... ,in The path transmission coefficient The estimated value, The number of sampling points within the watermark injection window; S3-4: Based on Calculate the estimated watermark contribution of the watermark injection power on the total surface of the distribution area. The watermark contribution estimate was used to estimate the total surface-side observed power of the station area. Watermark contribution deduction is performed to obtain a background load power estimate, which is then used in subsequent step S4 to reduce the line loss term. The difference term, which is limited to the background load power estimate, is constrained by the path transmission coefficient estimate, thereby reducing the coupling interference of the line loss rate on the online deviation solution result.

5. The online deviation calibration method for electricity meters according to claim 4, characterized in that: S4 includes: S4-1. Using the physical mechanism model of energy conservation in the transformer area as the physical constraint and the estimated background load power as the basis for line loss constraint, a coupled solution model of line loss term and online deviation is constructed so that a solvable set of equations can still be formed under the condition that the user's power load variation range is limited and the acquisition frequency is low. S4-2. In the coupled solution process, the online deviation of each user's electricity meter is used as the basis. As the solution object, a sparsity constraint is introduced to ensure that only a finite number of user electricity meter deviations appear as significant deviations in the solution results. This is to match the operational characteristics of a relatively limited number of misaligned meters in a batch misalignment replacement scenario. Simultaneously, measurement uncertainty is assessed and controlled in the solution results to minimize the relative expanded uncertainty. Meets preset threshold parameters The combined standard uncertainty With relative expanded uncertainty satisfy: , ,in For the first The standard uncertainty of the uncertainty component For the number of uncertainty components, As a coverage factor, the uncertainty components include at least the uncertainty components introduced by line loss estimation, clock skew, transformer area meter error, and reading error, and are satisfied by redistributing the standard uncertainty components. This ensures that the reliability of the online deviation solution results can be quantitatively controlled under the threshold parameter constraint, whereby... The parameters are determined based on the allowable error limit parameters corresponding to the metering level of the electricity meter and the preset proportional coefficient parameters.

6. The online deviation calibration method for electricity meters according to claim 5, characterized in that: S5 includes: S5-1, Output the online deviation for each user's electricity meter. and with corresponding It outputs the energy meter quality status evaluation results and quality status prediction results, and generates operation and maintenance measures corresponding to the quality status evaluation results, so that the operation and maintenance measures and the quantitative results of online deviation are established in a one-to-one correspondence. S5-2, when Not satisfied At that time, a closed-loop adjustment is performed to control the duration of the watermark injection window. and watermark injection power amplitude Make adjustments and re-execute steps S3 and S4 until the output result meets the requirements. .

7. An online deviation calibration system for electricity meters, characterized in that: include: The measurement freeze module is used to complete the acquisition of multi-source metering data, topology association, and time reference unification before freezing electricity collection within the transformer area topology level. The excitation inversion module is used to generate a readable controlled energy watermark within the transformer area and perform relevant decoding on the transformer area's overall table to invert path loss constraints, thereby improving the observability of the calibration equation set and reducing the impact of time-varying line loss rate on deviation calculation. The solution output module is used to construct a non-convex sparse multi-objective optimization solution process based on the energy conservation physical mechanism model and path loss constraints, and output the deviation calibration results, quality status evaluation results, prediction results, and operation and maintenance measures of each energy meter under uncertainty threshold constraints.

8. The online deviation calibration system for electricity meters according to claim 7, characterized in that: The measurement freezing module includes a data acquisition module, a topology association module, a frozen power triggering module, and a clock deviation recording module. The data acquisition module is used to acquire metering data from the main station, the distribution area concentrator, the meter box metering module, and the user's electricity meter. The topology association module is used to establish a hierarchical mapping relationship between the distribution area main meter and the user's electricity meter. The frozen power triggering module is used to complete the meter time synchronization before acquiring the frozen power. The clock deviation recording module is used to record the clock deviation introduced by the time synchronization. The excitation inversion module includes a calibration watermark beacon module, a watermark scheduling module, a correlation decoding module, and a path loss inversion module. The calibration watermark beacon module is used to generate controlled watermark energy at the hierarchical nodes corresponding to the meter box metering module and the branch box metering module. The watermark scheduling module is used to schedule the watermark injection window and injection amplitude. The correlation decoding module is used to perform correlation decoding on the watermark sequence at the main meter side of the transformer area. The path loss inversion module is used to form a path loss constraint based on the decoding result. The solution output module includes an anomaly handling module, an energy conservation modeling module, a non-convex sparse multi-objective solution module, an uncertainty assessment module, a testing and verification module, and an operation and maintenance strategy module. The anomaly handling module uses time-series information compression and anomaly detection to process anomaly factors. The energy conservation modeling module establishes a physical mechanism model of energy conservation in the transformer area. The non-convex sparse multi-objective solution module couples the solution of transformer area line loss and electricity meter operating error. The uncertainty assessment module redistributes the standard uncertainty components and satisfies uncertainty threshold constraints. The testing and verification module verifies the accuracy of the model output and uncertainty assessment based on scene simulation and digital twin mixed reality technology. The operation and maintenance strategy module outputs the calibration results, quality status evaluation results, prediction results, and operation and maintenance measures for each electricity meter.

Images