Smart calibration method and system for electricity meters

Abstract

This application relates to the field of electricity meter calibration technology, and provides an intelligent calibration method and system for electricity meters. The method includes: after calibration triggering, switching the bus-level electricity meter to a bus-level electrical isolation scenario via a switching relay; extracting the first set of bus-level metering values ​​after injecting a full-condition steady-state signal; calculating the first set of bus-level errors by comparing with theoretical values; determining whether to switch to a branch-level electrical isolation scenario based on error distribution characteristics; if switching, dynamically loading multi-dimensional disturbance load combinations in K branch circuits via a programmable load module to simultaneously measure the second set of bus-level metering values; performing branch-level energy consumption error tracing calculation on the second set of values ​​and executing closed-loop calibration. This addresses the technical problem of traditional electricity meter calibration processes being unable to respond to multi-dimensional disturbance load environments in real time, resulting in low calibration accuracy and inaccurate error distribution reflection. The method aims to improve the accuracy of electricity meter calibration in multiple scenarios and reduce equipment calibration time.

Description

Technical Field

[0001] This application relates to the field of electricity meter calibration technology, specifically to a smart calibration method and system for electricity meters. Background Technology

[0002] Electricity meters, as crucial devices for power metering and management, are widely used in power measurement, load management, and smart billing. Traditional calibration is typically performed under a single steady-state load, making it difficult to address the cumulative metering deviations caused by dynamic disturbances in multi-branch circuits during actual operation (such as harmonic pollution, load abrupt changes, and backfeeding from distributed power sources). Existing technologies lack the ability to trace the source of energy consumption interactions at the branch level and rely on manual offline calibration, leading to systematic error drift in electricity meters under real-world scenarios. Especially in multi-branch power supply systems in industrial and commercial applications, the metering accuracy of bus-based electricity meters is affected by nonlinear load coupling interference in branch circuits. Existing testing equipment cannot simultaneously simulate dynamic load combinations and electrical isolation scenarios, causing calibration results to deviate from actual operation. Summary of the Invention

[0003] This application provides an intelligent calibration method and system for electricity meters, aiming to solve the technical problems of low calibration accuracy and inaccurate reflection of error distribution in complex scenarios caused by the inability of traditional electricity meter calibration to respond to multi-dimensional disturbance load environments in real time. The method achieves the technical effect of improving the accuracy of electricity meter calibration in multiple scenarios and reducing equipment calibration time through phased electrical isolation and dynamic multi-dimensional disturbance load loading.

[0004] The first aspect disclosed in this application provides an intelligent calibration method for an energy meter, the method comprising: after calibration is triggered, switching the bus energy meter to a bus-level electrical isolation scenario via a switching relay; in the bus-level electrical isolation scenario, injecting a full-condition steady-state signal strategy into the bus energy meter, performing metering data extraction, and outputting a first set of bus-level metering values; calculating the test error by comparing the first set of bus-level metering values ​​with the full-condition theoretical values, and outputting a first set of bus-level errors; determining a scenario switching based on the error distribution characteristics of the first set of bus-level errors, and if a switching is determined, switching the bus energy meter to a branch-level electrical isolation scenario via the switching relay; in the branch-level electrical isolation scenario, simultaneously metering a second set of bus-level metering values ​​of the bus energy meter during the dynamic loading of multi-dimensional disturbance load combinations in K branch circuits by a programmable load module; performing branch-level energy consumption error tracing calculation on the second set of bus-level metering values, and performing closed-loop calibration on the bus energy meter based on the calculation results.

[0005] Another aspect of this application discloses an intelligent calibration system for electricity meters, the system comprising: a calibration triggering unit: after calibration triggering, switching the bus electricity meter to a bus-level electrical isolation scenario via a switching relay; a first metering data extraction unit: in the bus-level electrical isolation scenario, after injecting a full-condition steady-state signal strategy into the bus electricity meter, performing metering data extraction, and outputting a first set of bus-level metering values; a test error calculation unit: calculating the test error by comparing the first set of bus-level metering values ​​with the full-condition theoretical values, and outputting a first set of bus-level errors; and a scenario switching judgment unit. The first unit: determines the scene switching based on the error distribution characteristics of the first group of bus-level errors. If a switching is determined, the bus energy meter is switched to the branch-level electrical isolation scene via the switching relay. The second metering data extraction unit: in the branch-level electrical isolation scene, the second group of bus-level metering values ​​of the bus energy meter are simultaneously measured by the programmable load module during the dynamic loading of multi-dimensional disturbance load combinations in K branch circuits. The closed-loop calibration unit: performs branch-level energy consumption error tracing calculation on the second group of bus-level metering values ​​and performs closed-loop calibration on the bus energy meter based on the calculation results.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The aforementioned intelligent calibration method for electricity meters achieves automatic calibration through intelligent control. After calibration is initiated, the electricity meter is first switched to a bus-level electrical isolation scenario via a relay. Subsequently, a full-condition steady-state signal strategy is injected into the electricity meter, data extraction is performed, and the first set of bus-level metering values ​​is obtained. Then, by comparing these values ​​with the theoretical values ​​under full-condition conditions, the test error is calculated, and the first set of error data is output. Next, based on the error distribution characteristics, it is determined whether to switch to a branch-level electrical isolation scenario. If a switch is needed, the electricity meter is switched to the branch-level electrical isolation scenario via a relay. In this scenario, a multi-dimensional disturbance load is loaded through a programmable load module, and the second set of bus-level metering values ​​is simultaneously measured. Finally, based on the second set of metering values, energy consumption error tracing calculation is performed, and closed-loop calibration is ultimately executed to ensure the high accuracy and high stability of the electricity meter.

[0008] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0009] 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.

[0010] Figure 1 This is a flowchart illustrating a smart calibration method for an electricity meter in one embodiment.

[0011] Figure 2 This is an architecture diagram of a smart calibration system for an electricity meter in one embodiment.

[0012] Explanation of reference numerals in the attached figures: Calibration trigger unit 11, First measurement data extraction unit 12, Test error calculation unit 13, Scene switching judgment unit 14, Second measurement data extraction unit 15, Closed-loop calibration unit 16. Detailed Implementation

[0013] This application provides an intelligent calibration method and system for electricity meters, which solves the technical problems of low calibration accuracy and inaccurate error distribution in complex scenarios caused by the inability of traditional electricity meter calibration to respond to multi-dimensional disturbance load environments in real time. The method achieves the technical effect of improving the accuracy of electricity meter calibration in multiple scenarios and reducing equipment calibration time by using phased electrical isolation and dynamic multi-dimensional disturbance load loading.

[0014] The technical solutions of the embodiments of this application 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 this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0015] It should be noted that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product, or device.

[0016] Example 1, as Figure 1 As shown, this application provides a smart calibration method for electricity meters, the method comprising:

[0017] After calibration is triggered, the bus energy meter is switched to a bus-level electrical isolation scenario by switching relays.

[0018] In this embodiment, after calibration is initiated, the bus energy meter is first switched from its current operating state to an electrically isolated scenario via relay operation. This process ensures that the bus energy meter is disconnected from other circuits, thereby avoiding external interference and providing a clean and independent calibration environment. In this isolated state, the energy meter can perform accurate metering data acquisition and calibration without external influence, ensuring the accuracy of the calibration results.

[0019] Furthermore, this application provides a method for switching the bus energy meter to a bus-level electrical isolation scenario by switching relays after calibration triggering, the method comprising:

[0020] The switching relay performs a branch load blocking operation, electrically disconnecting the bus energy meter from the K branch circuits; after performing a grid input disconnection operation, the switching relay connects the bus energy meter to a standard source, wherein the standard source is used to inject the full-condition steady-state signal strategy into the bus energy meter.

[0021] Preferably, after calibration is triggered, the switching relay first performs a branch load blocking operation. This means it disconnects the bus energy meter from the K branch circuits, thus completely isolating the energy meter from these branch circuits and ensuring that the calibration process is not affected by external load changes. Subsequently, the switching relay performs a grid input disconnection operation, cutting off the connection between the bus energy meter and the grid. This step aims to eliminate the influence of the grid on the energy meter, providing a stable environment for subsequent accurate calibration. Afterward, the switching relay connects the bus energy meter to a standard source. The standard source provides a known and stable power signal to inject a full-condition steady-state signal strategy into the bus energy meter. This steady-state signal strategy includes different voltage, current, and power factor information to simulate the meter's operating state under various load conditions, ensuring accurate calibration of the energy meter under all operating conditions.

[0022] Table 1: Example Table of Steady-State Signaling Strategies under All Operating Conditions

[0023]

[0024] As shown in Table 1, the full-condition steady-state signal strategy example table covers typical operating conditions that the energy meter may encounter, including different voltage levels, load levels, load characteristics and power factor combinations, which are used to verify the metering accuracy and stability of the energy meter under full operating conditions.

[0025] Furthermore, this application provides a method for switching the bus energy meter to a bus-level electrical isolation scenario by switching relays after calibration triggering. Previously, the method included:

[0026] The system locally accesses the multi-source historical state data of the bus energy meter; after extracting the multi-scale steady-state running time from the multi-source historical state data, it compares and extracts the minimum value to generate a timing trigger threshold; it calculates the active power standard deviation of the multi-source historical state data through a sliding window to generate a first event trigger threshold; it calculates the range ratio of the effective values ​​of the branch circuit current based on the multi-source historical state data to generate a second event trigger threshold; and it inputs the timing trigger threshold, the first event trigger threshold, and the second event trigger threshold into the state machine to construct a calibration trigger condition matching engine.

[0027] Preferably, firstly, multi-source historical status data related to the bus energy meter is acquired via local calls. This data comes from different monitoring sources and includes the energy meter's operating status information at different time periods, such as voltage, current, and power. Then, multi-scale steady-state operating duration data is extracted from the acquired multi-source historical status data. This data reflects the duration of the energy meter in a stable operating state. By comparing these duration data, the minimum value is found, and a timing trigger threshold is generated based on this minimum value as the start time point for the subsequent calibration process. Next, according to a preset time window size, a sliding window is used to calculate the standard deviation of the active power in the multi-source historical status data. The standard deviation reflects the fluctuation range of the power data. The calculation of the standard deviation generates a first event trigger threshold, which is used to determine whether the power fluctuation exceeds a predetermined standard, thereby triggering a calibration event. Furthermore, based on the multi-source historical status data, the range ratio of the effective values ​​of the branch circuit current (i.e., the ratio of the current range to the current mean) is calculated to reflect the magnitude of current changes. The calculation of the range ratio generates a second event trigger threshold, which is used to determine whether the current fluctuation exceeds a predetermined standard, thereby triggering relevant calibration operations. Finally, the generated timed trigger threshold, first event trigger threshold, and second event trigger threshold are input into the state machine. The state machine performs logical judgments based on these threshold conditions to construct a calibration trigger condition matching engine. When one or more of these conditions are met, the calibration process will be triggered, ensuring that the electricity meter enters the calibration state at the appropriate time. Through this process, the automation and precise triggering of electricity meter calibration can be achieved, improving the reliability and real-time performance of calibration.

[0028] In a bus-level electrical isolation scenario, after injecting a full-condition steady-state signal strategy into the bus energy meter, metering data extraction is performed, and the first set of bus-level metering values ​​is output.

[0029] In one embodiment, in a bus-level electrical isolation scenario, a full-condition steady-state signal strategy is first injected into the bus energy meter via a standard source. This signal strategy includes parameters such as voltage, current, and power factor of the energy meter under various operating conditions, ensuring that the energy meter is in a stable operating state. After the signal injection, the energy meter begins metering operations based on these input signals. At this time, a metering data extraction operation is performed to read the metering data generated by the energy meter. This data includes the voltage, current, power, and other values ​​measured by the energy meter. After processing, the first set of bus-level metering values ​​is output. These metering values ​​serve as the basis for the calibration process, providing actual operating data of the energy meter under specific operating conditions.

[0030] Furthermore, this application provides a method for injecting a full-condition steady-state signal into the bus energy meter in a bus-level electrical isolation scenario, followed by metering data extraction and outputting a first set of bus-level metering values. The method includes:

[0031] According to the application scenario of the bus energy meter, a full-condition strategy information is matched, wherein the full-condition strategy information consists of a voltage condition rated sequence, a current condition range classification, and a power factor rated sequence; after randomly perturbing the full-condition strategy information using interactive multi-condition fluctuation scales, multiple sets of steady-state test signals are constructed based on orthogonal decomposition; the multiple sets of steady-state test signals are encapsulated in ascending voltage gradient structure to form a full-condition steady-state signal injection sequence, which serves as the full-condition steady-state signal strategy; during the process of injecting the full-condition steady-state signal injection sequence into the bus energy meter by controlling the standard source in discrete time steps, metering data extraction is performed synchronously, and the first set of bus-level metering values ​​is output.

[0032] Optionally, based on the application scenario of the bus energy meter, keyword matching is performed on the operating condition descriptions in the full-condition strategy library to extract full-condition strategy information that matches the current application scenario. This strategy information includes voltage operating condition rated sequences, current operating condition range classifications, and power factor rated sequences, which respectively represent the operating states of the energy meter under different voltage, current ranges, and power factors. Subsequently, the operating condition parameters in the matched full-condition strategy information are randomly perturbed to simulate the energy meter's response under different operating conditions, making the test signals closer to the various environmental changes that may be encountered in actual operation. Based on the perturbed full-condition strategy information, an orthogonal decomposition method is used to decompose it into multiple sets of independent steady-state test signals. Specifically, the perturbed full-condition strategy information is first centered, that is, the mean of each signal is subtracted from each data point of each signal to make the data have a zero mean. Then, based on the centered data, the covariance matrix is ​​calculated to reflect the linear relationship and variance between the signals. Next, eigenvalue decomposition is performed on the covariance matrix to obtain eigenvalues ​​and eigenvectors. Eigenvalues ​​represent the importance of each principal component, and eigenvectors represent the direction of the principal components. Based on the magnitude of the eigenvalues, the eigenvector corresponding to the largest eigenvalue is selected as the principal component; typically, the eigenvectors corresponding to the k largest eigenvalues ​​are chosen. By multiplying the centered data matrix with the selected eigenvector matrix, the orthogonal components (i.e., principal components) of each signal are obtained. These orthogonal components are independent of each other and serve as a set of steady-state test signals. These signals reflect the stable operating state of the energy meter under different conditions, which helps to accurately test various indicators of the energy meter during calibration. Then, the obtained multiple sets of steady-state test signals are structured and encapsulated according to the ascending order of voltage gradient. That is, these test signals are organized in order of voltage from low to high, forming a series of ordered test signal sequences to ensure the step-by-step nature of signal injection and the stability of the system. After encapsulation, all the steady-state test signals constitute a complete full-condition steady-state signal injection sequence. This sequence will serve as the full-condition steady-state signal strategy for subsequent accurate calibration of the energy meter. Finally, using standard source devices, the constructed full-condition steady-state signal is injected into the bus energy meter in a discrete timing and step-by-step control manner. During the signal injection process, the metering data extraction operation is performed synchronously to read the metering data of the energy meter in real time. Through these data, the first set of bus-level metering values ​​is generated, which provide data support for subsequent error calculation and calibration processes.

[0033] The test error is calculated by comparing the first set of bus-level measurement values ​​with the theoretical values ​​under all operating conditions, and the first set of bus-level errors is output.

[0034] In one embodiment, during the calibration process, the first set of acquired bus-level measurement values ​​is compared with the theoretical values ​​under full operating conditions. The theoretical values ​​under full operating conditions are the values ​​that the energy meter should display under known ideal operating conditions, typically calculated based on standard parameters such as voltage, current, and power factor. By comparing the differences between the actual measurement values ​​and the theoretical values, the first set of bus-level errors is obtained. This first set of bus-level errors reflects the performance deviation of the energy meter under the current operating conditions, providing a basis for subsequent calibration and adjustment.

[0035] Based on the error distribution characteristics of the first group of bus-level errors, a scene switching judgment is made. If a switching is determined, the bus energy meter is switched to the branch-level electrical isolation scene through the switching relay.

[0036] In one embodiment, after error calculation, the distribution characteristics of the first group of bus-level errors are analyzed to evaluate the form and extent of the errors. For example, statistical characteristics such as error fluctuation range and standard deviation are checked. If the distribution characteristics of these errors meet a preset switching standard, it is determined whether a scenario switch is needed. This determination decides whether the bus energy meter needs to be switched from the current bus-level electrical isolation scenario to a branch-level electrical isolation scenario. If a switch is determined to be necessary, an operation is performed via a switching relay to switch the bus energy meter to the branch-level electrical isolation scenario. This switching operation ensures that the energy meter can continue to be accurately calibrated in the new test environment, avoiding the influence of existing errors on subsequent processes.

[0037] Furthermore, this application provides a method for determining scene switching based on the error distribution characteristics of the first group of bus-level errors. If a switching is determined, the bus energy meter is switched to a branch-level electrical isolation scene via the switching relay. The method includes:

[0038] Based on preset correlation indicators, multidimensional features are extracted from the first group of bus-level errors, and bus-level error features are output. The bus-level error features include bus-level range, bus-level standard deviation, bus-level skewness coefficient, and bus-level over-threshold ratio. The error distribution pattern is determined based on the bus-level error features, and the distribution pattern category is output. The switching decision rule base is matched according to the distribution pattern category. If the decision output is a branch-level isolation command, the switching relay is driven to switch the bus energy meter to the branch-level electrical isolation scenario.

[0039] Preferably, firstly, based on preset correlation indicators, multi-dimensional feature data is extracted from the first group of bus-level errors to obtain bus-level error features. These bus-level error features include bus-level range, bus-level standard deviation, bus-level skewness coefficient, and bus-level over-threshold ratio. Among them, the bus-level range is the difference between the maximum and minimum error values, reflecting the overall range of error variation; the bus-level standard deviation is the average level of the difference between the error value and the error mean, used to measure the dispersion of error data. The larger the standard deviation, the greater the error fluctuation; the bus-level skewness coefficient is obtained by measuring the asymmetry of the error data distribution. If the data is biased to the right (more large error values), the skewness coefficient is positive; if the data is biased to the left (more small error values), the skewness coefficient is negative; if the data is symmetrically distributed, the skewness coefficient is close to zero; the bus-level over-threshold ratio is the proportion of errors exceeding the set error threshold, helping to determine whether the error is abnormal. Subsequently, based on the extracted bus-level error characteristics, the error distribution pattern is analyzed to determine the distribution pattern category. In this process, by comparing with known error distribution types, the current error distribution characteristics are determined. For example, if during calibration, the bus-level range is 0.1%~0.3%, the bus-level standard deviation is 0.02%~0.08%, the bus-level skewness coefficient is -0.2~+0.2, and the bus-level over-threshold ratio is 0.5%~2%, this may indicate that the error data follows an ideal normal distribution; if the bus-level range is 0.3%~0.8%, the bus-level standard deviation is 0.08%~0.2%, the bus-level skewness coefficient is -0.5~+0.5, and the bus-level over-threshold ratio is 2%~5%, this may indicate that the error data follows a conventional normal distribution. Subsequently, based on the determined error distribution pattern category, a matching switching decision rule is searched from the switching decision rule base. If the error pattern meets the criteria for switching to a branch-level electrical isolation scenario, a branch-level isolation command is generated through the switching decision rule base to activate the switching relay, thereby switching the bus energy meter from the current bus-level electrical isolation scenario to a branch-level electrical isolation scenario. This switching ensures that the energy meter can continue to be calibrated in the new scenario, avoiding adverse effects of errors in the current scenario on subsequent calibrations, and ensuring calibration accuracy and efficiency.

[0040] In a branch-level electrical isolation scenario, the second set of bus-level metering values ​​of the bus energy meter are simultaneously measured by the programmable load module during the dynamic loading of multi-dimensional disturbance load combinations on K branch circuits.

[0041] In one embodiment, in a branch-level electrical isolation scenario, a multi-dimensional disturbance load combination is first dynamically loaded onto K branch circuits via a programmable load module. These load combinations are ordered in time and are loaded progressively according to a preset timing sequence. The programmable load module precisely aligns the output of each load on each branch circuit according to the timing sequence, thereby simulating the energy meter response under different load conditions. During load loading, the bus energy meter begins synchronous metering operations, collecting load disturbance data from each branch circuit in real time. After processing, this data outputs a second set of bus-level metering values, reflecting the actual metering situation of the energy meter in the branch-level electrical isolation scenario. Through the timing control of the programmable load module, the loading of load disturbances is ensured to be synchronized with the metering process of the energy meter, thus providing accurate data support for subsequent error tracing and closed-loop calibration.

[0042] Furthermore, this application provides a method for simultaneously measuring a second set of bus-level metering values ​​of the bus energy meter in a branch-level electrical isolation scenario, during the dynamic loading of multi-dimensional disturbance load combinations on K branch circuits via a programmable load module. The method includes:

[0043] K historical electricity consumption data of the K branch circuits are collected; based on the recurrence frequency of load characteristics of the K historical electricity consumption data, characteristic load feature vectors are extracted to construct a K branch load feature library; the load feature vectors of the K branch load feature library are orthogonally decomposed and randomly combined to dynamically output a multidimensional disturbance load combination, wherein the multidimensional disturbance load combination consists of F multidimensional disturbance load vectors, F≥50; the programmable load module loads the F multidimensional disturbance load vectors in the K branch circuits in a discrete time sequence, and synchronously collects the metering data of the bus energy meter in the steady state range to obtain the second set of bus-level metering values, wherein the second set of bus-level metering values ​​includes F second bus-level metering values.

[0044] Preferably, firstly, historical power consumption data is collected from each of the K branch circuits. This data records the power consumption of each branch circuit over a period of time, reflecting the load fluctuations and operating status of each branch circuit. Then, based on the collected K historical power consumption data, the recurrence frequency of load characteristics in each branch circuit is analyzed. This means statistically analyzing the frequency of recurrence of load characteristics over time, extracting the most frequent resistive power factor as the resistive load power factor, the most frequent capacitive power factor as the resistive load power factor, the most frequent inductive power factor as the inductive auxiliary power factor, and the most frequent nonlinear load characteristic as the nonlinear load parameter. By concatenating the extracted resistive load power factor, capacitive load power factor, inductive auxiliary power factor, and nonlinear load parameter, a characteristic load feature vector of the K historical power consumption data is obtained, thus constructing a load feature library for the K branch circuits. Next, a similar orthogonal decomposition is performed on the characteristic load feature vector in the K branch load feature library, decomposing the complex load feature vector into mutually independent parts, facilitating subsequent combination and analysis. After orthogonal decomposition, the results are randomly combined to generate F different load combinations (F is greater than or equal to 50). Each combination corresponds to a multidimensional disturbance load vector to simulate load fluctuations that may occur under different operating conditions. These multidimensional disturbance load vectors are added to a set to form a multidimensional disturbance load combination. Then, the programmable load module, according to discrete timing control, gradually loads the F multidimensional disturbance load vectors from the multidimensional disturbance load combination into K branch circuits. Each multidimensional disturbance load vector is loaded at a specific time point to ensure that the dynamic fluctuations of the load are orderly and can simulate load changes in actual use. During the multidimensional disturbance load loading process, changes in each load combination will cause changes in the meter readings of the energy meters. Therefore, the meter readings of the bus energy meters are synchronously collected within the steady-state range of load loading to ensure data stability and reliability. Finally, the collected metering data are summarized to obtain the second set of bus-level metering values. This second set of bus-level metering values ​​includes F second bus-level metering values, which represent the performance of the bus energy meter under multi-dimensional disturbance load combinations, providing necessary data support for subsequent error calculation and calibration.

[0045] Furthermore, this application provides a method for orthogonally decomposing and randomly combining the load feature vectors of the K branch load feature libraries to dynamically output multidimensional perturbation load combinations. The method includes:

[0046] Based on the recurrence sequence of the K historical power consumption data according to the K branch load feature databases, cross-branch load conflict identification is performed as the first load combination limit; the bus capacity is obtained interactively as the second load combination limit; with the first load combination limit and the second load combination limit as constraints, the load feature vectors of the K branch load feature databases are orthogonally decomposed and randomly combined to dynamically output multi-dimensional disturbance load combinations.

[0047] Optionally, firstly, based on historical electricity consumption data contained in the K branch load feature libraries, the recurrence sequence of load features in each branch circuit is analyzed. The recurrence sequence refers to the recurring patterns of various load features over time. Analysis of this time-series data identifies potential load conflicts between different branch circuits. Load conflicts refer to the situation where multiple circuits may experience excessive loads or mutual interference within the same time period, preventing simultaneous stable operation. Therefore, a first load combination limit is set based on the recurring patterns to avoid these load conflicts and ensure load coordination between different circuits. After identifying load conflicts, the bus capacity is obtained interactively. The bus capacity refers to the maximum current or power that the bus energy meter can carry. Exceeding this bus capacity will affect the normal operation of the energy meter. Therefore, a second load combination limit is set based on the bus capacity to ensure that the load combination does not exceed the bus's carrying capacity. Once the first and second load combination limits are determined, the load feature vectors in the K branch load feature libraries are orthogonally decomposed. This process uses a similar method as described above, decomposing different load features into independent components through principal component analysis, facilitating subsequent combination and analysis. Finally, based on these orthogonally decomposed eigenvectors, F different load combinations are randomly generated, serving as F multidimensional disturbance load vectors. These multidimensional disturbance load vectors comply with both the first load combination constraint (avoiding cross-branch load conflicts) and the second load combination constraint (not exceeding the bus capacity). The multidimensional disturbance load combinations formed by these multidimensional disturbance load vectors will be used for subsequent testing and calibration to ensure that the bus energy meter can perform accurate metering and calibration under various load conditions.

[0048] Furthermore, this application provides a characteristic load feature vector including resistive load power factor, capacitive load power factor, inductive auxiliary power factor, and nonlinear load parameters.

[0049] Optionally, the characteristic load feature vector includes the power factor of resistive loads, the power factor of capacitive loads, the power factor of inductive auxiliary loads, and nonlinear load parameters. The power factor of a resistive load refers to the power factor of a purely resistive load, calculated by dividing the effective power of the resistive load portion by its apparent power. Effective power is the product of the voltage, current, and the cosine of the phase angle between the voltage and current in the resistive load portion; apparent power is the product of the voltage and current in the resistive load portion. The power factor of a capacitive load reflects the phase difference caused by the capacitive load, calculated by dividing the effective power of the capacitive load portion by its apparent power. The power factor of an inductive auxiliary load reflects the power factor of an inductive load, where current typically lags behind voltage; it is calculated by dividing the effective power of the inductive load and other auxiliary load portions by their apparent power. Nonlinear load parameters often lead to current waveform distortion and the generation of harmonics. In order to obtain nonlinear load parameters, Fourier transform and other methods are used to extract the amplitude and frequency of each harmonic. By calculating the harmonic content in the current and voltage waveforms, the harmonic parameters of the nonlinear load are obtained, including indicators such as total harmonic distortion (THD), to reflect the characteristics of the nonlinear load in the circuit.

[0050] Branch-level energy consumption error tracing calculation is performed on the second group of bus-level metering values, and closed-loop calibration is performed on the bus energy meter based on the calculation results.

[0051] In one embodiment, after obtaining the second set of bus-level metering values, the theoretical energy consumption value of the multi-dimensional disturbance load combination is combined with the second set of bus-level metering values ​​to perform error source tracing calculation. This analyzes the deviation between the actual measured values ​​and the theoretical values, thereby outputting the second set of bus-level errors. Subsequently, the fluctuation of the second set of bus-level errors is analyzed, and corresponding calibration strategy parameters are matched and generated. These parameters are based on the characteristics of the error and aim to compensate for measurement errors caused by load disturbances or environmental factors. Finally, based on the generated calibration strategy parameters, the bus energy meter is automatically calibrated in a closed loop. That is, the calibration parameters of the bus energy meter are dynamically adjusted until the error is minimized, ensuring that the metering accuracy of the bus energy meter under different load conditions meets the expected standard.

[0052] Furthermore, this application provides a method for performing branch-level energy consumption error tracing calculation on the second group of bus-level metering values, and performing closed-loop calibration on the bus energy meter based on the calculation results. The method includes:

[0053] A branch-level prediction array is constructed based on the theoretical energy consumption value of the multidimensional disturbance load combination; the branch-level prediction array is used to perform error source calculation on the second group of bus-level metering values, and the second group of bus-level errors is output; the calibration strategy parameters are matched and output according to the historical drift rate and the distribution fluctuation of the second group of bus-level errors; the calibration strategy parameters are used to perform automatic calibration on the bus energy meter.

[0054] Preferably, firstly, a branch-level prediction array is constructed based on the theoretical energy consumption value of the multi-dimensional perturbation load combination. This array reflects the ideal energy consumption of each branch circuit under different load combinations, providing a reference standard for subsequent error tracing calculations. Then, the branch-level prediction array is used to trace the error of the second set of bus-level metering values. That is, the difference between the actual measured second set of metering values ​​and the theoretical energy consumption value in the prediction array is compared to obtain the second set of bus-level errors. This second set of bus-level errors helps identify the measurement deviation of the bus energy meter under different load conditions. Next, a distribution fluctuation analysis is performed on the obtained second set of bus-level errors to calculate the bus-level error characteristics, including the bus-level range, bus-level standard deviation, bus-level skewness coefficient, and bus-level over-threshold ratio. Then, the error distribution pattern is determined based on the bus-level error characteristics to identify the current fluctuation distribution pattern. Finally, historical drift data within the most recent preset time period is extracted. This historical drift data reflects the drift rate of the energy meter under different operating conditions. The drift rate refers to the rate of change of the bus energy meter's metering value over time. By inputting historical drift rates and fluctuation distribution patterns into the calibration strategy matching rules, a calibration strategy that matches the current fluctuation is obtained, and the corresponding calibration strategy parameters are parsed out. These parameters include the adjustment magnitude, calibration frequency, and calibration timing. The aim is to dynamically adjust the calibration process based on the historical drift characteristics and current error distribution of the bus energy meter, ensuring that the metering accuracy of the bus energy meter is corrected and optimized under future load changes. Finally, using the generated calibration strategy parameters, an automated closed-loop calibration is performed on the bus energy meter. This process eliminates errors caused by load fluctuations and environmental factors by adjusting the internal calibration parameters of the bus energy meter, thereby ensuring that the bus energy meter can provide accurate metering data under various operating conditions.

[0055] In summary, the embodiments of this application have at least the following technical effects:

[0056] In this embodiment, after calibration is triggered, the bus energy meter is switched to a bus-level electrical isolation scenario via a switching relay. Then, in the bus-level electrical isolation scenario, a full-condition steady-state signal strategy is injected into the bus energy meter, and metering data extraction is performed, outputting a first set of bus-level metering values. Next, the first set of bus-level metering values ​​is compared with the theoretical values ​​under full-condition conditions to calculate the test error, outputting a first set of bus-level errors. Further, a scenario switching judgment is made based on the error distribution characteristics of the first set of bus-level errors. If a switching judgment is made, the bus energy meter is switched to a branch-level electrical isolation scenario via the switching relay. Then, in the branch-level electrical isolation scenario, the second set of bus-level metering values ​​is simultaneously measured by the programmable load module during the dynamic loading of multi-dimensional disturbance load combinations in K branch circuits. Finally, branch-level energy consumption error tracing calculation is performed on the second set of bus-level metering values, and closed-loop calibration is performed on the bus energy meter based on the calculation results. These technologies collectively solve the technical problems of traditional electricity meter calibration, which cannot respond to multi-dimensional disturbance load environments in real time, resulting in low calibration accuracy and inaccurate reflection of error distribution in complex scenarios. They achieve the technical effect of improving the accuracy of electricity meter calibration in multiple scenarios and reducing equipment calibration time through phased electrical isolation and dynamic multi-dimensional disturbance load loading.

[0057] Example 2, based on the same inventive concept as the smart calibration method for electricity meters in the foregoing examples, such as... Figure 2 As shown, this application provides an intelligent calibration system for electricity meters, the system comprising:

[0058] Calibration trigger unit 11: After calibration triggering, the bus energy meter is switched to the bus-level electrical isolation scenario via a switching relay; First metering data extraction unit 12: In the bus-level electrical isolation scenario, after injecting a full-condition steady-state signal strategy into the bus energy meter, metering data extraction is performed, and the first set of bus-level metering values ​​is output; Test error calculation unit 13: The first set of bus-level metering values ​​is compared with the full-condition theoretical values ​​to calculate the test error, and the first set of bus-level errors is output; Scenario switching judgment unit 14: Scenario switching judgment is performed based on the error distribution characteristics of the first set of bus-level errors. If a switching is judged, the bus energy meter is switched to the branch-level electrical isolation scenario via the switching relay; Second metering data extraction unit 15: In the branch-level electrical isolation scenario, the second set of bus-level metering values ​​of the bus energy meter is simultaneously measured by the programmable load module during the dynamic loading of multi-dimensional disturbance load combinations in K branch circuits; Closed-loop calibration unit 16: Branch-level energy consumption error tracing calculation is performed on the second set of bus-level metering values, and closed-loop calibration is performed on the bus energy meter based on the calculation results.

[0059] Furthermore, the calibration trigger unit 11 is also configured to perform the following method:

[0060] The switching relay performs a branch load blocking operation, electrically disconnecting the bus energy meter from the K branch circuits; after performing a grid input disconnection operation, the switching relay connects the bus energy meter to a standard source, wherein the standard source is used to inject the full-condition steady-state signal strategy into the bus energy meter.

[0061] Furthermore, the calibration trigger unit 11 is also configured to perform the following method:

[0062] The system locally accesses the multi-source historical state data of the bus energy meter; after extracting the multi-scale steady-state running time from the multi-source historical state data, it compares and extracts the minimum value to generate a timing trigger threshold; it calculates the active power standard deviation of the multi-source historical state data through a sliding window to generate a first event trigger threshold; it calculates the range ratio of the effective values ​​of the branch circuit current based on the multi-source historical state data to generate a second event trigger threshold; and it inputs the timing trigger threshold, the first event trigger threshold, and the second event trigger threshold into the state machine to construct a calibration trigger condition matching engine.

[0063] Furthermore, the first measurement data extraction unit 12 is also used to perform the following method:

[0064] According to the application scenario of the bus energy meter, a full-condition strategy information is matched, wherein the full-condition strategy information consists of a voltage condition rated sequence, a current condition range classification, and a power factor rated sequence; after randomly perturbing the full-condition strategy information using interactive multi-condition fluctuation scales, multiple sets of steady-state test signals are constructed based on orthogonal decomposition; the multiple sets of steady-state test signals are encapsulated in ascending voltage gradient structure to form a full-condition steady-state signal injection sequence, which serves as the full-condition steady-state signal strategy; during the process of injecting the full-condition steady-state signal injection sequence into the bus energy meter by controlling the standard source in discrete time steps, metering data extraction is performed synchronously, and the first set of bus-level metering values ​​is output.

[0065] Furthermore, the scene switching determination unit 14 is also used to perform the following method:

[0066] Based on preset correlation indicators, multidimensional features are extracted from the first group of bus-level errors, and bus-level error features are output. The bus-level error features include bus-level range, bus-level standard deviation, bus-level skewness coefficient, and bus-level over-threshold ratio. The error distribution pattern is determined based on the bus-level error features, and the distribution pattern category is output. The switching decision rule base is matched according to the distribution pattern category. If the decision output is a branch-level isolation command, the switching relay is driven to switch the bus energy meter to the branch-level electrical isolation scenario.

[0067] Furthermore, the second measurement data extraction unit 15 is also used to perform the following method:

[0068] K historical electricity consumption data of the K branch circuits are collected; based on the recurrence frequency of load characteristics of the K historical electricity consumption data, characteristic load feature vectors are extracted to construct a K branch load feature library; the load feature vectors of the K branch load feature library are orthogonally decomposed and randomly combined to dynamically output a multidimensional disturbance load combination, wherein the multidimensional disturbance load combination consists of F multidimensional disturbance load vectors, F≥50; the programmable load module loads the F multidimensional disturbance load vectors in the K branch circuits in a discrete time sequence, and synchronously collects the metering data of the bus energy meter in the steady state range to obtain the second set of bus-level metering values, wherein the second set of bus-level metering values ​​includes F second bus-level metering values.

[0069] Furthermore, the second measurement data extraction unit 15 is also used to perform the following method:

[0070] Based on the recurrence sequence of the K historical power consumption data according to the K branch load feature databases, cross-branch load conflict identification is performed as the first load combination limit; the bus capacity is obtained interactively as the second load combination limit; with the first load combination limit and the second load combination limit as constraints, the load feature vectors of the K branch load feature databases are orthogonally decomposed and randomly combined to dynamically output multi-dimensional disturbance load combinations.

[0071] Furthermore, the second measurement data extraction unit 15 is also used to perform the following method:

[0072] The key load characteristic vectors include the power factor of resistive loads, the power factor of capacitive loads, the power factor of inductive auxiliary loads, and nonlinear load parameters.

[0073] Furthermore, the closed-loop calibration unit 16 is also configured to perform the following method:

[0074] A branch-level prediction array is constructed based on the theoretical energy consumption value of the multidimensional disturbance load combination; the branch-level prediction array is used to perform error source calculation on the second group of bus-level metering values, and the second group of bus-level errors is output; the calibration strategy parameters are matched and output according to the historical drift rate and the distribution fluctuation of the second group of bus-level errors; the calibration strategy parameters are used to perform automatic calibration on the bus energy meter.

[0075] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0076] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0077] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A smart calibration method for electricity meters, characterized in that, The method includes: After calibration is triggered, the bus energy meter is switched to a bus-level electrical isolation scenario by switching relays; In a bus-level electrical isolation scenario, after injecting a full-condition steady-state signal strategy into the bus energy meter, metering data extraction is performed, and the first set of bus-level metering values ​​is output. The test error is calculated by comparing the first set of bus-level measurement values ​​with the theoretical values ​​under all operating conditions, and the first set of bus-level errors is output. Based on the error distribution characteristics of the first group of bus-level errors, a scene switching judgment is made. If a switching is judged, the bus energy meter is switched to the branch-level electrical isolation scene through the switching relay. In a branch-level electrical isolation scenario, the second set of bus-level metering values ​​of the bus energy meter are simultaneously measured by the programmable load module during the dynamic loading of multi-dimensional disturbance load combinations on K branch circuits. Branch-level energy consumption error tracing calculation is performed on the second group of bus-level metering values, and closed-loop calibration is performed on the bus energy meter based on the calculation results.

2. The intelligent calibration method for electricity meters as described in claim 1, characterized in that, The method includes performing branch-level energy consumption error tracing calculations on the second group of bus-level metering values, and performing closed-loop calibration on the bus energy meter based on the calculation results. A branch-level prediction array is constructed based on the theoretical energy consumption value of the multidimensional perturbation load combination; The branch-level prediction array is used to perform error source calculation on the second group of bus-level measurement values, and the second group of bus-level errors is output. Based on the historical drift rate and the distribution fluctuation of the second group of bus-level errors, the output calibration strategy parameters are matched; The bus energy meter is automatically calibrated using the calibration strategy parameters.

3. The intelligent calibration method for electricity meters as described in claim 1, characterized in that, After calibration is triggered, the bus energy meter is switched to a bus-level electrical isolation scenario by switching relays. The method includes: The switching relay performs a branch load blocking operation, electrically disconnecting the bus energy meter from the K branch circuits; After performing the grid input cut-off operation, the switching relay connects the bus energy meter to the standard source, wherein the standard source is used to inject the full-condition steady-state signal strategy into the bus energy meter.

4. The intelligent calibration method for electricity meters as described in claim 3, characterized in that, In a bus-level electrical isolation scenario, after injecting a full-condition steady-state signal strategy into the bus energy meter, metering data extraction is performed, and a first set of bus-level metering values ​​is output. The method includes: Match full-condition strategy information according to the application scenario of the bus energy meter, wherein the full-condition strategy information consists of voltage condition rated sequence, current condition range classification and power factor rated sequence. After randomly perturbing the full-condition strategy information using interactive multi-condition fluctuation scales, multiple sets of steady-state test signals are constructed based on orthogonal decomposition. The voltage gradient ascending sequence is used to encapsulate the multiple sets of steady-state test signals to form a full-condition steady-state signal injection sequence, which serves as the full-condition steady-state signal strategy. During the process of injecting the full-condition steady-state signal injection sequence into the bus energy meter by controlling the standard source in discrete time steps, the metering data extraction is performed synchronously, and the first set of bus-level metering values ​​is output.

5. The intelligent calibration method for electricity meters as described in claim 1, characterized in that, In a branch-level electrical isolation scenario, during the dynamic loading of multi-dimensional disturbance load combinations on K branch circuits via a programmable load module, the second set of bus-level metering values ​​of the bus energy meter are simultaneously measured. The method includes: Collect K historical power consumption data for the K branch circuits; Based on the frequency of load feature recurrence of the K historical electricity consumption data, the characteristic load feature vector is extracted to construct a K branch load feature library. The load feature vectors of the K branch load feature libraries are orthogonally decomposed and randomly combined to dynamically output a multidimensional perturbation load combination, wherein the multidimensional perturbation load combination is composed of F multidimensional perturbation load vectors, where F≥50; The programmable load module loads the F multidimensional disturbance load vectors in the K branch circuits in a discrete time sequence, and synchronously collects the metering data of the bus energy meter in the steady state range to obtain the second set of bus-level metering values, wherein the second set of bus-level metering values ​​includes F second bus-level metering values.

6. The intelligent calibration method for an electricity meter as described in claim 5, characterized in that, The method involves orthogonally decomposing and randomly combining the load feature vectors of the K branch load feature libraries to dynamically output a multidimensional perturbation load combination. Based on the recurrence sequence of the K historical electricity consumption data according to the K branch load feature databases, cross-branch load conflict identification is performed as the first load combination restriction; The bus capacity is obtained interactively and used as a second load combination limit; Using the first load combination constraint and the second load combination constraint as constraints, the load feature vectors of the K branch load feature libraries are orthogonally decomposed and randomly combined to dynamically output multidimensional disturbance load combinations.

7. The intelligent calibration method for an electricity meter as described in claim 1, characterized in that, Based on the error distribution characteristics of the first group of bus-level errors, a scenario switching judgment is made. If a switching is determined, the bus energy meter is switched to a branch-level electrical isolation scenario via the switching relay. The method includes: Based on preset correlation indicators, multidimensional features are extracted from the first group of bus-level errors to output bus-level error features, wherein the bus-level error features include bus-level range, bus-level standard deviation, bus-level skewness coefficient and bus-level over-threshold ratio. Based on the bus-level error characteristics, the error distribution pattern is determined, and the distribution pattern category is output. Based on the distribution pattern category matching switching decision rule library, if the decision output is a branch-level isolation command, then drive the switching relay to switch the bus energy meter to the branch-level electrical isolation scenario.

8. The intelligent calibration method for an electricity meter as described in claim 6, characterized in that, The characteristic load feature vector includes the power factor of resistive load, the power factor of capacitive load, the power factor of inductive auxiliary load, and nonlinear load parameters, wherein the power factor of inductive auxiliary load reflects the power factor of inductive load.

9. The intelligent calibration method for an electricity meter as described in claim 1, characterized in that, After calibration is triggered, the bus energy meter is switched to a bus-level electrical isolation scenario by switching relays. Prior to this, the method includes: Locally access the multi-source historical status data of the bus energy meter; After extracting the multi-scale steady-state runtime from the multi-source historical state data, the minimum value is extracted by comparison to generate a timed trigger threshold. The standard deviation of active power in the multi-source historical state data is calculated using a sliding window to generate a first event trigger threshold. Based on the multi-source historical state data, the range ratio of the effective value of the branch circuit current is calculated to generate the second event trigger threshold. The timed trigger threshold, the first event trigger threshold, and the second event trigger threshold are input into the state machine to construct a calibration trigger condition matching engine.

10. An intelligent calibration system for electricity meters, characterized in that, The system is used to perform the smart calibration method for an electricity meter according to any one of claims 1-9, including: Calibration trigger unit: After calibration triggering, the bus energy meter is switched to a bus-level electrical isolation scenario by switching relays; First metering data extraction unit: In a bus-level electrical isolation scenario, after injecting a full-condition steady-state signal strategy into the bus energy meter, it performs metering data extraction and outputs the first set of bus-level metering values; Test error calculation unit: The test error is calculated by comparing the first set of bus-level measurement values ​​with the theoretical values ​​under all operating conditions, and the first set of bus-level errors is output. Scene switching judgment unit: Based on the error distribution characteristics of the first group of bus-level errors, the scene switching judgment is made. If the switching is judged, the bus energy meter is switched to the branch-level electrical isolation scene through the switching relay. Second metering data extraction unit: In a branch-level electrical isolation scenario, the second set of bus-level metering values ​​of the bus energy meter are simultaneously measured by the programmable load module during the dynamic loading of multi-dimensional disturbance load combinations in K branch circuits. Closed-loop calibration unit: performs branch-level energy consumption error tracing calculation on the second group of bus-level metering values, and performs closed-loop calibration on the bus energy meter based on the calculation results.

Images