Electric energy error test method based on intelligent measurement switch

Abstract

The invention relates to the technical field of intelligent measurement switch testing, in particular to an electric energy error testing method based on an intelligent measurement switch, which comprises the following steps of: S1, carding multi-dimensional working conditions of the intelligent measurement switch, forming a working condition classification table, and constructing a testing management platform at the same time; s2, performing synchronization source precision calibration and compensation model construction according to the working condition classification table and the test management platform; when the method is used, errors caused by the environment and aging are conveniently reduced through synchronous source calibration, an environment error compensation model and an aging error compensation model, the problem that an existing method cannot correct systematic deviations is solved, full-range changes of loads, voltages and power factors are adapted by means of a working condition automatic identification and parameter dynamic matching mechanism, and the accuracy of the system is improved. The test adaptability is improved, the double models are synchronously called through the real-time data bus, the problem of time mismatch caused by environment and aging errors is solved, the correction response speed is improved conveniently, and a rapid load fluctuation scene is adapted.

Description

Technical Field

[0001] This invention relates to the field of intelligent measurement switch testing technology, and specifically to a method for testing electrical energy error based on an intelligent measurement switch. Background Technology

[0002] Intelligent measuring switches are electrical devices that combine circuit on / off control with intelligent detection of electrical parameters. Their core function is to control and collect data from circuit systems. Their basic function is to control the circuit on / off of traditional switches to ensure safe circuit operation. With a built-in measuring module, they can collect electrical parameter data such as voltage, current, power, and energy consumption in real time, supporting data storage and transmission. They can also realize remote control and fault alarms, and are suitable for smart grids, industrial control, smart homes, and other scenarios to meet the needs of automated monitoring.

[0003] Patent application number CN202410629809.4 describes in its specification that "this invention provides an FCT testing method, system, and medium based on intelligent measurement switches. The method includes: randomly generating simulated parameters for the current FCT test by a parameter random generation unit; configuring a corresponding simulated test circuit according to the generated simulated parameters for the current FCT test; placing multiple intelligent measurement switches under test in parallel into the simulated test circuit and performing power-on testing; acquiring measured FCT signal data of multiple intelligent measurement switches under test through a data acquisition unit; and processing the measured FCT signal data of multiple intelligent measurement switches under test through a preset FCT evaluation algorithm to obtain the FCT evaluation result of each intelligent measurement switch under test. This invention can effectively avoid the problem of inaccurate FCT testing of intelligent measurement switches caused by problems with the fixed test circuit itself in traditional methods."

[0004] While existing technical solutions have the advantages mentioned above, their disadvantages are as follows: With the rapid development of smart grids, the accuracy of smart metering switches, as key power metering devices, directly affects the fairness of power trading and the economic efficiency of grid operation. Traditional power error testing methods mainly use simple difference comparison methods, which cannot effectively eliminate systematic errors caused by environmental factors and equipment aging. At the same time, fixed test cycles and parameter settings cannot adapt to the test requirements under different operating conditions, resulting in low test efficiency.

[0005] In conclusion, developing a power error testing method based on intelligent measurement switches remains a key issue that urgently needs to be addressed in the field of intelligent measurement switch testing technology. Summary of the Invention

[0006] The purpose of this invention is to address the problems in existing technologies where, with the rapid development of smart grids, intelligent measuring switches, as key power metering devices, directly affect the fairness of power trading and the economic efficiency of grid operation. Traditional power error testing methods mainly employ simple difference comparison methods, which cannot effectively eliminate systematic errors caused by environmental factors and equipment aging. At the same time, fixed testing cycles and parameter settings cannot adapt to testing needs under different operating conditions, resulting in low testing efficiency.

[0007] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a method for testing electrical energy error based on an intelligent measuring switch, comprising the following steps: S1. Organize the multi-dimensional operating conditions of intelligent measurement switches and form an operating condition classification table, and at the same time build a test management platform; S2. Based on the working condition classification table and the test management platform, perform synchronous source accuracy calibration and compensation model construction; S3. The test management platform automatically identifies the current working condition and matches the test parameters, synchronously collects the synchronous source power and switch pulse signal, filters abnormal pulses and preprocesses them to obtain the measured power. S4. Correct the measured electrical energy by the compensation model to obtain the corrected electrical energy, and calculate the error at the same time point to obtain the average error and the maximum error. S5. By generating a test report containing working condition information, compensation comparison, and early warning records, iterate the working condition classification table and compensation model.

[0008] Furthermore, in step S1, the method for sorting out the multi-dimensional operating conditions of the intelligent measurement switch and forming an operating condition classification table, and simultaneously building a test management platform, is as follows: The multi-dimensional operating conditions include load current, voltage level, and power factor, with preset test parameters for each type of operating condition; the test management platform deploys a precision voltage / current synchronization source connected to the input terminal of the intelligent measurement switch; an environmental monitoring module is installed to collect temperature, humidity, and electromagnetic interference intensity at a distance of ≤1m from the intelligent measurement switch; a data acquisition terminal is configured to connect the output terminal of the intelligent measurement switch to the synchronization source; the test management platform integrates operating condition identification, data acquisition, pulse filtering, and error calculation functions, and also supports data interaction with the power dispatching system.

[0009] Further, in step S2, the method for performing synchronous source accuracy calibration and compensation model construction based on the operating condition classification table and the test management platform is as follows: The synchronization source uses a rubidium atomic clock as a time reference, and performs initial calibration of the clock system of the intelligent measurement switch and the synchronization source through the Beidou time synchronization module; the synchronization source is then activated, and the first voltage output time of the intelligent measurement switch is recorded. Calculate the time difference with the time of the first current output. ;exist When the value is greater than 5μs, the synchronization source clock is adjusted using the PID control algorithm of the built-in calibration module until... If the time limit is ≤5μs, and the synchronization source is replaced and the fault is recorded if the time limit is invalid, the compensation model includes an environmental error compensation model and an aging error compensation model. The construction of the environmental error compensation model includes testing the baseline error values ​​under three typical operating conditions in a standard environment. Environmental parameters were adjusted according to a gradient of ±10℃ temperature, ±20%RH humidity, and ±20dBμV electromagnetic interference. The system was then continuously tested for 12 hours under standard conditions, with measurement errors recorded every 10 minutes. ,in, For operating condition number, To determine the number of tests, we introduce... Remove outliers, by temperature ,humidity and electromagnetic interference Adjust environmental parameters and test the corresponding error for each combination of parameters. The temperature The step size is 2°C, with a total of 11 gradients; the humidity... The step size is 5%RH, with a total of 9 gradients; the electromagnetic interference The step size is 5 dBμV, with a total of 9 gradients; the changes in environmental parameters are calculated, and a three-dimensional environment-error database is constructed with a sample size ≥ 11 × 9 × 9 × 3 = 2673 groups; the linear compensation formula of the environmental error compensation model is trained: , In the formula, This represents the error value after compensation. , , The fitting coefficients corresponding to the changes in temperature, humidity, and electromagnetic interference were obtained using the weighted least squares method. This represents the baseline error value obtained under standard testing conditions. This is the temperature value. This represents the humidity change value. This represents the change in electromagnetic interference. , It is the cross term coefficient.

[0010] Furthermore, the method in step S2 also includes: The aging error compensation model includes collecting data on the running time of the intelligent measurement switch, the cumulative number of on / off cycles, and historical calibration records. A full-condition calibration is performed every three months, and the actual error is recorded. Error after environmental compensation during the same period Calculate the aging error offset, expression: , In the formula, and The coefficients are corrected based on the factory parameters of the intelligent measurement switch. For the running time of the intelligent measurement switch, This represents the cumulative number of on / off cycles.

[0011] Furthermore, in step S3, the test management platform automatically identifies the current operating condition and matches test parameters, synchronously acquires the synchronous source power and switching pulse signal, and filters abnormal pulses and preprocesses them to obtain the measured power as follows: The test management platform collects load current, voltage, and power factor in real time and compares them with the operating condition classification table to determine the current operating condition. A weighted Euclidean distance algorithm is used to calculate the similarity between the real-time parameters and the operating condition classification table. The formula is: , In the formula, For the weighting coefficients, satisfying , For reference parameters in the working condition classification table, When a match is found to be successful, the corresponding test parameters are automatically sent to the data acquisition terminal. This represents the maximum similarity across all operating conditions.

[0012] Furthermore, the method in step S3 also includes: Synchronous acquisition of synchronous source output power The power pulse signal is compared with the switching power pulse signal, and the number of pulses N per sampling cycle is counted. Pulse width, amplitude, and rise time characteristics are extracted and compared with standard values. Pulses with a similarity of <80% are considered abnormal. When the proportion of abnormal pulses R > 5%, environmental parameter differences are traced back to determine hardware faults and generate troubleshooting prompts. The preprocessing introduces a correction term for pulse characteristic deviation and uses normal pulses for power calculation, expressed as: , In the formula, The correction coefficients were obtained through least squares fitting. This is the normal pulse count. is the switching pulse constant.

[0013] Further, in step S4, the method for obtaining the corrected electrical energy by correcting the measured electrical energy through the compensation model, and calculating the error at the same time point, and accumulating the average error and the maximum error is as follows: The test management platform synchronously calls the environmental error compensation model and aging error compensation model through a real-time data bus, combined with pre-processed electrical energy. The error cancellation method is used to calculate the corrected electrical energy, and an uncertainty propagation mechanism for the compensation model is introduced. The expression is: , In the formula, Indicates time The electrical energy value after comprehensive correction This indicates the electrical energy value after preprocessing. Indicates at time The environmental error compensation amount is calculated using the environmental error compensation model. Indicates at time The aging error compensation amount is calculated using the aging error compensation model. Indicates at time The combined uncertainty of the compensation model.

[0014] Furthermore, the method in step S4 also includes: Based on the standard timestamp sequence generated by the BeiDou time synchronization module, the error at the same time node is calculated. ,expression: , In the formula, Indicates the first The relative error at each time point Indicates at a time node The electrical energy value of the time synchronization source Indicates at a time node The corrected energy value, based on Weighted average error and maximum error ,expression: , In the formula, This represents the weighted average error. The number of nodes representing valid data. Indicates the first The weight of each time point, This represents the maximum value of synchronous source electrical energy at the same point in time. This represents the maximum value of the relative error at the same point in time.

[0015] Furthermore, the method in step S4 also includes: exist When the value is less than or equal to a preset threshold, the system is considered normal and a report is generated. When the value exceeds the threshold, an audible and visual warning is triggered, and the operating condition, error value, and automatic 10-minute extension of the test for verification are pushed out. An error trend chart is plotted to mark nodes exceeding the threshold. The preset threshold is adaptively adjusted based on industry standards and operating conditions, expressed as: , In the formula, for Total harmonic distortion of current at any given time. The preset threshold representing the average error. It is an operation that takes the minimum value. This represents the preset threshold for the maximum error.

[0016] Furthermore, in step S5, the method for iterating the working condition classification table and the compensation model by generating a test report containing working condition information, compensation comparison, and early warning records is as follows: The system collects the measured power sequence after preprocessing in step S3, the corrected power sequence in step S4, the standard power sequence of the synchronous source, the operating parameters at corresponding time points, environmental monitoring data, and equipment aging parameters, aligns them by timestamp, and forms a structured dataset. It extracts feature vectors from the audible and visual warnings, performs dimensionality reduction through principal component analysis, calculates the principal component contribution weights, integrates the data to generate an intelligent measurement switch power error test report, including operating condition information, compensation comparison, and warning records. Based on the dataset, it uses a Gaussian mixture model to cluster the operating condition information and compensation comparison, and calculates the Mahalanobis distance between the cluster centers and existing operating conditions. When the Mahalanobis distance exceeds a preset distance threshold, the operating condition is added to the operating condition classification table, and the test parameter threshold is updated. It introduces a radial basis function neural network to optimize the environmental error compensation model and optimizes the aging error compensation formula through a multiple linear regression model.

[0017] Beneficial effects Compared with known public technologies, the technical solution provided by this invention has the following beneficial effects: When in use, this invention uses synchronous source calibration, environmental error compensation model, and aging error compensation model to reduce errors caused by environment and aging, which helps to solve the problem that existing methods cannot correct systematic deviations. By using automatic operating condition identification and dynamic parameter matching mechanism, it can adapt to the full range of load, voltage, and power factor changes, which helps to improve test adaptability. By synchronously calling the two models through real-time data bus, it solves the time mismatch problem caused by environmental and aging errors, which facilitates the improvement of correction response speed and adapts to scenarios with rapid load fluctuations. Attached Figure Description

[0018] Figure 1 This is a flowchart of an energy error testing method based on an intelligent measuring switch according to the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0020] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but includes other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0021] The present invention will now be described in further detail with reference to the accompanying drawings: like Figure 1 As shown, this invention provides a method for testing electrical energy error based on an intelligent measuring switch, comprising the following steps: S1, S1, sort out the multi-dimensional working conditions of intelligent measurement switches and form a working condition classification table, and at the same time build a test management platform; S2. Based on the working condition classification table and the test management platform, perform synchronous source accuracy calibration and compensation model construction; S3. The test management platform automatically identifies the current working condition and matches the test parameters, synchronously collects the synchronous source power and switch pulse signal, filters abnormal pulses and preprocesses them to obtain the measured power. S4. Correct the measured electrical energy by the compensation model to obtain the corrected electrical energy, and calculate the error at the same time point to obtain the average error and the maximum error. S5. By generating a test report containing working condition information, compensation comparison, and early warning records, iterate the working condition classification table and compensation model; Furthermore, in step S1, the method for sorting out the multi-dimensional operating conditions of the intelligent measurement switch and forming an operating condition classification table, and simultaneously building a test management platform, is as follows: The multi-dimensional operating conditions include load current, voltage level, and power factor, with preset test parameters for each type of operating condition; the test management platform deploys a precision voltage / current synchronization source connected to the input terminal of the intelligent measurement switch; an environmental monitoring module is installed to collect temperature, humidity, and electromagnetic interference intensity at a distance of ≤1m from the intelligent measurement switch; a data acquisition terminal is configured to connect the output terminal of the intelligent measurement switch to the synchronization source; the test management platform integrates operating condition identification, data acquisition, pulse filtering, and error calculation functions, and also supports data interaction with the power dispatching system; In this embodiment, typical application conditions are categorized according to load current (0-5%In, 5%-100%In, 100%-120%In), voltage level (220V / 380V / 10kV), and power factor (0.5 inductive-1.0 resistive), forming a condition classification table. Preset test parameters for each condition are: sampling frequency of 1Hz for low load conditions, 10Hz for high load conditions, a normal statistical period of 5min, and a sudden change period of 1min. A voltage / current synchronization source with an accuracy of 0.02 is deployed and connected to the input of the intelligent measurement switch, ensuring that the voltage and current synchronization error is ≤1μs. An environmental monitoring module is installed to collect temperature (-20~60℃), humidity (10%-90%RH), and electromagnetic interference intensity (0-100dBμV) at a distance of ≤1m from the switch. A data acquisition terminal supporting RS485 / Ethernet communication and a sampling rate ≥1000 points / s is configured and connected to the switch output and the synchronization source.

[0022] Further, in step S2, the method for performing synchronous source accuracy calibration and compensation model construction based on the operating condition classification table and the test management platform is as follows: The synchronization source uses a rubidium atomic clock as a time reference, and performs initial calibration of the clock system of the intelligent measurement switch and the synchronization source through the Beidou time synchronization module; the synchronization source is then activated, and the first voltage output time of the intelligent measurement switch is recorded. Calculate the time difference with the time of the first current output. ;exist When the value is greater than 5μs, the synchronization source clock is adjusted using the PID control algorithm of the built-in calibration module until... If the time limit is ≤5μs, and the synchronization source is replaced and the fault is recorded if the time limit is invalid, the compensation model includes an environmental error compensation model and an aging error compensation model. The construction of the environmental error compensation model includes testing the baseline error values ​​under three typical operating conditions in a standard environment of 25℃, 50%RH, and no electromagnetic interference. Environmental parameters were adjusted according to a gradient of ±10℃ temperature, ±20%RH humidity, and ±20dBμV electromagnetic interference. The system was then continuously tested for 12 hours under standard conditions, with measurement errors recorded every 10 minutes. ,in, For operating condition number, To determine the number of tests, we introduce... Remove outliers, by temperature ,humidity and electromagnetic interference Adjust environmental parameters and test the corresponding error for each combination of parameters. The temperature The step size is 2°C, with a total of 11 gradients; the humidity... The step size is 5%RH, with a total of 9 gradients; the electromagnetic interference The step size is 5 dBμV, with a total of 9 gradients; the changes in environmental parameters are calculated, and a three-dimensional environment-error database is constructed with a sample size ≥ 11 × 9 × 9 × 3 = 2673 groups; the linear compensation formula of the environmental error compensation model is trained: , In the formula, This represents the error value after compensation. , , The fitting coefficients corresponding to the changes in temperature, humidity, and electromagnetic interference were obtained using the weighted least squares method. This represents the baseline error value obtained under standard testing conditions. This is the temperature value. This represents the humidity change value. This represents the change in electromagnetic interference. , It is the cross term coefficient; In this embodiment, considering the field testing scenario of a 10kV intelligent measuring switch in a substation, a rubidium atomic clock is used as the synchronization source as the time reference, and the frequency stability of the rubidium atomic clock is ≤5× / day, the clock system of the intelligent measurement switch and the synchronization source is initially calibrated to UTC synchronization error ≤10ns through the Beidou time synchronization module; after the synchronization source outputs the rated signal, 300 sets of voltage first output time and current first output time are continuously collected, and the mean and standard deviation of the time difference are calculated. If the mean is >5μs, the clock offset is dynamically adjusted immediately through the PID control algorithm of the built-in calibration module. If the standard is still not met after 3 calibrations, the backup synchronization source is switched and the fault code is recorded. When constructing the environmental error compensation model, firstly, under the standard environment of 25℃, 50%RH, and 30dBμV, continuous testing was conducted for 12 hours on three typical operating conditions: 0.2In / 10kV / 0.9 power factor. Error values ​​were recorded every 10 minutes. After eliminating outliers using the 3σ criterion, the baseline error was determined. Then, the environmental parameters were adjusted according to gradient combinations of temperature 15℃~35℃ (step size 2℃), humidity 30%~70%RH (step size 5%), and electromagnetic interference 10~50dBμV (step size 5dBμV). The corresponding error was tested under each combination of parameters. A total of ≥2673 sets of data were collected to construct a three-dimensional environment-error database. The linear compensation formula of the environmental error compensation model containing cross terms was solved using the weighted least squares method. To address the insufficient synchronization accuracy caused by traditional static calibration and improve stability, the environmental error compensation model introduces cross terms of temperature-humidity and temperature-electromagnetic interference. Combined with multi-gradient parameter testing and weighted least squares fitting, it helps reduce measurement errors caused by environmental interference compared to the single-parameter compensation model.

[0023] Furthermore, the method in step S2 also includes: The aging error compensation model includes collecting data on the running time of the intelligent measurement switch, the cumulative number of on / off cycles, and historical calibration records. A full-condition calibration is performed every three months, and the actual error is recorded. Error after environmental compensation during the same period Calculate the aging error offset, expression: , In the formula, and The coefficients are corrected based on the factory parameters of the intelligent measurement switch. For the running time of the intelligent measurement switch, This represents the cumulative number of on / off cycles. In this embodiment, taking the operation and maintenance test scenario of a 10kV outdoor ring main unit intelligent measurement switch as an example, the switchgear, which is exposed to temperature fluctuations of -20℃ to 60℃ for a long time and is switched on and off 3 to 5 times per day, collects the cumulative number of on / off cycles during operation in real time through the IoT module of the test management platform, while simultaneously retrieving the equipment's historical calibration records for the past 2 years; a full-condition calibration is carried out every 3 months, covering typical operating conditions of 0.1In~1.2In load current, 0.5~1.0 power factor, and 10kV / 0.38kV voltage level, based on calculations using an environmental error compensation model. Based on this, the actual measurement error of the intelligent measurement switch is recorded, and then corrected according to the correction coefficient of the equipment's factory parameters and combined with the on-site operation and maintenance records. The error is then substituted into the formula. This approach helps to overcome the limitations of traditional methods that rely solely on runtime as a single parameter for calculating aging error. By statistically analyzing the cumulative number of on / off cycles and combining this with historical calibration records to correct the error, the accuracy of aging error calculation can be improved.

[0024] Furthermore, in step S3, the test management platform automatically identifies the current operating condition and matches test parameters, synchronously acquires the synchronous source power and switching pulse signal, and filters abnormal pulses and preprocesses them to obtain the measured power as follows: The test management platform collects load current, voltage, and power factor in real time and compares them with the operating condition classification table to determine the current operating condition. A weighted Euclidean distance algorithm is used to calculate the similarity between the real-time parameters and the operating condition classification table. The formula is: , In the formula, For the weighting coefficients, satisfying , For reference parameters in the working condition classification table, When a match is found to be successful, the corresponding test parameters are automatically sent to the data acquisition terminal. This represents the maximum similarity across all operating conditions. In this embodiment, taking the field test scenario of a 10kV distribution substation intelligent measurement switch as an example, the test management platform collects the effective values ​​of load current, effective values ​​of voltage, and power factor in real time through a high-precision sampling module. At the same time, it calls the pre-built operating condition classification table and uses a weighted Euclidean distance algorithm to calculate the similarity between the real-time parameters and the reference parameters of each operating condition. The operating condition is determined and the corresponding test parameters are automatically sent to the acquisition terminal. This approach helps to overcome the limitations of traditional single-parameter threshold comparison. By assigning higher weight to the load current through the weighted Euclidean distance algorithm, the problem of misjudgment of operating conditions caused by small voltage fluctuations or slight deviations in power factor is solved, which is conducive to improving the accuracy of operating condition identification.

[0025] Furthermore, the method in step S3 also includes: Synchronous acquisition of synchronous source output power The power pulse signal is compared with the switching power pulse signal, and the number of pulses N per sampling cycle is counted. Pulse width, amplitude, and rise time characteristics are extracted and compared with standard values. Pulses with a similarity of <80% are considered abnormal. When the proportion of abnormal pulses R > 5%, environmental parameter differences are traced back to determine hardware faults and generate troubleshooting prompts. The preprocessing introduces a correction term for pulse characteristic deviation and uses normal pulses for power calculation, expressed as: , In the formula, The correction coefficients were obtained through least squares fitting. This is the normal pulse count. The switching pulse constant; In this embodiment, taking the daily testing of a 10kV distribution substation intelligent measurement switch as an example, the test management platform synchronously collects the output power pulse signals of the synchronization source and the switch via an optical fiber link. A high-speed counter counts the total number of pulses within the cycle. Simultaneously, the pulse width, amplitude, and rise time of each pulse are extracted. A cosine similarity algorithm is used to calculate the similarity value between the real-time features and the standard features. The number of abnormal pulses within the cycle is calculated to be 520, with an abnormality rate R = 520 / 12480 ≈ 4.16% ≤ 5%, which is considered normal fluctuation. In a subsequent cycle, due to poor contact of the switch pulse module, an abnormality occurs... Random pulses with a duration of 80~150μs and an amplitude of 3~6V, with an abnormality rate of R=8%>5%, indicate a hardware fault. The platform immediately checks environmental parameters, calculates the environmental impact factor Denv=0.3<0.5, and generates a warning: "Potential risk of contact oxidation in the pulse module; it is recommended to check the pin impedance." In the preprocessing stage, the number of normal pulses is first screened out. =12480-520=11960, calculate the measured electrical energy according to the basic formula. =11.96kWh; Then, a pulse characteristic deviation correction term is introduced, and the correction coefficient is fitted using the least squares method. The real-time pulse characteristic mean is, and substituting it into the correction formula yields... =11.96×1.00055≈11.97kWh, the relative error with the synchronous source power of 12.5kWh is reduced from 4.32% before correction to 4.24%; it is convenient to break through the limitation of the traditional method of judging power by pulse counting alone, and introduce multiple features such as pulse width, amplitude and rise time to screen abnormal pulses, which is conducive to reducing the false judgment rate of abnormal pulses and improving anti-interference ability.

[0026] Further, in step S4, the method for obtaining the corrected electrical energy by correcting the measured electrical energy through the compensation model, and calculating the error at the same time point, and accumulating the average error and the maximum error is as follows: The test management platform synchronously calls the environmental error compensation model and aging error compensation model through a real-time data bus, combined with pre-processed electrical energy. The error cancellation method is used to calculate the corrected electrical energy, and an uncertainty propagation mechanism for the compensation model is introduced. The expression is: , In the formula, Indicates time The electrical energy value after comprehensive correction This indicates the electrical energy value after preprocessing. Indicates at time The environmental error compensation amount is calculated using the environmental error compensation model. Indicates at time The aging error compensation amount is calculated using the aging error compensation model. Indicates at time The combined uncertainty of the compensation model; In this embodiment, the dual models are synchronously invoked through a real-time data bus to solve the time mismatch problem caused by environmental and aging errors, which facilitates improved correction response speed and adapts to scenarios with rapid load fluctuations.

[0027] Furthermore, the method in step S4 also includes: Based on the standard timestamp sequence generated by the BeiDou time synchronization module, the error at the same time node is calculated. ,expression: , In the formula, Indicates the first The relative error at each time point Indicates at a time node The electrical energy value of the time synchronization source Indicates at a time node The corrected energy value, based on Weighted average error and maximum error ,expression: , In the formula, This represents the weighted average error. The number of nodes representing valid data. Indicates the first The weight of each time point, This represents the maximum value of synchronous source electrical energy at the same point in time. This represents the maximum value of the relative error at the same point in time. In this embodiment, the BeiDou time synchronization is used to generate nanosecond-level synchronized timestamps, which solves the problem of time node misalignment caused by traditional independent timing. It reduces time alignment error at the edge and ensures the spatiotemporal consistency of error calculation. In scenarios with rapid load fluctuations, it improves the accuracy of error statistics. The weighted average error highlights the impact of high-load nodes through load weights and avoids interference from random errors during low-load periods on the overall assessment. Compared with the arithmetic mean method, it is beneficial to improve the matching degree between error assessment and the importance of actual measurement.

[0028] Furthermore, the method in step S4 also includes: exist When the value is less than or equal to a preset threshold, the system is considered normal and a report is generated. When the value exceeds the threshold, an audible and visual warning is triggered, and the operating condition, error value, and automatic 10-minute extension of the test for verification are pushed out. An error trend chart is plotted to mark nodes exceeding the threshold. The preset threshold is adaptively adjusted based on industry standards and operating conditions, expressed as: , In the formula, for Total harmonic distortion of current at any given time. The preset threshold representing the average error. It is an operation that takes the minimum value. The preset threshold representing the maximum error; In this embodiment, the preset threshold of the average error is adaptively adjusted by the total harmonic distortion rate of the current, which helps to solve the problem of misjudgment due to an overly strict threshold or omission due to an overly lenient threshold under harmonic conditions. This helps to reduce the false alarm rate. The verification mechanism of automatic extended testing and high-frequency sampling avoids invalid operation and maintenance caused by single random errors, which helps to improve the verification accuracy. Compared with manual verification, the operation and maintenance response time is shortened and the efficiency is improved. The error trend graph's over-threshold node labeling and multi-parameter correlation realize the error-operating condition-environment traceability closed loop, which allows operation and maintenance personnel to quickly locate the root cause of the problem and shorten the fault diagnosis time.

[0029] Furthermore, in step S5, the method for iterating the working condition classification table and the compensation model by generating a test report containing working condition information, compensation comparison, and early warning records is as follows: The system collects the preprocessed measured power sequence from step S3, the corrected power sequence from step S4, the standard power sequence from the synchronous source, the operating parameters at corresponding time points, environmental monitoring data, and equipment aging parameters, aligning them by timestamp to form a structured dataset. It extracts feature vectors from the audible and visual warnings, performs dimensionality reduction using principal component analysis, calculates the principal component contribution weights, integrates the data to generate an intelligent measurement switch power error test report, including operating condition information, compensation comparison, and warning records. Based on the dataset, it uses a Gaussian mixture model to cluster the operating condition information and compensation comparison, calculating the Mahalanobis distance between the cluster centers and existing operating conditions. When the Mahalanobis distance exceeds a preset distance threshold, the operating condition is added to the operating condition classification table, and the test parameter threshold is updated. It introduces a radial basis function neural network to optimize the environmental error compensation model and optimizes the aging error compensation formula using a multiple linear regression model. In this embodiment, Gaussian mixture model clustering and Mahalanobis distance judgment are combined to achieve automatic identification of new working conditions and dynamic expansion of the classification table, which solves the lag of traditional manual supplementation of working conditions and improves the working condition coverage.

[0030] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for testing electrical energy error based on an intelligent measuring switch, characterized in that, Includes the following steps: S1. Organize the multi-dimensional operating conditions of intelligent measurement switches and form an operating condition classification table, and at the same time build a test management platform; S2. Based on the working condition classification table and the test management platform, perform synchronous source accuracy calibration and compensation model construction; S3. The test management platform automatically identifies the current working condition and matches the test parameters, synchronously collects the synchronous source power and switch pulse signal, filters abnormal pulses and preprocesses them to obtain the measured power. S4. Correct the measured electrical energy by the compensation model to obtain the corrected electrical energy, and calculate the error at the same time point to obtain the average error and the maximum error. S5. By generating a test report containing working condition information, compensation comparison, and early warning records, iterate the working condition classification table and compensation model.

2. The method for testing electrical energy error based on an intelligent measuring switch according to claim 1, characterized in that, In step S1, the method for sorting out the multi-dimensional operating conditions of the intelligent measurement switch and forming an operating condition classification table, and simultaneously building a test management platform, is as follows: The multi-dimensional operating conditions include load current, voltage level, and power factor, with preset test parameters for each type of operating condition; the test management platform deploys a precision voltage / current synchronization source connected to the input terminal of the intelligent measurement switch; an environmental monitoring module is installed to collect temperature, humidity, and electromagnetic interference intensity at a distance of ≤1m from the intelligent measurement switch; a data acquisition terminal is configured to connect the output terminal of the intelligent measurement switch to the synchronization source; the test management platform integrates operating condition identification, data acquisition, pulse filtering, and error calculation functions, and also supports data interaction with the power dispatching system.

3. The method for testing electrical energy error based on an intelligent measuring switch according to claim 2, characterized in that, In step S2, the method for performing synchronous source accuracy calibration and compensation model construction based on the operating condition classification table and the test management platform is as follows: The synchronization source uses a rubidium atomic clock as a time reference, and the clock system of the intelligent measurement switch and the synchronization source is initially calibrated through the Beidou time synchronization module. The synchronization source is activated, and the first voltage output time of the intelligent measuring switch is recorded. Calculate the time difference with the time of the first current output. ;exist When the value is greater than 5μs, the synchronization source clock is adjusted using the PID control algorithm of the built-in calibration module until... If the time limit is ≤5μs, and the synchronization source is replaced and the fault is recorded if the time limit is invalid, the compensation model includes an environmental error compensation model and an aging error compensation model. The construction of the environmental error compensation model includes testing the baseline error values ​​under three typical operating conditions in a standard environment. Environmental parameters were adjusted according to a gradient of ±10℃ temperature, ±20%RH humidity, and ±20dBμV electromagnetic interference. The system was then continuously tested for 12 hours under standard conditions, with measurement errors recorded every 10 minutes. ,in, For operating condition number, To determine the number of tests, we introduce... Remove outliers, by temperature ,humidity and electromagnetic interference Adjust environmental parameters and test the corresponding error for each combination of parameters. The temperature The step size is 2°C, with a total of 11 gradients; the humidity... The step size is 5%RH, with a total of 9 gradients; the electromagnetic interference The step size is 5 dBμV, with a total of 9 gradients; the changes in environmental parameters are calculated, and a three-dimensional environment-error database is constructed with a sample size ≥ 11 × 9 × 9 × 3 = 2673 groups; the linear compensation formula of the environmental error compensation model is trained: ， In the formula, This represents the error value after compensation. , , The fitting coefficients corresponding to the changes in temperature, humidity, and electromagnetic interference were obtained using the weighted least squares method. This represents the baseline error value obtained under standard testing conditions. This is the temperature value. This represents the humidity change value. This represents the change in electromagnetic interference. , It is the cross term coefficient.

4. The method for testing electrical energy error based on an intelligent measuring switch according to claim 3, characterized in that, The method in step S2 also includes: The aging error compensation model includes collecting data on the running time of the intelligent measurement switch, the cumulative number of on / off cycles, and historical calibration records. A full-condition calibration is performed every 3 months, and the actual error is recorded. Error after environmental compensation during the same period Calculate the aging error offset, expression: ， In the formula, and The coefficients are corrected based on the factory parameters of the intelligent measurement switch. For the running time of the intelligent measurement switch, This represents the cumulative number of on / off cycles.

5. The method for testing electrical energy error based on an intelligent measuring switch according to claim 4, characterized in that, In step S3, the test management platform automatically identifies the current operating condition and matches the test parameters, synchronously acquires the synchronous source power and switch pulse signal, and filters abnormal pulses and preprocesses them to obtain the measured power. The test management platform collects load current, voltage, and power factor in real time and compares them with the operating condition classification table to determine the current operating condition. A weighted Euclidean distance algorithm is used to calculate the similarity between the real-time parameters and the operating condition classification table. The formula is: , In the formula, For the weighting coefficients, satisfying , For reference parameters in the working condition classification table, When a match is found to be successful, the corresponding test parameters are automatically sent to the data acquisition terminal. This represents the maximum similarity across all operating conditions.

6. The method for testing electrical energy error based on an intelligent measuring switch according to claim 5, characterized in that, The method in step S3 also includes: Synchronous acquisition of synchronous source output power The power pulse signal is compared with the switching power pulse signal, and the number of pulses N per sampling cycle is counted. Pulse width, amplitude, and rise time characteristics are extracted and compared with standard values. Pulses with a similarity of <80% are considered abnormal. When the proportion of abnormal pulses R > 5%, environmental parameter differences are traced back to determine hardware faults and generate troubleshooting prompts. The preprocessing introduces a correction term for pulse characteristic deviation and uses normal pulses for power calculation, expressed as: ， In the formula, The correction coefficients were obtained through least squares fitting. This is the normal pulse count. is the switching pulse constant.

7. The method for testing electrical energy error based on an intelligent measuring switch according to claim 6, characterized in that, In step S4, the corrected electrical energy is obtained by correcting the measured electrical energy through the compensation model, and the error at the same time point is calculated. The method for accumulating the average error and the maximum error is as follows: The test management platform synchronously calls the environmental error compensation model and aging error compensation model through a real-time data bus, combined with pre-processed electrical energy. The error cancellation method is used to calculate the corrected electrical energy, and an uncertainty propagation mechanism for the compensation model is introduced. The expression is: ， In the formula, Indicates time The electrical energy value after comprehensive correction This indicates the electrical energy value after preprocessing. Indicates at time The environmental error compensation amount is calculated using the environmental error compensation model. Indicates at time The aging error compensation amount is calculated using the aging error compensation model. Indicates at time The combined uncertainty of the compensation model.

8. The method for testing electrical energy error based on an intelligent measuring switch according to claim 7, characterized in that, The method in step S4 also includes: Based on the standard timestamp sequence generated by the BeiDou time synchronization module, the error at the same time node is calculated. ,expression: ， In the formula, Indicates the first The relative error at each time point Indicates at a time node The electrical energy value of the time synchronization source Indicates at a time node The corrected energy value, based on Weighted average error and maximum error ,expression: ， In the formula, This represents the weighted average error. The number of nodes representing valid data. Indicates the first The weight of each time point, This represents the maximum value of synchronous source electrical energy at the same point in time. This represents the maximum value of the relative error at the same point in time.

9. The method for testing electrical energy error based on an intelligent measuring switch according to claim 8, characterized in that, The method in step S4 also includes: exist When the value is less than or equal to a preset threshold, the system is considered normal and a report is generated. When the value exceeds the threshold, an audible and visual warning is triggered, and the operating condition, error value, and automatic 10-minute extension of the test for verification are pushed out. An error trend chart is plotted to mark nodes exceeding the threshold. The preset threshold is adaptively adjusted based on industry standards and operating conditions, expressed as: ， In the formula, for Total harmonic distortion of current at any given time. The preset threshold representing the average error. It is an operation that takes the minimum value. This represents the preset threshold for the maximum error.

10. The method for testing electrical energy error based on an intelligent measuring switch according to claim 8, characterized in that, In step S5, the method for iterating the working condition classification table and the compensation model by generating a test report containing working condition information, compensation comparison, and early warning records is as follows: The measured power sequence after preprocessing in step S3, the power sequence after correction in step S4, the standard power sequence of the synchronous source, the operating parameters at the corresponding time nodes, the environmental monitoring data and the equipment aging parameters are collected, aligned by timestamps, and a structured dataset is formed. The feature vector of the sound and light warning is extracted, the dimensionality is reduced by principal component analysis, the contribution weight of the principal components is calculated, and the data is integrated to generate a smart measurement switch power error test report, which includes operating condition information, compensation comparison and warning records. Based on the dataset, a Gaussian mixture model is used to cluster the working condition information and compensation comparison, and the Mahalanobis distance between the cluster center and the existing working condition is calculated. When the Mahalanobis distance exceeds the preset distance threshold, the working condition is added to the working condition classification table and the test parameter threshold is updated. A radial basis function neural network is introduced to optimize the environmental error compensation model, and the aging error compensation formula is optimized through a multiple linear regression model.

Images