Error correction method and system for electric energy metering box

Abstract

The present application relates to the technical field of error correction, in particular to an error correction method and system for an electric energy metering box. The method acquires temperature data, current data, power factor and original electric quantity increment of the electric energy metering box in real time; determines the current ratio difference and angle difference in combination with the error characteristic table of the mutual inductor, acquires the power relative deviation degree based on the power factor, comprehensively obtains the total error degree, corrects the original electric quantity increment to obtain the first corrected electric quantity increment; constructs a temperature-current two-dimensional grid to locate the current target grid, acquires the corrected aging rate based on the difference of the time change of the first corrected electric quantity increment relative to the preset neighborhood grid, corrects the second corrected electric quantity increment in combination with the cumulative running time; filters and smooths the second corrected electric quantity increment to obtain the final corrected electric quantity increment, and acquires the final electric quantity in combination with the historical final cumulative electric quantity. The present application realizes dynamic decoupling and step-by-step stripping of multi-source errors, and significantly improves the long-term accuracy of electric energy metering.

Description

Technical Field

[0001] This invention relates to the field of error correction technology, specifically to an error correction method and system for electricity metering boxes. Background Technology

[0002] As the core equipment for end-point billing and load monitoring in the power system, the accuracy of electricity metering boxes directly affects the fairness of power grid settlement and the reliability of electricity consumption data. However, in actual operation, the metering accuracy of electricity metering boxes is affected by the superposition of various complex environmental and operating conditions, resulting in continuous metering deviations. If these deviations are not corrected in a timely and effective manner, small errors will be amplified over a long period of accumulated electricity consumption, eventually leading to settlement data that deviates significantly from the user's actual consumption. Currently, the sources of errors in electricity metering boxes are mainly divided into three categories: the first category is transient metering deviations of current transformers directly caused by clearly measurable physical quantities (such as drastic changes in ambient temperature, large fluctuations in current load, and shifts in power factor); the second category is the slow and irreversible aging drift of internal electronic components due to continuous heating and physical characteristic decay during long-term operation, and this aging rate varies significantly under different operating conditions (such as high temperature heavy load and low temperature light load); the third category is random errors caused by transient disturbances in the power grid, random load jumps, or sensor measurement noise.

[0003] Existing error correction technologies for electricity metering boxes typically rely on static compensation based on factory calibration parameters or simple linear correction using a single temperature sensor. These methods struggle to comprehensively cover and separate complex and intertwined error sources. In particular, when faced with non-uniform aging drift under different operating conditions and instantaneous random disturbances that cannot be directly quantified, the lack of effective dynamic tracking and adaptive stripping mechanisms leads to some real electricity consumption increments being incorrectly filtered out as errors, or random noise being misjudged as long-term drift. Ultimately, this results in a very unsatisfactory overall correction effect of electricity metering boxes in long-term complex operating environments. Summary of the Invention

[0004] To address the technical problem that existing error correction technologies for electricity metering boxes cannot fully decouple and adaptively separate multi-source metering deviations caused by complex and intertwined factors, resulting in unsatisfactory long-term comprehensive correction effects, the present invention aims to provide an error correction method and system for electricity metering boxes. The specific technical solution adopted is as follows: In a first aspect, one embodiment of the present invention provides an error correction method for an electricity metering box, the method comprising the following steps: Real-time acquisition of temperature data, current data, power factor, and raw power increment of the power metering box; Based on the current temperature and current data, and in conjunction with the transformer error characteristic table, determine the current ratio error and angle error; based on the current power factor and angle error, obtain the degree of relative power deviation; based on the current ratio error and degree of relative power deviation, obtain the total error degree, correct the current original power increment, and determine the current first corrected power increment; Construct a temperature-current two-dimensional grid and locate the current target grid. Based on the difference between the change of the first corrected power increment of the current target grid and the preset neighboring grid within a specified time period, obtain the corrected aging rate corresponding to the current target grid. Combine the cumulative running time of the current target grid to correct the current first corrected power increment and obtain the current second corrected power increment. The second corrected power increment within the current time period is filtered and smoothed to obtain the current final corrected power increment. Combined with the final accumulated power at the previous moment, the current final power is obtained.

[0005] Furthermore, the method for obtaining the degree of relative power deviation is as follows: Calculate the inverse cosine of the current power factor to obtain the true phase angle corresponding to the current power factor, and add the true phase angle to the current angle difference to obtain the offset phase angle; Subtract the cosine of the offset phase angle from the current power factor to obtain the deviation numerator; Take the maximum value between the absolute value of the current power factor and the preset minimum positive value as the denominator of the deviation; Divide the deviation numerator by the deviation denominator to obtain the relative power deviation.

[0006] Furthermore, the method for obtaining the total error level is as follows: The sum of the current ratio difference and the relative power deviation is taken as the current total error level.

[0007] Furthermore, the method for obtaining the first corrected power increment is as follows: Subtract the total error from the preset benchmark value to obtain the first correction coefficient; The product of the current original power increment and the first correction coefficient is used as the current first corrected power increment.

[0008] Furthermore, the method for constructing a temperature-current two-dimensional grid and locating the current target grid is as follows: Based on preset temperature intervals and preset current intervals, the operating space of the power metering box is divided into multiple two-dimensional grids. The two-dimensional grid into which the current temperature and current data fall is used as the current target grid.

[0009] Furthermore, the method for obtaining the corrected aging rate is as follows: Based on the trend of the first corrected power increment of the current target grid over a specified time period, the initial aging rate of the current target grid is obtained by linear regression. The average of the initial aging rates of the preset neighboring grids of the current target grid is used as the reference aging rate. The absolute value of the difference between the initial aging rate corresponding to the current target mesh and the reference aging rate is used as the relative numerator; The absolute value of the reference aging rate and the maximum value among the preset minimum change values ​​are used as the relative denominator; The ratio of the relative numerator to the relative denominator is used as the relative deviation degree of the current target grid. When the relative deviation is less than the preset relative deviation threshold, the initial aging rate corresponding to the current target mesh is used as the corrected aging rate corresponding to the current target mesh. When the relative deviation is greater than or equal to the preset relative deviation threshold, the initial aging rate and the reference aging rate corresponding to the current target mesh are weighted and summed to obtain the corrected aging rate corresponding to the current target mesh.

[0010] Furthermore, the method for obtaining the second corrected power increment is as follows: Multiply the correction aging rate corresponding to the current target mesh by the cumulative running time of the current target mesh to obtain the cumulative aging amount of the current target mesh; if the absolute value of the cumulative aging amount is greater than the preset aging upper limit, the cumulative aging amount is forcibly truncated to the preset aging upper limit with the same sign as the original. Subtract the accumulated aging amount from the current first corrected power increment to obtain the current second corrected power increment.

[0011] Furthermore, the method for obtaining the final power level is as follows: Arrange the second corrected power increments within the current time period in chronological order to obtain the power increment sequence; Median filtering is performed on the power increment sequence to obtain the current smoothed increment estimate; The absolute values ​​of the differences between the last specified number of second-corrected power increments in the power increment sequence and the smoothed increment estimate are all used as jump analysis values. If all jump analysis values ​​are greater than the preset load jump threshold, the current second corrected power increment is taken as the current final corrected power increment; if there is a jump analysis value less than or equal to the preset load jump threshold, the smoothed increment estimate is taken as the current final corrected power increment. The current final power increment is added to the previous final cumulative power to obtain the current final power.

[0012] Furthermore, the method for obtaining the ratio difference and angle difference is as follows: Retrieve the current temperature and current data from the current transformer error characteristic table. If a matching node exists, directly read the corresponding ratio error and angle error to determine the current ratio error and angle error. If no matching node exists, extract the data from the four adjacent operating condition nodes surrounding the current temperature and current data, and calculate the current ratio difference and angle difference using bilinear interpolation.

[0013] Secondly, another embodiment of the present invention provides an error correction system for an electricity metering box, the system comprising: a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, it implements the steps of any of the above methods.

[0014] The present invention has the following beneficial effects: This invention first determines the current ratio error and phase angle error based on current temperature and current data, combined with the transformer error characteristic table, to quantify the inherent measurement deviation of the transformer under the current environment and load. To accurately assess the impact of phase angle amplification under low power factor, the relative power deviation is obtained based on the current power factor and phase angle error, determining the relative active power error caused by phase asynchrony. Then, based on the current ratio error and relative power deviation, the total error is obtained, accurately reflecting the current comprehensive transient metering deviation ratio, thereby correcting the current original power increment and determining the current first corrected power increment, accurately eliminating deterministic metering errors caused by measurable transient operating conditions such as temperature, current, and power factor. To avoid mutual interference from non-uniform aging characteristics under different operating conditions, a temperature-current two-dimensional grid is constructed and the current target grid is located, achieving isolation and independent tracking of aging states in a multi-dimensional operating space. Finally, based on the first corrected power increment of the current target grid within a specified time period... The difference between the changes and the preset neighboring grid is used to obtain the corrected aging rate corresponding to the current target grid. This accurately reflects the long-term physical aging trend of the underlying components after removing abnormal noise under this specific operating condition. Then, combined with the cumulative running time of the current target grid, the total metering error offset generated by historical accumulation under this operating condition is determined. This corrects the current first corrected power increment and obtains the current second corrected power increment, accurately removing the unmeasurable long-term trend error caused by the slow aging of internal components. In order to restore the true power work value, the second corrected power increment in the current time period is filtered and smoothed to remove the random errors caused by high-frequency fluctuations in the power grid or sensor measurement noise, obtaining the current final corrected power increment. This achieves the safe reconstruction of the purest power increment. Then, combined with the final cumulative power at the previous moment, the current final power is obtained. Data convergence is completed on the integral dimension, ensuring the absolute accuracy and reliability of the settlement data of the equipment throughout its entire life cycle and under complex operating conditions. Attached Figure Description

[0015] Figure 1 A schematic flowchart illustrating an error correction method for an electricity metering box according to an embodiment of the present invention; Figure 2 This is a structural diagram of an error correction system for an electricity metering box provided in one embodiment of the present invention. Detailed Implementation

[0016] The following description, in conjunction with the accompanying drawings, details the specific scheme of the error correction method and system for the electricity metering box provided by this invention.

[0017] Example 1:

[0018] This invention proposes an error correction method for electricity metering boxes. Please refer to [link / reference]. Figure 1The diagram illustrates a schematic flowchart of an error correction method for an electricity metering box according to an embodiment of the present invention. The method includes the following steps: Step S1: Real-time acquisition of temperature data, current data, power factor, and raw power increment of the power metering box.

[0019] Specifically, this embodiment uses an energy metering box as an example for analysis, and the energy metering box mentioned below will refer to this energy metering box. In order to comprehensively capture the environmental and load conditions that affect the metering accuracy, the temperature data of the energy metering box is obtained in real time by a temperature sensor built into the energy metering box; the current data of the energy metering box is obtained in real time by a sampling circuit on the secondary side of the current transformer; the power factor of the energy metering box is obtained in real time by calculating the phase difference between the voltage and current waveforms by the energy metering chip, which is a real-time representation of the proportion of actual work done by electrical energy under the load; and the original energy increment of the energy metering box, that is, the initial metering value without any error characteristic compensation, is obtained in real time by reading the active power integration register of the energy metering chip in the current sampling period. It should be noted that, in order to prevent the operating conditions from being misaligned during error compensation due to the time difference in the acquisition of different physical quantities, this embodiment synchronously acquires the temperature data, current data, power factor and original power increment of the power metering box, and sets the data acquisition frequency to 1 time / second to ensure that the transient fluctuations of the power grid load can be accurately captured and the error status updated in real time. The implementer can set the data acquisition frequency according to the computing power of the internal chip and the actual accuracy requirements on site, which is not limited here.

[0020] Step S2: Based on the current temperature and current data, and in conjunction with the transformer error characteristic table, determine the current ratio error and angle error; based on the current power factor and angle error, obtain the degree of relative power deviation; based on the current ratio error and degree of relative power deviation, obtain the degree of total error, correct the current original power increment, and determine the current first corrected power increment.

[0021] Specifically, in order to quantify the inherent measurement deviation of the current transformer under different environments and loads, this embodiment constructs a current transformer error characteristic table by acquiring the current transformer's factory type test data or offline calibration data, and records the ratio difference and phase difference under different combinations of discrete calibration temperature nodes and calibration current nodes. Among them, the ratio difference characterizes the relative amplitude deviation between the actual transformation ratio and the rated transformation ratio of the current transformer, reflecting the reading error caused by the inaccuracy of the transformation ratio; the phase difference characterizes the phase shift of the secondary current relative to the primary current of the current transformer, reflecting the work calculation error caused by phase asynchrony. To accurately obtain the error parameters under the current operating condition, and then determine the current ratio error and angle error based on the current temperature and current data and the transformer error characteristic table, the following steps are taken: First, the current temperature and current data are retrieved from the transformer error characteristic table. If a matching node exists, the corresponding ratio error and angle error are directly read to determine the current ratio error and angle error. If no matching node exists, it is first determined whether the current temperature or current data exceeds the coordinate boundary of the transformer error characteristic table. If it does, it is forcibly clamped to the boundary extreme value to prevent the interpolation algorithm from crashing. When extracting adjacent operating condition node data, if adjacent nodes greater than (or less than) the extreme value cannot be obtained due to boundary constraints, the bidirectional enclosing requirement for that dimension is cancelled, and the extreme value node data at the boundary line or endpoint is directly called for one-dimensional linear interpolation or endpoint extreme value replacement. If the boundary extreme value is not reached, the enclosing node data is extracted. The current temperature and current data are obtained from four adjacent operating condition nodes (combined nodes of adjacent calibration temperature data and adjacent calibration current data that are lower than the current temperature data, adjacent calibration temperature data and adjacent calibration current data that are higher than the current temperature data, adjacent calibration temperature data and adjacent calibration current data that are higher than the current temperature data, and adjacent calibration current data that are lower than the current temperature data, and adjacent calibration current data that are higher than the current temperature data). The current ratio difference and angle difference are calculated using bilinear interpolation (specifically, two linear interpolations are performed along the temperature direction to obtain the intermediate error values ​​corresponding to the two boundary temperatures, and then a linear interpolation is performed along the current direction on these intermediate error values ​​to finally obtain the ratio difference and angle difference corresponding to the current temperature and current data). The bilinear interpolation algorithm is a well-known technique and will not be described in detail here.

[0022] Considering that phase difference can cause the secondary current of the transformer to shift relative to the primary current, especially under low power factor, a small phase difference can be amplified and seriously affect the calculation of active power. In order to accurately assess the impact of the current phase shift on the final work measurement, and then obtain the degree of relative power deviation based on the current power factor and phase difference, it can reflect the proportion of relative measurement error caused by the current phase difference to active power. The current ratio difference indicates the relative amplitude deviation between the actual transformer ratio and the rated ratio. To comprehensively evaluate the overall impact of measurable physical quantities on the electricity reading, the total error level is obtained based on the current ratio difference and the relative power deviation. This is achieved by adding the current ratio difference and the relative power deviation as the current total error level. The sign of the total error level indicates the direction of the deviation of the reading relative to the actual electricity consumption; a positive number indicates that the current meter reading is too high, and a negative number indicates that the current meter reading is too low. Based on this, the current comprehensive transient metering deviation ratio is determined. To restore the true electricity consumption, the current original electricity increment is corrected using the total error level to determine the current first corrected electricity increment, eliminating the deterministic metering errors caused by measurable transient conditions such as temperature data, current data, and power factor.

[0023] Preferably, in one feasible embodiment, the method for obtaining the degree of relative power deviation is as follows: by taking the inverse cosine of the current power factor, the true phase angle corresponding to the current power factor is obtained, while ensuring that the true phase angle and the current angle difference are both in radians. The sum of the true phase angle and the current angle difference is used as the offset phase angle, which characterizes the distorted phase state actually participating in the active power calculation after being affected by the transformer angle difference. In order to quantify the absolute power loss caused by the phase deviation, the cosine value of the offset phase angle is subtracted from the current power factor, which is used as the deviation numerator to reflect the absolute difference between the ideal work and the actual metered work. In order to convert the absolute loss into a relative deviation ratio... To prevent calculation crashes or zero denominator loopholes caused by power factor approaching zero or reverse power supply from the grid leading to four-quadrant metering, the algorithm takes the maximum value between the absolute value of the current power factor and a preset minimum positive value as the deviation denominator. The absolute value is used to ensure that the denominator is always positive, and the maximum value is used to set a safety lower limit. In this embodiment, the preset minimum positive value is set to 0.005 to ensure that the algorithm can still output effective values ​​stably under extremely low power factor conditions. The implementer can set the preset minimum positive value according to the floating-point arithmetic accuracy requirements of the system, which is not limited here. Then, the deviation numerator is divided by the deviation denominator to obtain the relative power deviation degree and accurately extract the relative error of the active power caused by the angle difference.

[0024] Preferably, in one feasible embodiment, the method for obtaining the first corrected power increment is as follows: In order to convert the error ratio into a correction coefficient to accurately eliminate the transient measurement deviation caused by the measurable physical quantity, the total error degree is subtracted from the preset benchmark value to obtain the first correction coefficient, and the actual effective proportion of the real power in the original measured power is determined; in this embodiment, the preset benchmark value is set to 1 to satisfy the physical logic of using the original measured value as the whole unit and deducting the error ratio to restore the real value. The implementer can set the preset benchmark value according to the data normalization format, which is not limited here; in order to determine the pure power increment after removing the influence of the above-mentioned measurable transient operating conditions, the product of the current original power increment and the first correction coefficient is taken as the current first corrected power increment.

[0025] Step S3: Construct a temperature-current two-dimensional grid and locate the current target grid. Based on the difference between the change of the first corrected power increment of the current target grid within a specified time period and the preset neighboring grid, obtain the corrected aging rate corresponding to the current target grid. Combine the cumulative running time of the current target grid to correct the current first corrected power increment and obtain the current second corrected power increment.

[0026] Specifically, considering that metering drift caused by the aging of internal components in the electricity metering box is a non-uniform accumulation process depending on the operating conditions (for example, the aging rate in high-temperature heavy-load areas is much faster than in low-temperature light-load areas), in order to avoid mutual interference of aging characteristics under different operating conditions, a temperature-current two-dimensional grid is constructed and the current target grid is located, which facilitates the isolation and independent tracking of the aging state in a multi-dimensional operating condition space. In order to extract the aging trend under the current operating conditions from historical data, and then analyze the change of the first correction power increment of the current target grid within a specified time period, the specific calculation of this change occurs during regular operation and maintenance, and is obtained by inserting a high-precision external standard reference meter into the loop to obtain the ratio. The model is used to calculate the data, while during daily operation, the latest corrected aging rate saved by the grid point is directly called to evaluate the aging amount and determine the aging characteristics of the current target grid. Considering that the aging characteristics of the original analysis may be mixed with long-term fluctuations or occasional noise interference from the actual power consumption mode, in order to ensure that the calculated aging characteristics conform to the objective law that the aging of physical devices should smoothly transition between similar operating conditions, the changes corresponding to the current target grid are compared with the changes corresponding to the preset neighboring grids. The analysis is conducted to see if the aging characteristics of the current grid have undergone abnormal changes and to determine the abnormal smoothing intervention strategy, thereby obtaining the corrected aging rate corresponding to the current target grid. Considering that the aging rate only represents the drift per unit time, and combining this with the cumulative operating time of the current target grid, the total offset caused by historical accumulation under the current operating conditions is determined. This is used to correct the current first corrected power increment and obtain the current second corrected power increment, further eliminating the unmeasurable long-term trend error caused by the aging of internal components. It should be noted that this embodiment sets the specified time period to cover the span of 30 valid historical sampling points (i.e., the historical power increment and corresponding timestamp records falling under the current target grid operating conditions during the historical operation of the power metering box), and the time interval between adjacent historical sampling points is not less than 24 hours. This ensures that the extracted changes truly reflect the long-term physical aging trend rather than short-term load fluctuations. The implementer can set the specified time period according to the data storage capacity, but the end time of the specified time period is always the current time. If there are no 30 valid historical sampling points before the current time, the aging correction calculation for the target grid is not performed temporarily; that is, the corrected aging rate of the target grid is assumed to be zero until the accumulated valid historical data is used. The number of sampling points is increased until the requirement is met to prevent the aging features from diverging due to insufficient data samples. The preset neighborhood grid is set to the eight adjacent grids (i.e., the perimeter of the nine-square grid) of the current target grid in two-dimensional space. The implementer can set the preset neighborhood grid according to the granularity of the grid division, which is not limited here. If the current target grid is a boundary grid or a corner grid and its number of adjacent grids is less than eight, only the effective adjacent grids that actually exist in the two-dimensional working condition space are selected as the preset neighborhood grid. In addition, the cumulative running time of the current target grid refers to the total effective physical record time span recorded within the specific working condition grid since the equipment was put into operation, which can be directly called from the historical accumulation cache module of the main control microprocessor.

[0027] Preferably, in one feasible embodiment of this method, the method for constructing a temperature-current two-dimensional grid and locating the current target grid is as follows: First, based on a preset temperature interval and a preset current interval, the operating condition space of the energy metering box is divided into multiple two-dimensional grids; then, the two-dimensional grid into which the current temperature data and current data fall is taken as the current target grid. In this embodiment, the preset temperature interval is set to 5°C to ensure that the operating condition division has sufficient temperature sensitivity while avoiding excessive grid fragmentation; the preset current interval is set to 10% of the rated current (i.e., the maximum allowable continuous operating current marked on the nameplate of the energy metering box) to ensure that the load characteristics are accurately isolated; the implementer can set the preset temperature interval and preset current interval according to the thermal characteristics of the transformer and the computing performance of the processor, which is not limited here.

[0028] Preferably, in one feasible method of this embodiment, the method for obtaining the corrected aging rate is as follows: First, based on the trend of the first corrected power increment of the current target grid over a specified time period, the initial aging rate corresponding to the current target grid is obtained through linear regression. That is, during regular maintenance, the cumulative value of the first corrected power increment of the power metering box under the corresponding operating condition is obtained, and the cumulative power of the external high-precision standard reference meter in the same time period is obtained simultaneously. The two are subtracted to obtain the power deviation value. The power deviation value of different historical sampling points is used as the dependent variable, and the corresponding timestamp is used as the independent variable. The linear drift slope of the power deviation value with the running time is calculated. To prevent the denominator variance of the regression slope calculation from being zero due to the small time difference of the sampling points, a preset minimum positive number (e.g., 0.001, which is not limited here) is added to the time variance term of the regression formula. The larger the absolute value of the initial aging rate, the faster the performance of the internal components degrades under the current operating condition, resulting in a larger metering deviation. The linear regression processing method is well-known and will not be elaborated further. Considering that the aging of physical devices should exhibit a smooth transition characteristic under similar operating conditions, in order to construct a local benchmark for identifying abnormal noise, the average initial aging rate corresponding to the preset neighboring grids of the current target grid is used as the reference aging rate. It should be noted that when calculating the average initial aging rate corresponding to the preset neighboring grids, if a preset neighboring grid does not currently meet the requirement of 30 historical sampling points (i.e., it fails to effectively extract the initial aging rate), then in the averaging calculation, the preset neighboring grid is removed as a null value and is not included in the total number of valid grids. Only the preset neighboring grids with valid initial aging rates are summed and averaged to avoid invalid data diluting the true reference benchmark. The absolute value of the difference between the initial aging rate corresponding to the current target grid and the reference aging rate is used as the relative numerator. The absolute value of the reference aging rate and the maximum value among the preset minimum change values ​​are used as the relative denominator. In this embodiment, the preset minimum change value is set to be... To prevent calculation crashes when the reference aging rate approaches zero in the early stages of aging or under normal conditions, implementers can set a preset minimum change value based on the floating-point arithmetic precision of the system; no limit is imposed here. The ratio of the relative numerator to the relative denominator is used as the relative deviation of the current target grid, accurately reflecting the severity of the deviation of the local aging rate of the current target grid from the overall smooth trend. The greater the relative deviation, the more severe the interference of abnormal noise or specific power consumption behavior on the initial aging rate calculated by the current target grid, and the lower the reliability. To distinguish between real physical operating condition differences and abnormal noise interference, this embodiment sets a preset relative deviation threshold of 0.5. This ensures that algorithmic intervention only occurs when local offsets significantly violate physical smoothness rules. Implementers can set the preset relative deviation threshold according to the complexity of the equipment deployment environment; no limitation is imposed here. When the relative deviation is less than the preset relative deviation threshold, it indicates that the initial aging rate calculated by the current target grid conforms to the physical smoothness rules between operating conditions, and the data is reliable. In this case, the initial aging rate corresponding to the current target grid is used as the corrected aging rate corresponding to the current target grid. When the relative deviation is greater than or equal to the preset relative deviation threshold, it indicates that the initial aging rate calculated by the current target grid is subject to abnormal interference. To retain some real local physical differences while filtering out abnormal noise, the initial aging rate corresponding to the current target grid and the reference aging rate are weighted and summed, i.e., by... , to obtain the corrected aging rate corresponding to the current target mesh, where S is the initial aging rate corresponding to the current target mesh; For reference aging rate; To pre-set the reference weight, this embodiment sets it to 0.7. This ensures that while suppressing transient noise, it still provides a high degree of confidence in the true local aging characteristics of the independent mesh, preventing smoothing from completely erasing local operating condition characteristics. Implementers can set this weight based on the noise distribution level of historical data. No restrictions are imposed here.

[0029] Preferably, in one feasible embodiment, the method for obtaining the second corrected power increment is as follows: Considering that the corrected aging rate only represents the attenuation ratio per unit operating time, in order to restore it to the actual offset value to be compensated, after strictly unifying the time unit of the corrected aging rate and the time unit of the current target grid's cumulative operating time, the corrected aging rate corresponding to the current target grid is multiplied by the cumulative operating time of the current target grid to obtain the cumulative aging amount of the current target grid, and the total metering error value accumulated since the equipment was put into operation under the current operating conditions is determined; considering that extreme data noise or regression model divergence may lead to an abnormally high calculated cumulative aging amount, if the cumulative aging amount is too high... If the absolute value of the aging amount is greater than the preset aging limit, the accumulated aging amount will be forcibly truncated to the preset aging limit with the same sign as the original, to avoid the polarity reversal of the corrected power increment due to overcompensation, which would cause a fatal logic error of reverse meter reading. In this embodiment, the preset aging limit is set to 5% of the absolute value of the current first corrected power increment, to ensure that the single aging compensation range is always within a safe and physically reasonable range. The implementer can set the preset aging limit according to the limit aging parameters allowed by the equipment, which is not limited here. In order to restore the pure power data after deducting aging attenuation, the current first corrected power increment is subtracted from the accumulated aging amount to obtain the current second corrected power increment.

[0030] Step S4: Filter and smooth the second corrected power increment within the current time period to obtain the current final corrected power increment. Combine this with the final accumulated power at the previous moment to obtain the current final power.

[0031] Specifically, considering that the second corrected power increment, after separating measurable and aging errors, may still contain unmeasurable transient random interference caused by high-frequency transient fluctuations in the power grid, random load changes, or sensor measurement noise, in order to eliminate such occasional pulse errors and restore the purest power output value within the current cycle, the second corrected power increment within the current time period is filtered and smoothed. Combined with a preset load change threshold, the final corrected power increment is obtained. This involves arranging the second corrected power increments within the current time period in chronological order to obtain a power increment sequence. Then, median filtering is performed on the power increment sequence to remove extreme outliers (high values). (Frequency noise or sudden spikes) are filtered out, and the median value is selected to obtain the current smoothed increment estimate. Considering that directly applying the medium-range filter may misjudge the real large load step change of the power grid as spike noise and erase it, in order to strictly distinguish between isolated occasional noise and real continuous load access and avoid omissions or underestimations of metering data, the absolute values ​​of the differences between the last specified number of second-corrected power increments in the power increment sequence and the smoothed increment estimate are obtained, and all are used as jump analysis values ​​to extract the fluctuation characteristics within the latest continuous sampling window. In this embodiment, the specified number is set to 3 to ensure that short-term isolated pulses can be effectively intercepted while real step changes are quickly released. The implementer can adjust the calculation according to the power grid. The typical duration of the interference pulse is set to a specified number, which is not limited here. To quantify whether the jump amplitude is a reasonable jump caused by the start-up and shutdown of real physical electrical equipment, this embodiment introduces a preset load jump threshold based on the rated maximum step load of the equipment as a precondition. That is, if the jump analysis values ​​are all greater than the preset load jump threshold, it means that the current power deviation is not only huge in magnitude but also has temporal continuity, and it is determined to be a real step load access. Then, the current second corrected power increment is taken as the current final corrected power increment. If there is a jump analysis value less than or equal to the preset load jump threshold, it means that the current power change is not continuous or fluctuating. If the amplitude is within the range of the base noise, and is determined to be a high-frequency transient disturbance in the power grid or an isolated spike interference from a sensor, then the smoothed increment estimate is used as the current final corrected power increment, thereby achieving safe reconstruction and smooth extraction of power increment characteristics. In this embodiment, the duration of the current time period is set to cover the current moment and the past 6 consecutive sampling moments (i.e., a total of 7 data points). This ensures that the filtering window can effectively cover and suppress occasional transient disturbances, and will not cause normal load step changes to be smoothed and delayed due to an excessively long window. The implementer can set the duration of the current time period according to the sampling frequency and the characteristics of power grid fluctuations; there is no limitation here, but the end time of the current time period is always the current moment. It should be noted that if the initial stage of equipment power-on and commissioning results in insufficient historical data to include the past 6 consecutive sampling moments within the time period, the missing historical gaps will be filled by the second corrected power increment of the current moment until 7 data points are accumulated.The median filtering algorithm is well-known and will not be described in detail here.

[0032] To restore the real-time extracted pure increment to an absolute meter reading that can be directly read by users or the billing system, and then combine it with the final cumulative electricity consumption from the previous moment to obtain the current final electricity consumption, the current final corrected electricity consumption increment is added to the final cumulative electricity consumption from the previous moment. This completes the final convergence and output of electricity consumption data on the integral dimension. It should be noted that the final cumulative electricity consumption from the previous moment is obtained by reading the corrected final cumulative historical register maintained in the main control microprocessor of the electricity metering box, and can be directly accessed; if the current moment is the initial power-on moment after the equipment is put into operation, the final cumulative electricity consumption from the previous moment is initialized to 0.

[0033] In summary, this embodiment acquires temperature data, current data, power factor, and original power increment from the power metering box; determines the current ratio error and angle error based on the transformer error characteristic table; obtains the relative power deviation based on the power factor; comprehensively calculates the total error; and corrects the original power increment to obtain the first corrected power increment. A temperature-current two-dimensional grid is constructed to locate the current target grid. Based on the difference between the time variation of its first corrected power increment and the preset neighboring grid, the correction aging rate is obtained, and combined with the cumulative operating time, a second corrected power increment is obtained. The second corrected power increment is filtered and smoothed to obtain the final corrected power increment, and the final power is obtained by combining the historical final cumulative power. This invention achieves dynamic decoupling and step-by-step stripping of multi-source errors, significantly improving the long-term accuracy of power metering.

[0034] Example 2:

[0035] This invention also proposes an error correction system for an electricity metering box; please refer to [link / reference]. Figure 2 The diagram illustrates a structural diagram of an error correction system for an energy metering box according to an embodiment of the present invention. The system includes: a data acquisition module 10, a first correction energy increment acquisition module 20, a second correction energy increment acquisition module 30, and a final energy acquisition module 40.

[0036] The data acquisition module 10 is used to acquire the temperature data, current data, power factor and raw power increment of the power metering box in real time.

[0037] The first corrected power increment acquisition module 20 is used to determine the current ratio error and angle error based on the current temperature data and current data, combined with the transformer error characteristic table; to obtain the power relative deviation degree based on the current power factor and angle error; to obtain the total error degree based on the current ratio error and power relative deviation degree; to correct the current original power increment; and to determine the current first corrected power increment.

[0038] The second correction power increment acquisition module 30 is used to construct a temperature-current two-dimensional grid and locate the current target grid. Based on the difference between the change of the first correction power increment of the current target grid and the preset neighboring grid within a specified time period, the correction aging rate corresponding to the current target grid is obtained. Combined with the cumulative running time of the current target grid, the current first correction power increment is corrected to obtain the current second correction power increment.

[0039] The final power acquisition module 40 is used to filter and smooth the second corrected power increment within the current time period to obtain the current final corrected power increment, and combine it with the final accumulated power at the previous moment to obtain the current final power.

[0040] It should be noted that the error correction system for an electricity metering box and the error correction method for an electricity metering box provided in the above embodiments belong to the same concept. The specific implementation process can be found in the method embodiments, and will not be repeated here.

Claims

1. An error correction method for an electricity metering box, characterized in that, The method includes the following steps: Real-time acquisition of temperature data, current data, power factor, and raw power increment of the power metering box; Based on the current temperature and current data, and in conjunction with the transformer error characteristic table, determine the current ratio error and angle error; based on the current power factor and angle error, obtain the degree of relative power deviation; based on the current ratio error and degree of relative power deviation, obtain the total error degree, correct the current original power increment, and determine the current first corrected power increment; Construct a temperature-current two-dimensional grid and locate the current target grid. Based on the difference between the change of the first corrected power increment of the current target grid and the preset neighboring grid within a specified time period, obtain the corrected aging rate corresponding to the current target grid. Combine the cumulative running time of the current target grid to correct the current first corrected power increment and obtain the current second corrected power increment. The second corrected power increment within the current time period is filtered and smoothed to obtain the current final corrected power increment. Combined with the final accumulated power at the previous moment, the current final power is obtained. The method for obtaining the degree of relative power deviation is as follows: Calculate the inverse cosine of the current power factor to obtain the true phase angle corresponding to the current power factor, and add the true phase angle to the current angle difference to obtain the offset phase angle; Subtract the cosine of the offset phase angle from the current power factor to obtain the deviation numerator; Take the maximum value between the absolute value of the current power factor and the preset minimum positive value as the denominator of the deviation; Divide the deviation numerator by the deviation denominator to obtain the relative power deviation. The method for obtaining the corrected aging rate is as follows: Based on the trend of the first corrected power increment of the current target grid over a specified time period, the initial aging rate of the current target grid is obtained by linear regression. The average of the initial aging rates of the preset neighboring grids of the current target grid is used as the reference aging rate. The absolute value of the difference between the initial aging rate corresponding to the current target mesh and the reference aging rate is used as the relative numerator; The absolute value of the reference aging rate and the maximum value among the preset minimum change values ​​are used as the relative denominator; The ratio of the relative numerator to the relative denominator is used as the relative deviation degree of the current target grid. When the relative deviation is less than the preset relative deviation threshold, the initial aging rate corresponding to the current target mesh is used as the corrected aging rate corresponding to the current target mesh. When the relative deviation is greater than or equal to the preset relative deviation threshold, the initial aging rate and the reference aging rate corresponding to the current target mesh are weighted and summed to obtain the corrected aging rate corresponding to the current target mesh.

2. The error correction method for an electricity metering box as described in claim 1, characterized in that, The method for obtaining the total error level is as follows: The sum of the current ratio difference and the relative power deviation is taken as the current total error level.

3. The error correction method for an electricity metering box as described in claim 1, characterized in that, The method for obtaining the first corrected power increment is as follows: Subtract the total error from the preset benchmark value to obtain the first correction coefficient; The product of the current original power increment and the first correction coefficient is used as the current first corrected power increment.

4. The error correction method for an electricity metering box as described in claim 1, characterized in that, The method for constructing a temperature-current two-dimensional grid and locating the current target grid is as follows: Based on preset temperature intervals and preset current intervals, the operating space of the power metering box is divided into multiple two-dimensional grids. The two-dimensional grid into which the current temperature and current data fall is used as the current target grid.

5. The error correction method for an electricity metering box as described in claim 1, characterized in that, The method for obtaining the second corrected power increment is as follows: Multiply the correction aging rate corresponding to the current target mesh by the cumulative running time of the current target mesh to obtain the cumulative aging amount of the current target mesh; If the absolute value of the accumulated aging amount is greater than the preset aging limit, the accumulated aging amount will be forcibly truncated to the preset aging limit with the same sign as the original. Subtract the accumulated aging amount from the current first corrected power increment to obtain the current second corrected power increment.

6. The error correction method for an electricity metering box as described in claim 1, characterized in that, The method for obtaining the final battery level is as follows: Arrange the second corrected power increments within the current time period in chronological order to obtain the power increment sequence; Median filtering is performed on the power increment sequence to obtain the current smoothed increment estimate; The absolute values ​​of the differences between the last specified number of second-corrected power increments in the power increment sequence and the smoothed increment estimate are all used as jump analysis values. If all jump analysis values ​​are greater than the preset load jump threshold, then the current second corrected power increment will be used as the current final corrected power increment. If a jump analysis value is less than or equal to a preset load jump threshold, the smoothed increment estimate will be used as the current final corrected power increment. The current final power increment is added to the previous final cumulative power to obtain the current final power.

7. The error correction method for an electricity metering box as described in claim 1, characterized in that, The methods for obtaining the ratio difference and angle difference are as follows: Retrieve the current temperature and current data from the current transformer error characteristic table. If a matching node exists, directly read the corresponding ratio error and angle error to determine the current ratio error and angle error. If no matching node exists, extract the data from the four adjacent operating condition nodes surrounding the current temperature and current data, and calculate the current ratio difference and angle difference using bilinear interpolation.

8. An error correction system for an electricity metering box, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the error correction method for an energy metering box as described in any one of claims 1-7.

Images