Method for correcting and selecting daily timing error of electric energy meter

Abstract

The invention discloses a method for realizing daily timing error correction and selection of an electric energy meter, relates to the technical field of electric energy meter metering, and realizes online and real-time intelligent judgment on whether the electric energy meter has an adjustment value or not by constructing a dynamic screening mechanism of prediction, verification and decision. A theoretical reference is established according to a standard frequency-temperature curve of a crystal oscillator, at the initial stage of low-temperature adjustment, actual measurement error data of first several temperature points are only utilized to be fitted with the theoretical curve, an initial error prediction range is rapidly calculated, the prediction range is dynamically verified and adaptively updated along with data input of subsequent test points, and the error prediction range is accurately and accurately adjusted. When new data falls within the range, it is confirmed that the characteristics of the electric energy meter are stable, and when the new data continuously exceed the range, the electric energy meter is judged and removed in advance, so that the electric energy meter without good temperature consistency can be recognized and removed in time in the early stage of the adjustment process, and follow-up invalid high-temperature testing and compensation value calculation operation are avoided.

Description

Technical Field

[0001] This invention relates to the field of electricity meter measurement technology, specifically a method for correcting and selecting daily timing errors in electricity meters. Background Technology

[0002] With the deepening of smart grid construction and refined energy metering management, the long-term accuracy and stability of electricity meters, as the cornerstone of electricity trade settlement and data collection, are of paramount importance. Daily timing error is a core indicator for evaluating the clock performance of electricity meters, mainly caused by the drift of the internal clock crystal frequency due to temperature changes. To ensure the metering reliability of electricity meters across the entire temperature range, temperature compensation calibration of daily timing error must be performed during the production process. This is a key process for improving product consistency and pass rate.

[0003] Currently, existing methods for calibrating daily timekeeping errors in electricity meters typically involve measuring the error at multiple temperature points and then performing a simple weighted average of the acquired data to calculate the compensation value. These methods have significant limitations: First, at the data processing level, simple weighted averaging cannot effectively identify and eliminate jump errors and gross errors caused by transient characteristics of crystal oscillators or measurement interference, resulting in distortion of the compensation reference. Secondly, in terms of process logic, the existing model of "calibrating all first and then uniformly recalibrating and screening" lacks predictive ability and cannot identify and eliminate electricity meters with unstable error-temperature characteristics and no calibration value in the early stage of calibration, resulting in a large amount of ineffective working hours and wasted resources. Finally, the complete "high-normal-low" three-temperature zone testing process is lengthy, which restricts production cycle and efficiency.

[0004] These defects lead to high calibration costs and low efficiency for electricity meters, and make it difficult to accurately control the consistency of the final product.

[0005] In summary, existing technical solutions have systemic shortcomings in terms of predictability, intelligence, and overall efficiency in the calibration process. Therefore, there is an urgent need for a new method that can deeply clean and optimize error data, assess the calibration potential of electricity meters in real time and achieve intelligent selection, and optimize the testing process, so as to significantly improve production efficiency and product quality while ensuring calibration accuracy. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method for correcting and selecting daily timing errors in electricity meters. This method utilizes a dynamic screening mechanism involving prediction, verification, and decision-making to achieve online, real-time intelligent judgment of whether an electricity meter is calibrable. Based on a theoretical benchmark established using the standard frequency-temperature curve of a crystal oscillator, in the initial stage of low-temperature calibration, only the measured error data from the first few temperature points are fitted to the theoretical curve to quickly calculate an initial error prediction range. This prediction range is dynamically verified and adaptively updated with subsequent test data input. When new data falls within the range, the meter's characteristics are confirmed to be stable; if it continuously exceeds the range, it is prematurely rejected. The working principle of this mechanism essentially transforms traditional post-event result verification into online process diagnosis based on real-time data streams. This allows electricity meters lacking good temperature consistency to be identified and rejected early in the calibration process, avoiding subsequent ineffective high-temperature tests and compensation value calculations, thereby fundamentally optimizing the allocation of calibration resources.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: On one hand, a method for correcting and selecting daily timing errors in electricity meters, the specific steps of which are as follows: S100. Establish a standard curve: Analyze the frequency-temperature characteristics of the clock crystal used in the energy meter under test, and plot the standard frequency-temperature curve of the clock crystal through computer simulation. S200, Low Temperature Calibration and Initial Prediction Screening: Controlling the temperature drop of the high and low temperature chamber to the preset value. At several low-temperature test points, daily timing error data of the electricity meters at each test point were obtained, based on the previous... The daily timing error data of the aforementioned low-temperature test points are used to calculate the fitting degree with the standard frequency-temperature curve, generating a first prediction range, which is then used. The daily timing error data of each low temperature test point is used to verify the first prediction range. Based on the verification results, the electricity meters are initially screened, and the final prediction range is updated and output. S300, High Temperature Calibration and Final Judgment: Control the high and low temperature chamber to rise to the preset temperature. At each high-temperature test point, the daily timing error data of the electricity meter is obtained. It is determined whether the daily timing error data of each high-temperature test point is within the final prediction range after S200 update. Based on the determination result, the electricity meter is finally corrected and eliminated. S400, Error Data Processing: The daily timing error data obtained in S200 and S300 are processed, including outlier removal, data smoothing, and adaptive filtering correction, to obtain an error value sequence for calculating compensation values. S500, Write Compensation Value: For energy meters that have been determined to be qualified by S300, calculate the daily timing error compensation value based on the error value sequence obtained by S400, and write the daily timing error compensation value into the corresponding qualified energy meter.

[0008] Furthermore, the specific steps for the S200 low-temperature calibration and initial prediction screening are as follows: S210: Control the high and low temperature chambers to descend sequentially to the first set of low temperature test points, and obtain the daily timing error data of the electricity meter at each test point through the daily timing device; S220, Before Selection The daily timing error data of each low-temperature test point is used to calculate the root mean square error between the error and the standard frequency-temperature curve. Based on the root mean square error and the preset confidence interval, the first prediction range is calculated. S230, Continue to obtain subsequent information Based on the daily timing error data of each low-temperature test point, it is determined whether the actual error value of each low-temperature test point is within the first prediction range. The specific logic is as follows: S231, when all subsequent If the actual error of each low-temperature test point is within the first prediction range, then the daily timing error of the electricity meter is determined to be stable and has calibration value, and the first prediction range is used as the final prediction range for S300. S232, when the follow-up If the actual error of any low-temperature test point exceeds the first prediction range but does not exceed 1.5 times its range, then the data of that low-temperature test point is included in the calculation, the first prediction range is updated, and subsequent points are verified. S233, when the follow-up If the actual error at any low-temperature test point exceeds 1.5 times the first prediction range, the energy meter is deemed to have poor error stability and is not worth adjusting, and is therefore rejected.

[0009] Furthermore, in S200, the root mean square error ,in, This refers to the number of daily timing error data points used in the fitting process, i.e., the number of low-temperature test points. Represents the root mean square error, used before quantization. The degree of deviation between the measured error data of each low-temperature test point and the standard curve. Represented on the standard frequency-temperature curve, the first The theoretical error value corresponding to each test temperature point Representing the The measured daily timing error value for each test temperature point, the first prediction range is set based on the calculated RMSE value, and represents a deviation interval of the corresponding value of the standard frequency-temperature curve, i.e. The smaller the RMSE value, the higher the fit between the actual error data and the standard curve, the better the error stability, and the higher the reliability of the prediction range.

[0010] Furthermore, the specific steps for the S300 high-temperature calibration and final determination are as follows: S310, control the high and low temperature chambers to rise sequentially to At each high-temperature test point, the daily timing error data of the electricity meter is obtained. S320. Compare the daily timing error data of each high-temperature test point with the final prediction range output by S200, specifically including: S321. When the daily timing error data of all high-temperature test points are within the final prediction range, the energy meter is deemed to have qualified error consistency. S322. If the daily timing error data of only one high-temperature test point exceeds the final prediction range, then the energy meter is determined to be a candidate meter. S323. If the error data of two or more high-temperature test points exceeds the final prediction range, the energy meter is determined to have poor error consistency and is discarded. S330. For electricity meters that are deemed qualified, execute S500.

[0011] Furthermore, the specific steps of the S400 error data processing are as follows: S410. Collect daily timing error data at a single test temperature point at fixed time intervals to form the original error data sequence of that temperature point. S420. Convert the original error data sequence into a standard normal distribution and remove outlier data points; S430. The original error data sequence after removing outliers is smoothed using the moving average method. S440. The least mean square adaptive filtering algorithm is applied to correct the smoothed original error data sequence in order to optimize the original error data sequence and extract the optimal estimate for compensation.

[0012] Furthermore, in step S420, statistical analysis is performed on the original error data sequence acquired at a single temperature point to calculate the mean of the original error data sequence. and standard deviation ; Remove outlier data points that deviate from the mean by more than three standard deviations, i.e., remove data points that meet the following criteria: - |>3 Error data ,in, For the first The raw error data collected this time.

[0013] Furthermore, the moving average method in S430 is used to perform moving average processing on the original error data sequence after removing outlier data points, and the smoothed error value is... ,in, Indicates time Error value after smoothing For the first The original error data collected after removing outlier data points. For the time of data collection, The preset sliding window size is used to control the smoothness, and its value is a positive integer.

[0014] Furthermore, in S440, the least mean square adaptive filtering optimizes the original error data sequence after smoothing in S430 to generate the optimal estimate for calculating the compensation value. The least mean square adaptive filtering algorithm uses time... Filter weights, time Smoothing error value and time Based on the input signal, the filter weights are iteratively updated, and the corrected error is calculated. The specific optimization steps are as follows: Based on the initial values ​​of the filter weights and the learning rate, for each time step in the smoothed original error data sequence... Obtain the current input signal The input signal This is the normalized temperature value for the current test temperature point; Calculation time intermediate error value ,in, For S430 output Error value after time smoothing for The filter weights at each time step; Update the filter weights according to the least mean square criterion: ,in, For the updated The time-major filter weights are used to predict and correct subsequent errors; Output the corrected error value at the current time. ,Should This refers to a data point in the error data. Through iterative calculation, the corrected error value sequence of the entire original error data sequence is finally obtained, which serves as the input for calculating the daily timing error compensation value in the S500.

[0015] On the other hand, a system for correcting and selecting daily timing errors of electricity meters is provided. This system consists of host computer software, a high and low temperature chamber, a daily timing device, a control circuit, and multiple electricity meters to be tested. The host computer software is used to execute process control, data calculation and logical judgment, and has functions such as data acquisition, curve plotting, error calculation, prediction range generation, high and low temperature chamber temperature control and energy meter switching control. The high and low temperature chamber is controlled by the host computer software and is used to provide the temperature environment for the low temperature test point and the high temperature test point. The daily timing device is used to collect the daily timing error data of the energy meter under test and transmit it to the host computer software in real time. The control circuit receives switching instructions from the host computer software to switch the test of the energy meter under test, and simultaneously receives the compensation value output by the host computer software and writes it into the corresponding energy meter. The energy meter under test: establishes a connection with the system through the control circuit, and accepts error data acquisition and compensation value writing operations.

[0016] Compared with existing technologies, this method for correcting and selecting daily timing errors in electricity meters has the following advantages: This invention achieves online, real-time intelligent judgment on whether an energy meter is calibrated by constructing a dynamic screening mechanism of prediction, verification, and decision-making. Based on the standard frequency-temperature curve of the crystal oscillator, a theoretical benchmark is established. In the initial stage of low-temperature calibration, only the measured error data of the first few temperature points are fitted with the theoretical curve to quickly calculate an initial error prediction range. This prediction range is dynamically verified and adaptively updated with the input of data from subsequent test points. When new data falls within the range, the meter's characteristics are confirmed to be stable; if it continuously exceeds the range, it is prematurely rejected. The working principle of this mechanism is essentially to transform traditional post-event result verification into online process diagnosis based on real-time data streams. This allows energy meters lacking good temperature consistency to be identified and rejected early in the calibration process, avoiding subsequent ineffective high-temperature tests and compensation value calculations, thereby fundamentally optimizing the allocation of calibration resources.

[0017] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0019] Figure 1 A flowchart illustrating the steps of a method for correcting and selecting daily timing errors in electricity meters; Figure 2 This is a schematic diagram of a system structure for correcting and selecting daily timing errors in electricity meters; Figure 3 This is an operation flowchart of a method for correcting and selecting daily timing errors in electricity meters. Detailed Implementation

[0020] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0021] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a method for implementing daily timing error correction and selection of an electricity meter,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plurality forms, unless the context clearly indicates otherwise; “plural” generally includes at least two.

[0022] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0023] To address the problems of existing technologies, this invention first describes the scenario of daily timing error correction and selection for electricity meters. This invention is primarily applied to the final calibration and screening stage of electricity meter manufacturing, aiming to solve the problem of daily timing errors caused by temperature drift of the internal clock crystal oscillator frequency. In mass production environments, traditional methods involve testing all electricity meters across the entire temperature range and uniformly compensating them. This not only results in long testing cycles and low efficiency but also fails to identify and eliminate products with unstable temperature characteristics and no calibration value in advance, leading to resource waste and increased costs. This invention constructs a dynamic intelligent screening mechanism based on prediction, verification, and decision-making. It evaluates the error stability of electricity meters in real time during testing, achieving early screening and accurate compensation, significantly improving calibration efficiency and product consistency.

[0024] This invention provides a method for correcting and selecting daily timing errors in electricity meters. It establishes a crystal oscillator standard frequency-temperature curve as a theoretical benchmark. During the low-temperature testing phase, partial data is used to predict and verify error behavior, dynamically selecting electricity meters with stable characteristics. The final judgment is made during the high-temperature testing phase. Furthermore, the error data collected throughout the process is deeply cleaned and optimized, and finally, accurate compensation values ​​are calculated and written. This method transforms traditional post-event inspection into online process diagnosis, achieving intelligent and efficient calibration procedures.

[0025] Specifically, such as Figure 1 As shown, a method for correcting and selecting daily timing errors in electricity meters is described. The specific steps of this method are as follows: S100. Analyze the frequency-temperature characteristics of the clock crystal used in the energy meter under test, and plot the standard frequency-temperature curve of the clock crystal using computer simulation. S200, controls the high and low temperature chamber to cool down to the preset temperature. At several low-temperature test points, daily timing error data of the electricity meters at each test point were obtained, based on the previous... Daily timing error data from several low-temperature test points are used to calculate the fit with the standard frequency-temperature curve, generating a first prediction range. This range is then used... The daily timing error data of each low-temperature test point is used to verify the first prediction range. Based on the verification results, the electricity meters are initially screened, and the final prediction range is updated and output. S300, controls the high and low temperature chamber to rise to the preset temperature. A number of high-temperature test points are used to obtain the daily timing error data of the electricity meter at each high-temperature test point. It is determined whether the error of each high-temperature test point is within the final prediction range after S200 update, and the electricity meter is finally corrected and eliminated based on the judgment result. S400: Process the daily timing error data obtained in S200 and S300. The processing includes removing outliers, data smoothing, and adaptive filtering correction to obtain an error value sequence for calculating compensation values. S500: For the energy meters that have been determined to be qualified by S300, calculate the daily timing error compensation value based on the error value sequence obtained by S400, and write the daily timing error compensation value into the corresponding qualified energy meter.

[0026] In the specific implementation process, such as Figure 2 As shown, the implementation of the method of the present invention relies on a system consisting of host computer software, a high and low temperature chamber, a daily timing device, a control circuit, and multiple energy meters under test. The host computer software serves as the control core, responsible for process scheduling, temperature control, data acquisition, calculation and analysis, and instruction issuance. The high and low temperature chamber provides a precise and controllable high and low temperature testing environment. The daily timing device is used to collect daily timing error data of the energy meters with high precision. The control circuit realizes the automatic switching test of multiple energy meters and the writing of compensation values. The energy meters under test are mass-produced products awaiting calibration.

[0027] Specifically, the first step is to establish a standard frequency-temperature characteristic curve for the clock crystal oscillator used in the energy meter under test. This curve is obtained through computer simulation or actual measurement and calibration of crystal oscillator samples, reflecting the typical law of crystal oscillator frequency change with temperature. The curve data is pre-stored in the host computer software in the form of function tables or fitting formulas, serving as a theoretical benchmark for subsequent error prediction and evaluation. The establishment of the standard curve ensures that the calibration process has a unified and objective reference, avoiding evaluation deviations caused by individual differences in crystal oscillators and inconsistent theoretical models.

[0028] The high and low temperature chamber is controlled to start from room temperature and gradually decrease to n preset low temperature test points, such as -25°C, -15°C, -5°C, etc. After the temperature stabilizes at each point, the daily timing error data of the current energy meter under test is collected by the daily timing device and uploaded to the host computer software in real time.

[0029] The host computer software first uses the error data collected from the first n / 2 low-temperature test points to perform a fitting analysis with the standard frequency-temperature curve stored in S100. Specifically, it calculates the root mean square error (RMSE) between these first n / 2 measured error values ​​and the theoretical values ​​at the corresponding temperature points on the standard curve. The formula for calculating RMSE is: ,in, This refers to the number of error data points used in the fitting process, i.e., the number of low-temperature test points. Represents the root mean square error, used before quantization. The degree of deviation between the measured error data of each low-temperature test point and the standard curve. This represents the first frequency-temperature curve on the standard frequency-temperature curve. The theoretical error value corresponding to each test temperature point Representing the The measured daily timing error value for each test temperature point. RMSE quantifies the degree of deviation between the measured data and the standard model.

[0030] Based on the calculated RMSE value, the host computer software generates a first prediction range. In this embodiment, the first prediction range is set to the interval between the theoretical value of the standard curve and the range of fluctuation by twice the RMSE, i.e. This range represents a preliminary prediction of the error behavior of the energy meter at subsequent temperature points.

[0031] Subsequently, the measured error data of the next n / 2 low-temperature test points were obtained and compared with the first prediction range one by one for verification. The verification logic is as follows: If the measured error of all subsequent test points falls within the first prediction range, it indicates that the error-temperature characteristics of the energy meter are highly consistent with the standard model, have good stability, and are calibrated. In this case, the first prediction range is directly used as the final prediction range for subsequent high temperature determination. If the measured error of any subsequent test point exceeds the first prediction range but does not exceed 1.5 times its range, it is considered that there is a slight deviation. The host computer software will include the data of that point, recalculate the RMSE and update the prediction range, and then continue to verify the remaining points. If the measured error at any subsequent test point exceeds 1.5 times the first predicted range, the meter is deemed to have unstable and inconsistent error-temperature characteristics, making it unworthy of calibration. Subsequent testing of the meter is immediately terminated, and it is marked as a rejected product. This mechanism allows defective products to be eliminated early in the low-temperature testing phase, avoiding ineffective high-temperature testing and data processing, and saving significant time and resources.

[0032] For the selected electricity meters, the high and low temperature chamber is controlled to sequentially rise to m preset high temperature test points, such as +45°C, +60°C, +70°C, etc., and daily timing error data at each point is collected. A final consistency check is performed under high temperature conditions. The host computer software compares the measured error data of each high temperature test point with the output final prediction range. The judgment logic is as follows: If the error data of all high-temperature test points fall within the final prediction range, the energy meter is judged to have excellent error consistency across the entire temperature range and is a qualified meter. If the error data of only one high-temperature test point exceeds the final prediction range, then the table is determined to be a candidate table; If the error data of two or more high-temperature test points exceeds the final prediction range, the electricity meter is judged to have poor error consistency and will be ultimately rejected.

[0033] Only electricity meters that are deemed qualified or as candidates will proceed to the subsequent compensation value writing process. This ensures that the electricity meters that ultimately receive compensation all have good temperature stability.

[0034] For the selected electricity meters, all valid daily timing error raw data collected in S200 and S300 need to undergo data processing to obtain a clean and reliable error value sequence for calculating compensation values. The processing steps are as follows: Data acquisition and serialization: At each test temperature point, daily timing error data is continuously collected over a period of time at fixed time intervals to form an original error data sequence for that temperature point; Outlier removal: Statistical analysis is performed on the original error data sequence for each temperature point, calculating its mean μ and standard deviation σ. Outliers deviating from the mean by more than 3 times the standard deviation (i.e., satisfying |σ / σ) are removed. - |>3 ) data points It can effectively remove errors caused by transient interference or measurement noise; Data smoothing: The moving average method is used to smooth the sequence after removing outliers. The smoothing formula is: ,in, Indicates time Error value after smoothing For the first The original error data collected after removing outlier data points. For the time of data collection, The preset sliding window size is used to control the smoothness, and its value is a positive integer. This moving average method reduces random fluctuations and makes the error trend clearer. Adaptive Filtering Correction: To further optimize the data, a least mean square adaptive filtering algorithm is employed. This algorithm optimizes the original error data sequence after S430 smoothing to generate the optimal estimated error value for calculating the compensation value. The least mean square adaptive filtering algorithm uses time... Filter weights, time Smoothing error value and current time Based on the input signal, the filter weights are iteratively updated, and the corrected error is calculated. The specific optimization steps are as follows: Based on the initial values ​​of the filter weights and learning rate The learning rate A preset positive decimal value is used to control the rate of weight updates and convergence stability for each time step in the smoothed original error data sequence. Obtain the current input signal Input signal This is the normalized temperature value for the current test temperature point; Calculation time intermediate error value ,in, For S430 output Error value after time smoothing for The filter weights at each time step; Update the filter weights according to the least mean square criterion: ,in, For the updated The time-major filter weights are used to predict and correct subsequent errors; Output the corrected error value at the current time. ,Should This refers to a data point in the error data. Through iterative calculation, the corrected error value sequence of the entire original error data sequence is finally obtained, which serves as the input for calculating the daily timing error compensation value in the S500.

[0035] Based on the obtained final error value sequence covering high and low temperature test points, the host computer software uses curve fitting to calculate a globally optimal daily timing error compensation value. The control circuit receives the instructions and compensation value issued by the host computer software and writes them into the memory of the corresponding qualified energy meter. After writing, the energy meter will call this compensation value to perform real-time correction of the internal clock during its working life, thereby maintaining high-precision timing function across the entire temperature range.

[0036] In summary, this invention achieves intelligent and efficient daily timing error correction of electricity meters through a closed-loop process of standard curve modeling, low-temperature prediction screening, high-temperature final judgment, deep data processing, and precise compensation writing. It moves the screening process forward and makes it dynamic, making real-time decisions during the testing process, eliminating defective products in advance, and performing a complete compensation process only on electricity meters with stable characteristics. As a result, it significantly improves production calibration efficiency and reduces overall costs while ensuring the high quality of the final product.

[0037] like Figure 3 As shown, this paper describes a method for correcting and selecting daily timing errors in electricity meters. The specific process is as follows: (1) Initialization and preparation Establish a standard curve: Analyze the characteristics of the clock crystal oscillator used in the current batch of electricity meters, and plot its standard frequency-temperature curve through computer simulation as a benchmark for subsequent predictions.

[0038] (2) Low temperature calibration and predictive initial screening Start low temperature test: The host computer software controls the high and low temperature chamber to start cooling down.

[0039] Obtain the first batch of low temperature data: The high and low temperature chamber is stabilized at the preset first half of the low temperature point in sequence. At each temperature point, the power meter is switched by the control circuit, and the daily timing error data is obtained by the daily timing device.

[0040] Calculate the initial prediction range: Using the error data of the first few low temperature points, calculate their fit with the standard curve to generate an initial prediction range.

[0041] Verify predictions and dynamically filter: The high and low temperature chamber continues to cool down to the latter part of the low temperature point, and the actual error data of these points is obtained. Each newly obtained actual error value is compared with the current prediction range, and a judgment is made based on the comparison results: If the new error values ​​are all within the prediction range, the table is considered to have stable errors, and it is retained and proceeds to the next step.

[0042] If the new error value exceeds the prediction range but does not exceed 1.5 times, the prediction range is updated based on the new data, and subsequent points are examined.

[0043] If the new error value exceeds the prediction range by 1.5 times, the table is immediately determined to have poor error stability and is not worth adjusting. It is then marked and removed from the subsequent process.

[0044] Output final prediction range: After completing all low-temperature point tests, generate a final prediction range for high-temperature calibration for the selected energy meters.

[0045] (3) High temperature calibration and final judgment Start high temperature test: The host computer software controls the high and low temperature chamber to start heating.

[0046] Acquire high temperature data: The high and low temperature chamber is stabilized at all preset high temperature points in sequence, and the daily timing error data of each point is acquired.

[0047] Final consistency determination: Compare the error data for each high-temperature point with the final prediction range: Qualified: If the error of all high-temperature points is within the predicted range, the consistency of the table is judged to be excellent.

[0048] Selection of alternatives: If only one high-temperature point has an error exceeding the range, this table can be considered as an alternative.

[0049] Judgment and Removal: If two or more high-temperature points have errors exceeding the range, the table is judged to have poor consistency and is removed.

[0050] (4) Data processing and compensation Optimize error data: For the electricity meters that have passed the final judgment, process the raw error data collected at all temperature points: Outlier removal: Use statistical methods to remove erroneous data points that deviate significantly from the normal range.

[0051] Data smoothing: Using methods such as moving averages to smooth data and reduce the impact of random fluctuations.

[0052] Adaptive filtering: The algorithm is applied to further correct the error trend and obtain the optimized error value sequence that best reflects the true characteristics of the crystal oscillator.

[0053] Calculate and write the compensation value: Based on the obtained optimized error value sequence, calculate an optimal compensation value that can compensate for the error across the entire temperature range, and write this value into the memory of the energy meter through the control circuit.

[0054] (5) End of process Calibration complete: All qualified electricity meters have been processed, and the process is finished.

[0055] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for correcting and selecting daily timing errors in electricity meters, characterized in that, The specific steps of this method are as follows: S100. Establish a standard curve: Analyze the frequency-temperature characteristics of the clock crystal used in the energy meter under test, and plot the standard frequency-temperature curve of the clock crystal through computer simulation. S200, Low Temperature Calibration and Initial Prediction Screening: Controlling the temperature drop of the high and low temperature chamber to the preset value. At several low-temperature test points, daily timing error data of the electricity meters at each test point were obtained, based on the previous... The daily timing error data of the aforementioned low-temperature test points are used to calculate the fitting degree with the standard frequency-temperature curve, generating a first prediction range, which is then used. The daily timing error data of each low temperature test point is used to verify the first prediction range. Based on the verification results, the electricity meters are initially screened, and the final prediction range is updated and output. S300, High Temperature Calibration and Final Judgment: Controlling the high and low temperature chamber to rise to the preset temperature. At each high-temperature test point, the daily timing error data of the electricity meter is obtained. It is determined whether the daily timing error data of each high-temperature test point is within the final prediction range after S200 update. Based on the determination result, the electricity meter is finally corrected and eliminated. S400, Error Data Processing: The daily timing error data obtained in S200 and S300 are processed, including outlier removal, data smoothing, and adaptive filtering correction, to obtain an error value sequence for calculating compensation values. S500, Write Compensation Value: For energy meters that have been determined to be qualified by S300, calculate the daily timing error compensation value based on the error value sequence obtained by S400, and write the daily timing error compensation value into the corresponding qualified energy meter.

2. The method for correcting and selecting daily timing errors in electricity meters according to claim 1, characterized in that, The specific steps for the S200 low-temperature calibration and initial prediction screening are as follows: S210: Control the high and low temperature chambers to descend sequentially to the first set of low temperature test points, and obtain the daily timing error data of the electricity meter at each test point through the daily timing device; S220, Before Selection The daily timing error data of each low-temperature test point is used to calculate the root mean square error between the error and the standard frequency-temperature curve. Based on the root mean square error and the preset confidence interval, the first prediction range is calculated. S230, Continue to obtain subsequent information Based on the daily timing error data of each low-temperature test point, it is determined whether the actual error value of each low-temperature test point is within the first prediction range. The specific logic is as follows: S231, when all subsequent If the actual error of each low-temperature test point is within the first prediction range, then the daily timing error of the electricity meter is determined to be stable and has calibration value, and the first prediction range is used as the final prediction range for S300. S232, when the follow-up If the actual error of any low-temperature test point exceeds the first prediction range but does not exceed 1.5 times its range, then the data of that low-temperature test point is included in the calculation, the first prediction range is updated, and subsequent points are verified. S233, when the follow-up If the actual error at any low-temperature test point exceeds 1.5 times the first prediction range, the energy meter is deemed to have poor error stability and is not worth adjusting, and is therefore rejected.

3. The method for correcting and selecting daily timing errors in electricity meters according to claim 2, characterized in that, In S200, the root mean square error ,in, This refers to the number of daily timing error data points used in the fitting process, i.e., the number of low-temperature test points. Represents the root mean square error, used before quantization. The degree of deviation between the measured error data of each low-temperature test point and the standard curve. Represented on the standard frequency-temperature curve, the first The theoretical error value corresponding to each test temperature point Representing the The measured daily timing error value for each test temperature point, the first prediction range is set based on the calculated RMSE value, and represents a deviation interval of the corresponding value of the standard frequency-temperature curve, i.e. .

4. The method for correcting and selecting daily timing errors in electricity meters according to claim 1, characterized in that, The specific steps for the S300 high-temperature calibration and final determination are as follows: S310, control the high and low temperature chambers to rise sequentially to At each high-temperature test point, the daily timing error data of the electricity meter is obtained. S320. Compare the daily timing error data of each high-temperature test point with the final prediction range output by S200, specifically including: S321. When the daily timing error data of all high-temperature test points are within the final prediction range, the energy meter is deemed to have qualified error consistency. S322. If the daily timing error data of only one high-temperature test point exceeds the final prediction range, then the energy meter is determined to be a candidate meter. S323. If the error data of two or more high-temperature test points exceeds the final prediction range, the energy meter is determined to have poor error consistency and is discarded. S330. For electricity meters that are deemed qualified, execute S500.

5. The method for correcting and selecting daily timing errors in electricity meters according to claim 1, characterized in that, The specific steps for processing the S400 error data are as follows: S410. Collect daily timing error data at a single test temperature point at fixed time intervals to form the original error data sequence of that temperature point. S420. Convert the original error data sequence into a standard normal distribution and remove outlier data points; S430. The original error data sequence after removing outliers is smoothed using the moving average method. S440. The least mean square adaptive filtering algorithm is applied to correct the smoothed original error data sequence in order to optimize the original error data sequence and extract the optimal estimate for compensation.

6. The method for correcting and selecting daily timing errors in electricity meters according to claim 5, characterized in that, In step S420, statistical analysis is performed on the raw error data sequence acquired at a single temperature point, and the mean of the raw error data sequence is calculated. and standard deviation ; Remove outlier data points that deviate from the mean by more than three standard deviations, i.e., remove data points that meet the following criteria: - |>3 Error data ,in, For the first The raw error data collected this time.

7. The method for correcting and selecting daily timing errors in electricity meters according to claim 5, characterized in that, The moving average method in S430 is used to perform moving average processing on the original error data sequence after removing outlier data points. The smoothed error value is... ,in, Indicates time Error value after smoothing For the first The original error data collected after removing outlier data points. For the time of data collection, This is the preset size of the sliding window.

8. The method for correcting and selecting daily timing errors in electricity meters according to claim 5, characterized in that, In step S440, the least mean square adaptive filtering optimizes the original error data sequence after smoothing in step S430 to generate the optimal estimate for calculating the compensation value. The least mean square adaptive filtering algorithm uses time... Filter weights, time Smoothing error value and time Based on the input signal, the filter weights are iteratively updated, and the corrected error is calculated. The specific optimization steps are as follows: Based on the initial values ​​of the filter weights and the learning rate, for each time step in the smoothed original error data sequence... Obtain the current input signal The input signal This is the normalized temperature value for the current test temperature point; Calculation time intermediate error value ,in, For S430 output The error value after time smoothing for The filter weights at each time step; Update the filter weights according to the least mean square criterion: ,in, For the updated The time-major filter weights are used to predict and correct subsequent errors; Output the corrected error value at the current time. ,Should This refers to a data point in the error data. Through iterative calculation, the corrected error value sequence of the entire original error data sequence is finally obtained, which serves as the input for calculating the daily timing error compensation value in the S500.

9. A system for correcting and selecting daily timing errors in electricity meters, applicable to the method for correcting and selecting daily timing errors in electricity meters as described in any one of claims 1-8, characterized in that, The system consists of host computer software, a high and low temperature chamber, a daily timing device, a control circuit, and multiple energy meters to be tested; The host computer software is used to perform process control, data calculation and logical judgment, and has functions such as data acquisition, curve plotting, error calculation, prediction range generation, high and low temperature chamber temperature control and energy meter switching control. The high and low temperature chamber is controlled by the host computer software and is used to provide the temperature environment for the low temperature test point and the high temperature test point. The daily timing device is used to collect the daily timing error data of the energy meter under test and transmit it to the host computer software in real time. The control circuit receives switching instructions from the host computer software to switch the test of the energy meter under test, and simultaneously receives the compensation value output by the host computer software and writes it into the corresponding energy meter. The energy meter under test: establishes a connection with the system through the control circuit, and accepts error data acquisition and compensation value writing operations.

Images