A method and system for monitoring operation error of electric energy meter
By identifying noise points by segmenting and iterative threshold segmentation algorithms of the current data sequence, combined with cubic spline interpolation technology, the problem of poor noise denoising effect in the monitoring of electricity meter errors is solved, and efficient monitoring and error recognition of the power system is realized to ensure that Data quality and system stability are achieved.
Patent Information
- Application Number
- CN202411074087.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-07
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-08-07
AI Technical Summary
The existing electrical energy meter error monitoring methods have poor results in the denoising process, resulting in the accumulation of metering errors and affecting the economy and stability of the power system.
By segmenting the current data sequence, identifying noise points using suspected noise possibility relationship and iterative threshold segmentation algorithm, and denoising processing is performed in combination with cubic spline interpolation technology, data quality and reliability are ensured.
It improves the noise recognition accuracy and data quality, realizes real-time monitoring of the power meter and accurate operation error calculation, timely generates alarm signals, and ensures the economy and reliability of the power system.
Smart Images

Figure CN118604717B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of electric energy meter monitoring, and more particularly to a method and system for monitoring electric energy meter operation errors. Background Art
[0002] As an indispensable component of the power system, electricity meters primarily measure and record energy consumption, providing accurate data for power trading and energy management. With the development of smart grid technology and the widespread adoption of smart metering devices, the role of electricity meters has become increasingly prominent. Their accuracy and reliability are directly related to the economic efficiency and stability of the power system. Over the long term, electricity meters can experience metering errors due to factors such as device aging, mechanical wear, degradation of electronic components, and environmental changes such as temperature fluctuations, humidity changes, and unstable power supply voltage. If these errors are not detected and corrected promptly, they will gradually accumulate, affecting the accuracy of electricity metering and, in turn, the economic benefits of users and power companies. Therefore, monitoring electricity meter operational errors is crucial.
[0003] A Chinese patent application, publication number CN116859321A, discloses a method for monitoring meter errors in energy meters based on an energy controller. By performing singular value decomposition on the household meter data matrix, this method can perform calculations even when the number of data points is smaller than the number of users. This method also mitigates the problem of inaccurate judgments due to correlations between users. It also uses an iterative process to reduce the impact of noise in the data, ultimately achieving a relatively stable least-squares analytical solution.
[0004] However, since the noise sources in the surrounding environment of the electricity meter are different and the noise intensity is different at different times, different data points in the current data sequence are affected by noise in different ways. Directly denoising the overall data will result in some noise points similar to normal data being ignored, resulting in poor denoising effect and affecting the accuracy of the monitoring results. Summary of the Invention
[0005] In order to solve the problem of poor denoising effect, the present invention proposes a method and system for monitoring operation errors of an electric energy meter.
[0006] In a first aspect, the present invention discloses a method for monitoring an operation error of an electric energy meter, comprising: acquiring a current data sequence according to a preset sampling interval, taking any current data as target data, and constructing a window with the target data as the center and a preset length as the radiation range of the target data; randomly selecting a preset reference number of current data as reference current data, and obtaining the radiation range of each reference current data, wherein the reference current data does not include the target data; and calculating the suspected noise probability of the target data, wherein the suspected noise probability satisfies the relationship:
[0007] ,in, Indicates the The suspected noise possibility of the current data, Indicates the total number of reference current data, Indicates the The radiation range of the current data is Indicates the The radiation range of the reference current data, Indicates similarity, Represents an exponential function. In the relation of suspected noise probability, The current data is the target data.
[0008] The current data series is segmented based on the likelihood of suspected noise; each segment is denoised to complete operational error monitoring.
[0009] By setting the target data and its radiation range and comparing it with randomly selected reference current data, the possibility of suspected noise in the target data can be quantified. By calculating the possibility of suspected noise, the current data sequence can be intelligently segmented and the data intervals that may contain noise can be identified. This not only improves the pertinence and efficiency of data monitoring, but also provides a basis for denoising processing. After denoising each segment, the random fluctuations and noise in the current data can be removed, thereby improving the quality and reliability of the data.
[0010] Preferably, the window size of the radiation range of the reference current data is consistent with the window size of the radiation range of the target data.
[0011] Preferably, the suspected noise possibility further includes: taking the ratio of the standard deviation of the current value to the mean value in the radiation range of the target data as independence; the suspected noise possibility satisfies the relationship:
[0012] ,in, Indicates the The suspected noise possibility of the current data, Indicates the The independence of the current data, Indicates the total number of reference current data, Indicates the The radiation range of the current data is Indicates the The radiation range of the reference current data, Indicates similarity, Represents an exponential function. In the relation of suspected noise probability, The current data is the target data.
[0013] By introducing the concept of independence, the volatility and dispersion of the current data are quantified by the ratio of the standard deviation to the mean within its radiation range, thereby evaluating the stability of each current data point. This quantitative method not only captures the inherent variability of the data, but also enhances the measurement of data consistency by comparing with the mean.
[0014] Preferably, the segmentation of the current data sequence includes: taking the left endpoint or the right endpoint of the current data sequence as the starting point and a preset number of current data as the segment length to obtain the first segment; taking the minimum value of the current data in the segment as the grouping threshold, dividing the current data in the segment into two categories, calculating the preference of the segment, traversing the grouping threshold from the minimum value to the maximum value, obtaining the preference of each grouping, and taking the maximum preference as the actual preference of the current segment length; iterating the segment length according to the preset step size, calculating the actual preference of the first segment at each segment length, and when the actual preference of the segment length of the first segment decreases twice in a row during the iteration process, the segment length before the decrease is taken as the final segment length of the first segment.
[0015] Starting from the endpoints of the current data sequence, the data is segmented into segments of preset lengths, ensuring an orderly and comprehensive analysis. The data is then divided into two categories by setting the minimum current value within a segment as the grouping threshold. The effectiveness of the segmentation is evaluated by calculating the degree of preference. By iterating through all possible grouping thresholds, the optimal grouping point is found, thereby maximizing the segment's discriminability. By introducing an iterative mechanism to adjust the segment length and calculate the corresponding degree of preference, the most appropriate segment length can be identified.
[0016] Preferably, the degree of preference satisfies the relationship:
[0017] ,in, Indicates the segment length is Time The preference of the subgroup, and Respectively represent the segment length Time The mean value of the suspected noise probability of the two types of current data obtained by the sub-grouping, and Respectively represent the segment length Time The variance of the suspected noise probability of the two types of current data obtained by the sub-grouping.
[0018] By comparing the intra-class and inter-class differences between the current data of the two groups, the degree of dispersion of each category of current data is measured, ensuring that the groups not only have significant differences in mean but also have stability in variance. This improves the automation and accuracy of the segmentation and grouping process. By optimizing the grouping threshold, the ability to identify outliers in the current data series is enhanced, which helps to achieve more effective data denoising and quality control.
[0019] Preferably, the dividing the current data in the segment into two categories includes: dividing the current data in the segment that is less than or equal to the grouping threshold into one category of current data, and dividing the current data in the segment that is greater than the grouping threshold into another category of current data.
[0020] Preferably, the denoising of each segment to complete the operation error monitoring includes: using an iterative threshold segmentation algorithm to segment each segment, completing the identification of noise points in each segment, and using cubic spline interpolation to reassign the current data at the noise points to complete the denoising; after denoising, the accurate power of the electric energy meter is calculated based on the voltage data of the electric energy meter, and the operation error is calculated; comparing the operation error with a preset error threshold, and generating and sending an alarm signal in response to the operation error being greater than or equal to the preset error threshold.
[0021] Through cubic spline interpolation technology, the noise points are smoothed and reassigned, which restores the continuity of the data and maintains the smoothness and physical properties of the curve. The denoised current data and the voltage data of the electricity meter are used to evaluate the operating error of the electricity meter and realize real-time monitoring of the electricity meter performance.
[0022] In a second aspect, the present invention discloses an electric energy meter operation error monitoring system, comprising: a processor; and a memory, wherein the memory stores computer instructions. When the computer instructions are executed by the processor, the system executes the above-mentioned electric energy meter operation error monitoring method.
[0023] Beneficial effects of the present invention:
[0024] 1. This invention achieves a highly flexible and accurate monitoring method by segmenting the current data sequence into multiple segments and independently identifying noise points for each segment. The core advantage of this method is that it allows each segment to independently determine the optimal threshold based on its own data characteristics, thereby adapting to local variations and fluctuations in the data and significantly improving the accuracy and reliability of noise identification. By utilizing an iterative threshold segmentation algorithm, the threshold can be adaptively adjusted to cope with data of different distributions or different noise levels, ensuring effective noise identification and processing under various conditions.
[0025] 2. By calculating the meter's operating error in real time and comparing it with a preset error threshold, a highly efficient monitoring mechanism is implemented. When an error exceeds the threshold, an alarm signal is promptly generated and sent, providing real-time monitoring and a rapid response mechanism for meter operation and maintenance. This timely feedback mechanism not only helps quickly identify and resolve potential faults or anomalies, but also reduces energy and economic losses caused by errors, ensuring the economic efficiency and reliability of the power system.
[0026] 3. The introduction of cubic spline interpolation technology provides a smooth and continuous solution for denoising. This method not only effectively removes noise but also maintains the physical properties and continuity of the data, ensuring that the denoised data still accurately reflects the actual operating status of the electricity meter. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an illustrative and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:
[0028] Figure 1 The present invention provides a flow chart of an electric energy meter operation error monitoring method. DETAILED DESCRIPTION
[0029] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.
[0030] It should be understood that when the terms "first," "second," and the like are used in the claims, description, and drawings of the present invention, they are merely used to distinguish between different objects, rather than to describe a specific order. The terms "comprise" and "comprising" used in the description and claims of the present invention indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or combinations thereof.
[0031] The present invention provides a method for monitoring the operation error of an electric energy meter. Figure 1 As shown, a method for monitoring operation errors of an electric energy meter includes steps S1 to S4, which are described in detail below.
[0032] S1, obtaining a current data sequence according to a preset sampling interval.
[0033] It's important to note that single-phase energy meters are key devices used in power systems to measure and record single-phase AC energy consumption. They are widely used in homes, businesses, and industries. Using built-in current transformers, energy meters accurately measure the current in a circuit, a fundamental element of energy metering. Current transformers convert high currents into low currents, facilitating measurement and processing in energy meters.
[0034] In one embodiment, current data is collected through a current transformer. Exemplarily, the sampling interval is set to 5 seconds.
[0035] S2, calculates the suspected noise probability of the target data.
[0036] In one embodiment, any current data is used as target data, and a window constructed with the target data as the center and a preset length as the size is used as the radiation range of the target data. Exemplarily, the size of the radiation range is set to 7.
[0037] A preset reference number of current data is randomly selected as reference current data, and a radiation range of each reference current data is obtained, wherein the reference current data does not include the target data, and the window size of the radiation range of the reference current data is consistent with the window size of the radiation range of the target data. Exemplarily, the preset reference number is set to 20, and the specific setting can be made by those skilled in the art.
[0038] When the radiation range of the target data does not meet the preset length, the data contained in the window is retained. For example, the target data is the third current data, the window size is 6, there are two current data on the left side of the target data, and three current data on the right side. At this time, the window size of the radiation range of the reference current data is also 6, and it is guaranteed that any reference current data has two current data on the left side and three current data on the right side.
[0039] Calculate the suspected noise probability of the target data, and the suspected noise probability satisfies the relationship:
[0040] ,in, Indicates the The suspected noise possibility of the current data, Indicates the total number of reference current data, Indicates the The radiation range of the current data is Indicates the The radiation range of the reference current data, Indicates similarity, Represents an exponential function. In the relation of suspected noise probability, The current data is the target data.
[0041] It's important to note that noise points in current data sequences are often random, lacking a fixed pattern or location. Furthermore, within a certain range of noise points, they often appear uncorrelated with surrounding data points. Therefore, the less correlated a data point is with other data points within a certain range, the more likely it is a noise point. In a massive amount of current data, if the data changes near a data point are normal, similar changes will inevitably occur in other time periods.
[0042] In one embodiment, the suspected noise possibility calculation further includes: taking the ratio of the standard deviation to the mean of the current values in the radiation range of the target data as independence.
[0043] The probability of suspected noise satisfies the relationship:
[0044] ,in, Indicates the The suspected noise possibility of the current data, Indicates the The independence of the current data, Indicates the total number of reference current data, Indicates the The radiation range of the current data is Indicates the The radiation range of the reference current data, Indicates similarity, Represents an exponential function. In the relation of suspected noise probability, The current data is the target data.
[0045] S3, segmenting the current data sequence based on the likelihood of suspected noise.
[0046] In one embodiment, the left endpoint or the right endpoint of the current data sequence is used as the starting point, and a preset number of current data is used as the segment length to obtain the first segment. The minimum value of the current data in the segment is used as the grouping threshold, and the current data in the segment is divided into two categories. The preference degree of the segment is calculated, and the preference degree satisfies the relationship:
[0047] ,in, Indicates the segment length is Time The preference of the subgroup, and Respectively represent the segment length Time The mean value of the suspected noise probability of the two types of current data obtained by the sub-grouping, and Respectively represent the segment length Time The variance of the suspected noise probability of the two types of current data obtained by the sub-grouping.
[0048] Can represent the gap between classes, and and It represents the intra-class gap.
[0049] The grouping threshold is traversed from the minimum value to the maximum value to obtain the preference degree of each grouping, and the maximum preference degree is used as the actual preference degree of the current segment length.
[0050] The segment lengths are iterated according to a preset step size, and the actual preference level of the first segment at each segment length is calculated. For example, the iteration step size is set to 3, meaning that three current data points are added to the segment at a time. If the actual preference level of the first segment length decreases twice during the iteration process, the segment length before the decrease is used as the final segment length of the first segment.
[0051] For example, the length of the current data sequence is 100. The left endpoint of the current data sequence is used as the starting point, and the 15 current data in chronological order are used as the first segment. The grouping threshold is traversed from the minimum value to the maximum value of the 15 current data. The current data in the segment that is less than or equal to the grouping threshold is regarded as one type of current data, and the current data in the segment that is greater than the grouping threshold is regarded as another type of current data. The preference degree after each grouping is calculated, and the maximum preference degree is selected as the actual preference degree of the first segment when the segment length is 15. The segment degree is then updated according to the preset iterative step size, that is, the segment length is changed to 18, and the actual preference degree of the first segment when the segment length is 18 is obtained. The actual preference degree of the first segment is calculated when the segment length is 21. If the actual preference degree decreases continuously, the final segment length of the first segment is 15, and the second segment is constructed with the 16th data as the starting point.
[0052] At this point, the current data sequence can be segmented.
[0053] In one embodiment, the current data series is segmented using a Fisher optimal solution.
[0054] S4, denoise each segment to complete the running error monitoring.
[0055] In one embodiment, based on the segmentation result of step S3, an iterative threshold segmentation algorithm is used to segment each segment, noise points in each segment are identified, and the current data at the noise points are reassigned using cubic spline interpolation, thereby achieving the purpose of denoising.
[0056] After denoising, the accurate power of the electric energy meter can be calculated based on the voltage data of the electric energy meter. Then the operating error of the electric energy meter is calculated. The operating error satisfies the relationship:
[0057] ,in, Indicates the running error, Indicates rated power, Indicates actual power.
[0058] Compare the operating error with the preset error threshold. When the operating error is greater than or equal to the preset error threshold, it means that the current operating error of the electricity meter is too large, and an alarm signal is generated and sent to complete the operating error monitoring.
[0059] An embodiment of the present invention further discloses an electric energy meter operation error monitoring system, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an electric energy meter operation error monitoring method according to the present invention is implemented.
[0060] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.
[0061] In the present invention, the aforementioned memory can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic storage medium or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium can be part of, accessible to, or connected to the device.
[0062] While this specification has shown and described several embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Numerous modifications, variations, and alternatives will occur to those skilled in the art without departing from the concept and spirit of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in practicing the present invention.
[0063] The above are all preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the scope of protection of the present invention.
Claims
1. A method for monitoring the operation error of an electric energy meter, It is characterized in that include: Acquire a current data sequence according to a preset sampling interval, take any current data as target data, and take a window constructed with the target data as the center and a preset length as the size as the radiation range of the target data; Randomly selecting a preset reference number of current data as reference current data, and obtaining a radiation range of each reference current data, wherein the reference current data does not include target data, and a window size of the radiation range of the reference current data is consistent with a window size of the radiation range of the target data; The suspected noise probability of the target data is calculated, and the suspected noise probability satisfies the relationship: ,in, Indicates The possibility of suspected noise in the current data, Indicates the total number of reference current data, Indicates The radiation range of the current data is Indicates The radiation range of the reference current data is Indicates similarity, represents an exponential function. In the relation of the suspected noise probability, The current data is the target data; Segment the current data series based on the likelihood of suspected noise, including: Taking the left end point or the right end point of the current data sequence as the starting point and taking a preset number of current data as the segment length, the first segment is obtained; The minimum value of the current data in the segment is used as the grouping threshold, and the current data in the segment is divided into two categories, including: The current data in the segment that is less than or equal to the grouping threshold is regarded as one type of current data, and the current data in the segment that is greater than the grouping threshold is regarded as another type of current data; Calculate the segment's preference to satisfy the relationship: ,in, Indicates the segment length is Time The preference of the subgroup, and Respectively represent the segment length Time The mean value of the suspected noise probability of the two types of current data obtained by the grouping, and Respectively represent the segment length Time The variance of the suspected noise probability of the two types of current data obtained by the sub-grouping; Traverse the grouping threshold from the minimum value to the maximum value to obtain the preference degree of each grouping, and take the maximum preference degree as the actual preference degree of the current segment length; Iterate the segment length according to the preset step length, calculate the actual preference of the first segment at each segment length, and when the segment length of the first segment decreases twice in succession during the iteration process, take the segment length before the decrease as the final segment length of the first segment; Each segment is denoised to complete the running error monitoring.
2. A method for monitoring operation errors of an electric energy meter according to claim 1, It is characterized in that The window size of the radiation range of the reference current data is consistent with the window size of the radiation range of the target data.
3. The method for monitoring the operation error of an electric energy meter according to claim 1, It is characterized in that The suspected noise possibilities also include: The ratio of the standard deviation to the mean of the current value in the radiation range of the target data is taken as indifference; The probability of suspected noise satisfies the relationship: ,in, Indicates The possibility of suspected noise in the current data, Indicates The independence of the current data, Indicates the total number of reference current data, Indicates The radiation range of the current data is Indicates The radiation range of the reference current data is Indicates similarity, represents an exponential function. In the relation of the suspected noise probability, The current data is the target data.
4. The method for monitoring the operation error of an electric energy meter according to claim 1, It is characterized in that The denoising of each segment to complete the operation error monitoring includes: Each segment is segmented using an iterative threshold segmentation algorithm to identify the noise points in each segment, and the current data at the noise points is reassigned using cubic spline interpolation to complete denoising; After denoising, the accurate power of the electric energy meter is calculated based on the voltage data of the electric energy meter, and the operation error is calculated; The operating error is compared with a preset error threshold, and in response to the operating error being greater than or equal to the preset error threshold, an alarm signal is generated and sent.
5. An electric energy meter operation error monitoring system, It is characterized in that include: Processor; and A memory storing computer instructions, wherein when the computer instructions are executed by a processor, the system executes an electric energy meter operation error monitoring method according to any one of claims 1 to 4.
Citation Information
Patent Citations
Electric energy meter metering error monitoring method based on energy controller
CN116859321A
Wind farm output power data restoration method considering segmental characteristics of wind speed data
CN106570790A
Abnormal data detection method and device, electronic equipment and computer storage medium
CN113435464A
Cited By
96-point electric quantity multi-source fusion fitting and anomaly correction method for new energy power plant
CN121543023A
