Method and device for evaluating metering state of electric energy meter

Abstract

The invention discloses a method and equipment for evaluating the metering state of an electric energy meter. The method comprises the following steps: determining a technical line loss value of a to-be-counted object in a line loss statistical period; classifying the initial line loss samples according to the technical line loss values to obtain classified line loss samples; removing abnormal sample data in the classified line loss samples to obtain target line loss samples; based on the electric quantity level of each user in the target line loss sample and the electric quantity curve correlation, carrying out user merging to obtain a merged target user; determining a model regression coefficient by adopting the multiple linear regression model for counting the line loss, the electric quantity related parameters in the target line loss sample and the actual electricity consumption of each target user; and analyzing the metering state of the electric energy meter corresponding to each user by adopting an anomaly detection criterion and a model regression coefficient to obtain an abnormal electric energy meter. According to the invention, effective detection of the electric energy meter with an abnormal metering state is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply, and in particular to a method and device for evaluating the measurement status of an electric energy meter. Background Art

[0002] As the operation time increases, metering equipment failures and out-of-tolerances often occur due to factors such as aging of metering equipment, poor operating environment, over-capacity operation, and power theft by users. Traditional electronic energy meters do not have the function of self-monitoring metering errors. Since it is impossible to monitor the metering status of the energy meter online, faulty and out-of-tolerance metering equipment and power theft by users cannot be discovered and handled in a timely manner for a long time, resulting in a large number of power errors and power losses, making it difficult to refund and compensate for power, and potential marketing risks. Summary of the invention

[0003] The present invention provides a method and a device for evaluating the metering state of an electric energy meter, which can effectively detect an electric energy meter with an abnormal metering state.

[0004] According to one aspect of the present invention, there is provided a method for evaluating the measurement status of an electric energy meter, comprising:

[0005] Determine the technical line loss value of the object to be counted within the line loss statistical period;

[0006] Classifying and eliminating initial line loss samples with associated statistical line loss values ​​according to the technical line loss values ​​to obtain classified line loss samples;

[0007] Using a data outlier detection method to remove abnormal sample data from the classified line loss samples to obtain target line loss samples;

[0008] If the number of samples of the target line loss sample does not satisfy the preset sample margin set for the number of users included in the statistical object, users are merged based on the power level and power curve correlation of each user in the target line loss sample to obtain the merged target user;

[0009] The corresponding model regression coefficient is determined by using a pre-created multivariate linear regression model of statistical line loss, power-related parameters in the target line loss sample, and actual power consumption of each target user;

[0010] The abnormal detection criterion and the model regression coefficient are used to analyze the metering state of the electric energy meter corresponding to each user to obtain the abnormal electric energy meter.

[0011] According to another aspect of the present invention, there is provided an electric energy meter measurement state evaluation device, comprising:

[0012] A line loss value determination module, used to determine the technical line loss value of the object to be counted within the line loss statistical period;

[0013] A classification and elimination module, used for classifying and eliminating initial line loss samples with associated statistical line loss values ​​according to the technical line loss values, to obtain classified line loss samples;

[0014] A detection and elimination module is used to eliminate abnormal sample data in the classified line loss samples by using a data outlier detection method to obtain a target line loss sample;

[0015] A user merging module, configured to merge users based on the power level and power curve correlation of each user in the target line loss sample to obtain a merged target user if the sample number of the target line loss sample does not meet a preset sample margin set for the number of users included in the statistical object;

[0016] A coefficient determination module, used to determine the corresponding model regression coefficient using a pre-created multivariate linear regression model of statistical line loss and power-related parameters in the target line loss sample and the actual power consumption of each target user;

[0017] The state analysis module is used to analyze the metering state of the electric energy meter corresponding to each user by using the abnormal detection criterion and the model regression coefficient to obtain the abnormal electric energy meter.

[0018] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0019] at least one processor; and

[0020] a memory communicatively connected to the at least one processor; wherein,

[0021] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for evaluating the measurement status of an electric energy meter described in any embodiment of the present invention.

[0022] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for evaluating the measurement status of an electric energy meter described in any embodiment of the present invention when executed.

[0023] According to another aspect of the present invention, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, the method for evaluating the metering state of an electric energy meter according to any embodiment of the present invention is implemented.

[0024] The technical solution of the embodiment of the present invention ensures that the attributes of the line loss samples are consistent by classifying the initial line loss samples of the associated statistical line loss values ​​according to the determined technical line loss values; and adopts a data anomaly detection method to eliminate abnormal sample data in the classified line loss samples, thereby reducing the influence of unstable independent variables on the evaluation results; and combines and solves the independent variables based on the principles of similar electricity levels and the highest correlation with electricity curves, thereby achieving the dimensionality reduction effect of the independent variables of the multivariate linear regression model; and, adopts an anomaly detection criterion to detect significant regression coefficients to determine the electric energy meters with abnormal metering states, thereby realizing effective detection of the electric energy meters with abnormal metering states.

[0025] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0027] Figure 1 It is a flow chart of a method for evaluating the measurement status of an electric energy meter provided by an embodiment of the present invention;

[0028] Figure 2 is a flow chart of another method for evaluating the measurement status of an electric energy meter provided by an embodiment of the present invention;

[0029] Figure 3 It is a structural schematic diagram of an electric energy meter measurement state evaluation device provided by an embodiment of the present invention;

[0030] Figure 4 It is a structural block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0033] In one embodiment, Figure 1 This is a flow chart of a method for evaluating the metering status of an electric energy meter provided by an embodiment of the present invention. This embodiment is applicable to the case of online automatic evaluation of the metering status of an electric energy meter. The method can be executed by an electric energy meter metering status evaluation device. The electric energy meter metering status evaluation device can be implemented in the form of hardware and / or software. The electric energy meter metering status evaluation device can be configured in an electronic device. For example, the electronic device can be a terminal device such as a tablet computer and a personal computer. Figure 1 As shown, the method includes:

[0034] S110, determining the technical line loss value of the object to be counted within the line loss statistical period.

[0035] In one example, the object to be counted refers to a power supply and use unit that needs to perform line loss statistics. For example, the object to be counted can be a unit such as a substation and a line. Generally speaking, a single object to be counted can include hundreds of electric meters, that is, hundreds of users. The line loss statistical period refers to a time period for statistically analyzing the line loss value of the object to be counted. The line loss statistical period can be calculated in hours, days, and weeks. For example, the line loss statistical period can be 1 day. The technical line loss value refers to the power loss caused by the physical characteristics of the power supply equipment itself during the transmission and distribution of power. In one example, the technical line loss value can include: fixed line loss and variable line loss. Among them, the fixed line loss can also be called no-load loss, which refers to the loss generated by the power equipment when there is no load current passing through, that is, the fixed line loss is a constant value and does not change significantly with the change of load current; the variable line loss can also be called load loss, which refers to the loss generated by the power equipment when there is load current passing through, that is, the variable loss will change with the change of load current and power factor.

[0036] In one example, based on the line loss calculation method of voltage lossless conduction theory, the technical line loss value of the statistical object within a line loss statistical period can be statistically analyzed.

[0037] S120 , classifying and eliminating initial line loss samples of associated statistical line loss values ​​according to technical line loss values ​​to obtain classified line loss samples.

[0038] The statistical line loss value refers to the amount of electricity lost in the process of the input of electric energy to the statistical object to be counted and the energy consumption of the user; the statistical line loss value can be the difference between the total power supply and the total power consumption. The initial line loss sample refers to the power-related parameters associated with a statistical line loss value based on a pre-created multivariate linear regression model of statistical line loss. For example, the initial line loss sample may include the following variables: actual line loss statistical value y, fixed loss x b , the number of users n, the power consumption of n users u n , user regression coefficient x corresponding to n users n , total power supply S, total power supply regression coefficient x s , variable line loss w t , variable line loss regression coefficient x t Among them, x n Used to represent the attributes and weights of the user's power consumption in the online loss; x t Used to indicate the degree of adjustment to the technical line loss value.

[0039] The classified line loss sample refers to a line loss anomaly type corresponding to the abnormal line loss sample in the initial line loss sample. For example, the line loss anomaly type of the classified line loss sample may include: a high-loss anomaly sample and a low-loss anomaly sample.

[0040] In an embodiment, the classification upper and lower thresholds of the initial line loss samples can be determined in a preconfigured manner, and the initial line loss samples whose statistical line loss values ​​are between the lower and upper thresholds are taken as normal line loss samples, and the normal line loss samples are eliminated from the initial line loss samples to obtain abnormal line loss samples in the initial line loss samples; then the initial line loss samples whose statistical line loss values ​​are less than the lower threshold are taken as low-loss abnormal samples, and the initial line loss samples whose statistical line loss values ​​are greater than the upper threshold are taken as high-loss abnormal samples; then the sample number of low-loss abnormal samples and the sample number of high-loss abnormal samples are counted, and if the sample number of high-loss abnormal samples is greater than the sample number of low-loss abnormal samples, the low-loss abnormal samples are eliminated from the initial line loss samples, and the high-loss abnormal samples are taken as classified line loss samples; if the sample number of high-loss abnormal samples is less than the sample number of low-loss abnormal samples, the high-loss abnormal samples are eliminated from the initial line loss samples, and the low-loss abnormal samples are taken as classified line loss samples.

[0041] S130: Using a data outlier detection method to remove abnormal sample data from the classified line loss samples to obtain target line loss samples.

[0042] In one example, a data anomaly detection method is used to identify an algorithm for identifying data points in a classified line loss sample that do not meet expectations or show obvious differences. For example, the data anomaly detection method may be a quartile method or a standard deviation method. Exemplarily, the process of removing abnormal sample data is described using a data anomaly detection method. Specifically, the quartile method is used to determine the abnormal upper boundary value and the abnormal lower boundary value, and then the abnormal sample data whose statistical line loss rate is greater than the abnormal upper boundary value and the abnormal sample data whose statistical line loss rate is less than the abnormal lower boundary value are removed from the classified line loss sample to obtain a target line loss sample.

[0043] In one example, the number of initial line loss samples is determined based on the number of users included in the object to be counted and a preset sample margin. In order to better extract sample features, the number of initial line loss samples needs to be greater than the number of users included in the object to be counted. If a sufficient number of samples cannot be obtained, the maximum value of the actually obtainable number of samples is taken.

[0044] S140: If the number of samples of the target line loss sample does not satisfy the preset sample margin set for the number of users included in the statistical object, merge the users based on the power level of each user in the target line loss sample and the correlation between the power curve to obtain the merged target user.

[0045] Among them, the power level is used to characterize the power consumption of a user; the power curve is used to characterize the power change curve of a user in a statistical time period; the power curve correlation is used to characterize the consistency of the change of the power curves of two users. In one example, the users whose power consumption in the whole period is 0 in the target line loss sample can be eliminated, and then the power curve correlation coefficient between each user's power curve and the power curve of other single users is calculated, and the power curve correlation between each user's power curve and the power curve of other single users is determined based on the power curve correlation coefficient, and another user with the highest correlation with the power curve of one user is found, and two users whose correlation coefficients reach the correlation coefficient threshold are merged into one user group. If the number of users composed of the user group and the unmerged single users does not meet the preset sample margin, two unmerged single users with similar power levels are merged to obtain the merged target user.

[0046] In one example, the preset sample margin is used to represent the number of samples that needs to be greater than the number of variables. For example, if the number of variables is 10 and the preset sample margin is 4, the number of samples needs to be 14.

[0047] S150: Determine corresponding model regression coefficients using a pre-created multivariate linear regression model of statistical line loss, power-related parameters in the target line loss sample, and actual power consumption of each target user.

[0048] In one example, the model regression coefficient refers to the regression coefficient required to correspond to each variable in the multivariate linear regression model of statistical line loss. For example, the model regression coefficient may include: user regression coefficient x n , total power supply regression coefficient x s and variable line loss regression coefficient x t .

[0049] In one example, the total power supply in the target line loss sample and the actual power consumption of each target user can be obtained, and the sum of the actual power consumption of all target users in the target line loss sample can be determined as the total actual power consumption; then the difference between the total power supply and the total actual power consumption is used as the actual value of the statistical line loss, and then the actual value of the statistical line loss, the actual power consumption and the technical line loss value are input into the pre-constructed multivariate linear regression model solution formula of the statistical line loss to obtain the corresponding model regression coefficient.

[0050] S160, using anomaly detection criteria and model regression coefficients to analyze the metering status of the electric energy meter corresponding to each user to obtain an abnormal electric energy meter.

[0051] In one example, the model regression coefficient refers to the regression coefficient involved in the pre-created multivariate linear regression model of statistical line loss, for example, the model regression coefficient may include the user regression coefficient. The anomaly detection criterion refers to the criterion for anomaly detection of the user regression coefficient, for example, the anomaly detection criterion may be the 3σ criterion. In an embodiment, the solution value of the user regression coefficient is classified into positive and negative categories according to positive and negative values, that is, the user regression coefficient is divided into two categories: positive and negative values; then, anomaly detection is used to accurately and cyclically detect the significantly different regression coefficients in a group of data consisting of positive values, and the anomaly detection criterion is used to cyclically detect the significantly different regression coefficients in a group of data consisting of negative values, then the significantly different regression coefficient is the abnormal user regression coefficient, the electric energy meter corresponding to the abnormal user regression coefficient is the abnormal electric energy meter, and the corresponding user is the abnormal user.

[0052] The technical solution of this embodiment ensures that the attributes of the line loss samples are consistent by classifying the initial line loss samples of the associated statistical line loss values ​​according to the determined technical line loss values; and adopts a data anomaly detection method to eliminate abnormal sample data in the classified line loss samples, thereby reducing the impact of unstable independent variables on the evaluation results; and combines and solves the independent variables based on the principles of similar electricity levels and the highest correlation with electricity curves, thereby achieving the dimensionality reduction effect of the independent variables of the multivariate linear regression model; and, adopts anomaly detection criteria to detect significant regression coefficients to determine the electricity meters with abnormal metering states, thereby realizing effective detection of the electricity meters with abnormal metering states.

[0053] In one embodiment, Figure 2This is a flowchart of another method for evaluating the metering status of an electric energy meter provided by an embodiment of the present invention. This embodiment is based on the above embodiment and further explains the process of determining the technical line loss value, the process of screening the classified line loss samples, the process of screening the target line loss samples, the process of user merging, the process of determining the model regression coefficient, and the process of screening the abnormal electric energy meter. Figure 2 As shown, the method includes:

[0054] S210: Acquire the actual active power, user voltage, and gateway power supply voltage of each user in the statistical object in each line loss statistical sub-period.

[0055] Among them, the line loss statistical sub-period is a subset of the line loss statistical period, that is, the line loss statistical period is divided into multiple sub-segments, and each sub-segment is used as the line loss statistical sub-period; the actual active power refers to the power corresponding to the active power actually consumed by the user during the power consumption process; the user voltage refers to the actual voltage value transmitted to the user; the gateway power supply voltage refers to the voltage value of the metering gateway at the input of the electric energy to be counted, for example, the gateway power supply voltage can be the voltage value at the outgoing line side of the substation or the incoming line side of the user's distribution room.

[0056] In one example, the user voltage can be obtained using an energy meter or a multifunctional power detector, or an oscilloscope or a voltmeter. In one example, the gateway power supply voltage can be obtained from a substation monitoring system or using a gateway metering device. In one example, the actual active power can be obtained through an energy meter or from a power supply management system.

[0057] S220 , determining the theoretical active power of the associated user within the line loss statistical sub-period according to the actual active power, the user voltage, and the gateway power supply voltage.

[0058] In one example, assuming that in the t-th Δt period, the theoretical active power of the i-th user is recorded as ΔW ti ; Correspondingly, the calculation formula for theoretical active power can be: ΔW ti =ΔW' ti *U t0 / U ti Among them, U ti is the user voltage of the i-th user in the t-th Δt period, U t0 is the gate supply voltage during the tth Δt period, and ΔW' ti is the actual active power of the i-th user in the t-th Δt period.

[0059] S230. Determine the theoretical power of the associated user in the line loss statistical period according to the theoretical active power of each user in each line loss statistical sub-period.

[0060] In one example, the theoretical power of a user in a line loss statistical period can be the sum of the theoretical active power in each line loss statistical sub-period. Exemplarily, assuming that the line loss statistical period is divided into N line loss statistical sub-periods, the formula for the theoretical power of a user can be: Among them, W i is the theoretical power consumption of the i-th user, N is the number of line loss statistical sub-periods contained in the line loss statistical period, ΔW ti is the theoretical active power of the i-th user in the t-th Δt period.

[0061] S240 , determining the theoretical power loss of each user in the online loss statistical period according to the theoretical power consumption and the measured power consumption of each user in the online loss statistical period.

[0062] The measured power refers to the power consumption obtained by actually monitoring the power consumption of each user during the online loss statistical period; the theoretical power loss refers to the power value theoretically lost by each user during the online loss statistical period. In an example, the difference between the theoretical power consumption and the measured power consumption of each user during the online loss statistical period can be used as the theoretical power loss of each user during the online loss statistical period. For example, the calculation formula for the theoretical power loss of the i-th user during the online loss statistical period can be: W iloss =W i -W' i ; Among them, W iloss represents the theoretical power loss of the i-th user during the online loss statistical period; W i represents the theoretical power consumption of the i-th user in the online loss statistical period; W' i It represents the measured power consumption of the i-th user during the online loss statistical period.

[0063] S250: Determine the technical line loss value of the object to be counted within the line loss statistical period according to the theoretical power loss of each user and the number of users included in the object to be counted.

[0064] In one example, the theoretical power loss of each user included in the object to be counted can be added together to obtain the technical line loss of the object to be counted within the line loss statistical period. Exemplarily, the calculation formula for the technical line loss value of the object to be counted within the line loss statistical period can be: Among them, W loss represents the variable line loss in the technical line loss value of the object to be counted within the line loss statistical period; n represents the number of users included in the object to be counted; W iloss represents the theoretical power loss of the ith user during the online loss statistical period. loss With w t Equivalent.

[0065] In one example, S210-S250 is a technical line loss statistical method that uses a voltage lossless conduction theoretical line loss calculation method to count the technical line loss value. This technical line loss statistical method is based on the lossless conduction of the power supply voltage to the user side, and the theoretical power of the user under lossless voltage is calculated from the perspective of metering. The difference between the theoretical power of the user and the measured power is the theoretical power loss of the user. The total theoretical power loss of the user is accumulated to obtain the theoretical power loss of the object to be counted as the technical line loss value.

[0066] S260. Determine the upper and lower thresholds for the initial line loss sample classification based on the technical line loss value, the power consumption of all electric energy meters in the statistical object, the power consumption of related equipment not included in the metering, and the accuracy level of the electric energy meter.

[0067] Among them, the total amount of electricity meter power consumption refers to the electricity consumed when the electricity meter is running; the power consumption of the associated equipment that is not included in the metering refers to the fixed loss; the threshold basic value can be the sum of the theoretical loss power calculated and the quantifiable fixed loss of the equipment. In one example, the upper limit threshold of the initial line loss sample classification can be the uncertainty caused by the sum of the threshold basic value and the accuracy level of the electricity meter; the lower limit threshold can be the uncertainty caused by the difference between the threshold basic value and the accuracy level of the electricity meter.

[0068] S270, screening out high-loss abnormal samples from the initial line loss samples according to the upper threshold, and screening out low-loss abnormal samples from the initial line loss samples according to the lower threshold.

[0069] In one example, the initial line loss samples whose statistical line loss values ​​are between the lower threshold and the upper threshold are normal line loss samples; the initial line loss samples whose statistical line loss values ​​exceed the upper threshold are high-loss abnormal samples; and the initial line loss samples whose statistical line loss values ​​are lower than the lower threshold are low-loss abnormal samples. In an embodiment, the normal line loss samples in the initial line loss samples are removed to obtain abnormal line loss samples; then, the line loss samples in the abnormal line loss samples that are greater than the upper threshold are regarded as high-loss abnormal samples, and the line loss samples that are less than the lower threshold are regarded as low-loss abnormal samples.

[0070] S280. If the number of high-loss abnormal samples included in the initial line loss samples is greater than the number of low-loss abnormal samples, the low-loss abnormal samples are removed from the initial line loss samples, and the high-loss abnormal samples are used as classified line loss samples.

[0071] The sample number of high-loss abnormal samples and the sample number of low-loss abnormal samples contained in the initial line loss samples are compared, and the abnormal samples with a larger sample number are retained. For example, if the sample number of high-loss abnormal samples is greater than the sample number of low-loss abnormal samples, the low-loss abnormal samples are removed from the initial line loss samples, and the high-loss abnormal samples are used as classified line loss samples.

[0072] S290. If the number of high-loss abnormal samples included in the initial line loss samples is less than the number of low-loss abnormal samples, the high-loss abnormal samples are removed from the initial line loss samples, and the low-loss abnormal samples are used as classified line loss samples.

[0073] In one example, the number of high-loss abnormal samples and the number of low-loss abnormal samples included in the initial line loss samples are compared, and the abnormal samples with a larger number of samples are retained. For example, if the number of high-loss abnormal samples is less than the number of low-loss abnormal samples, the high-loss abnormal samples are removed from the initial line loss samples, and the low-loss abnormal samples are used as classified line loss samples.

[0074] S2100. Determine a corresponding statistical line loss rate according to the statistical line loss power and the total power supply in the classified line loss sample.

[0075] In one example, the statistical line loss rate is used to characterize the operating efficiency and management level of the power system. Generally speaking, the statistical line loss rate refers to the percentage of statistical line loss electricity to total power supply within a certain period of time. The statistical line loss electricity is the difference between the total power supply and the total actual power consumption; the statistical line loss rate is the percentage of the ratio between the statistical line loss electricity and the total power supply. Among them, the total actual power consumption is the sum of the actual power consumption of all users included in the classified line loss sample. It should be noted that the statistical line loss electricity can also be called the actual value of the statistical line loss, or the statistical line loss value.

[0076] S2110. Use the quartile method and statistical line loss rate to determine the corresponding abnormal upper boundary value and abnormal lower boundary value.

[0077] In one example, the statistical line loss rate corresponding to each classified line loss sample is arranged in order from small to large; then the quartile positions, i.e., the positions of the lower quartile, the median, and the upper quartile, are determined from the arranged statistical line loss rates; then, since the positions of the quartiles are usually not integers, the specific quartile values ​​can be determined by interpolation; then the interquartile range is determined based on the difference between the upper quartile and the lower quartile; then the abnormal lower boundary value is determined based on the lower quartile and the interquartile range, and the abnormal upper boundary value is determined based on the upper quartile and the interquartile range.

[0078] S2120, remove the abnormal sample data whose statistical line loss rate is greater than the abnormal upper boundary value and the abnormal sample data whose statistical line loss rate is less than the abnormal lower boundary value from the classified line loss samples to obtain the target line loss samples.

[0079] The statistical line loss rate corresponding to each classified line loss sample is compared with the abnormal upper boundary value and the abnormal lower boundary value, and the abnormal sample data with a statistical line loss rate greater than the abnormal upper boundary value is removed from the classified line loss samples, and the abnormal sample data with a statistical line loss rate less than the abnormal lower boundary value is removed from the classified line loss samples to obtain the target line loss sample.

[0080] S2130: Determine the correlation between the power curve of each user in the target line loss sample and the power curves of other single users.

[0081] In one example, the power curve of each user in the target line loss sample can be plotted in the same coordinate system as the power curves of other single users, and the trends and changing trends of the two curves can be intuitively analyzed. If the trends of the power curves corresponding to the two users are relatively similar, then the power curves representing the two users have a certain correlation.

[0082] S2140 , searching for another user with the highest correlation with a power curve of a user and the corresponding correlation coefficient.

[0083] In one example, if the power curves of two users are linearly related, the Pearson correlation coefficient may be selected for calculation; if the curve relationship is more complex, the Spearman correlation coefficient may be more suitable. In an embodiment, the correlation coefficient between the power curves of two users may be calculated according to the selected correlation coefficient calculation method; if the correlation coefficient ranges from -1 to 1, and the closer the absolute value is to 1, the stronger the correlation is; conversely, the closer the absolute value is to 0, the weaker the correlation is.

[0084] In an embodiment, another user having the highest correlation with a power curve of a user is found, and a correlation coefficient between the power curve of the user and the power curve of another user is determined.

[0085] S2150: Merge two users whose correlation coefficients reach a correlation coefficient threshold into one user group.

[0086] The correlation coefficient threshold is used to indicate whether to merge two users. If the correlation coefficient reaches the correlation coefficient threshold, it can be considered that the power curves of the two users are similar and can be directly merged into one power curve, and the two users can be regarded as one user group.

[0087] The purpose of merging users according to the correlation of electricity curves is to reduce the multicollinearity problem of independent variables and achieve dimensionality reduction of model independent variables. Merging may cause the actual abnormal independent variable characteristics to fail to be presented and ultimately fail to be detected. Generally speaking, users with high correlation of electricity curves have high similarity in electricity usage characteristics. Merging two users with high similarity in electricity usage characteristics can ensure that the overall characteristics are maintained after the merger, thereby ensuring the detection of abnormal independent variables.

[0088] S2160: If the number of users consisting of the user group and the unmerged single users does not meet the preset sample margin, merge the users according to the power levels of the unmerged single users to obtain a merged target user.

[0089] Among them, the electricity measured by the electricity meters with abnormal metering status is mostly at a medium to low level. Users with low correlation are merged according to similar load levels to avoid the characteristics of abnormal low-power users being masked and unable to be detected after being merged with users with high power load.

[0090] S2170. Obtain the total power supply in the target line loss sample and the actual power consumption of each target user.

[0091] Among them, the total power supply refers to the total amount of electric energy supplied to the object to be counted during the line loss statistical period; the actual power consumption refers to the amount of electric energy consumed by each target user in the target line loss sample during the line loss statistical period.

[0092] S2180: Determine the actual value of the statistical line loss according to the total power supply and the actual power consumption of each target user.

[0093] In one example, the actual power consumption of all target users in the target line loss sample is counted to obtain the total actual power consumption; then the difference between the total power supply and the total actual power consumption is used as the actual value of the statistical line loss. For example, it is assumed that the calculation formula for the actual value of the statistical line loss can be: Where y represents the actual value of statistical line loss; S represents the total power supply; u i represents the actual power consumption of the i-th user; n represents the total number of target users included in the target line loss sample.

[0094] S2190. Input the actual value of statistical line loss, actual power consumption and technical line loss value into a pre-constructed multivariate linear regression model solution formula of statistical line loss to obtain the corresponding model regression coefficient.

[0095] In one example, the solution formula of the multivariate linear regression model for statistical line loss is: y = x b +u1*x1+u2*x2+u3*x3+…+u n *x n +S*x s +w t*x t ; where y represents the actual value of statistical line loss; S represents the total power supply; u n represents the actual power consumption of the nth user; x b represents fixed loss; x n represents the user regression coefficient corresponding to n users; w t Indicates variable line loss; x t Represents the variable line loss regression coefficient.

[0096] The target users after the above processing (including the merged user group and the unmerged single user) are solved according to the multivariate linear regression model of the above statistical line loss; then the multivariate linear regression model is solved by the least square method, as shown in the following formula: X = (U T U) -1 U T Y; can be equivalent to the following formula:

[0097]

[0098] Among them, X is a column vector, which is the regression coefficient x b 、x n 、x s and x t The solution result; U is composed of constant term, independent variable u n , S, w t The parameter matrix composed of samples, the constant term x b The coefficient of is 1, and the independent variable u n is the actual power consumption of each user (for example, u 11 Refers to the actual electricity consumption of the first user on the first day, u 1n refers to the actual power consumption of the nth user on the first day), the independent variable S is the total power supply of the object to be counted, and the independent variable w t is the technical line loss value; Y is a column vector consisting of power loss; m is the number of target line loss samples. If the line loss statistical cycle is 1 day, m can also be understood as the number of days. m1 Refers to the actual power consumption of the first user in the mth day, u mn Refers to the actual electricity consumption of the nth user in the mth day.

[0099] In an embodiment, the The calculated y value, as well as u1, u2, ... n , S and w t Substitute the solution formula of the multivariate linear regression model of statistical line loss into X=(U T U) -1 U T Y can get x b, x1, x2, ... x n 、x s and x t .

[0100] S2200 , classify the user regression coefficients according to positive and negative values ​​to obtain first-category user regression coefficients and second-category user regression coefficients.

[0101] In one example, the regression coefficient of the first type of user and the regression coefficient of the second type of user are different in sign. For example, if the regression coefficient of the first type of user is a value greater than 0, the regression coefficient of the second type of user is correspondingly a value less than 0. In one example, the solution value of the user regression coefficient is divided into two categories according to the positive and negative values, that is, it is determined whether the solution value of the user regression coefficient is greater than 0. If it is greater than 0, it is classified as the regression coefficient of the first type of user; if it is less than 0, it is classified as the regression coefficient of the second type of user.

[0102] S2210, using the Laida criterion to perform cyclic detection on a group of data consisting of the regression coefficients of the first type of users and a group of data consisting of the regression coefficients of the second type of users, and screen out abnormal user regression coefficients.

[0103] In one example, the Laida criterion is referred to as the 3σ criterion. The 3σ criterion is used to detect the user regression coefficient x as an independent variable in a set of data consisting of the first type of user regression coefficients. n After the outlier is found, the outlier in the set of data is removed and tested again until no outlier is found in the regression coefficient of the first type of user. Similarly, the 3σ criterion is used to detect the user regression coefficient x as the independent variable in the set of data consisting of the regression coefficient of the second type of user. n After finding the outlier in the data, remove the outlier and re-test the data until no outlier is detected in the regression coefficient of the second category of users.

[0104] S2220. The electric energy meter of the user associated with the abnormal user regression coefficient is taken as the abnormal electric energy meter.

[0105] In one example, an abnormal electric energy meter with a positive error metering state is output in a low-loss abnormal sample; and an abnormal electric energy meter with a negative error metering state is output in a high-loss abnormal sample.

[0106] In one embodiment, when the abnormal user corresponding to the abnormal electric energy meter is an abnormal user group, it also includes: treating each user in the abnormal user group as a single user, and returning to the step of determining the technical line loss value of the object to be counted within the line loss statistical period, and screening out the abnormal users in the abnormal user group. In the case where the abnormal user is an abnormal user group, it is impossible to clearly identify the abnormal user in the abnormal user group. At this time, the abnormal user group is decomposed into single users and re-solved. The single abnormal user detected in the first solution and the single user decomposed into the abnormal user group are treated as single users in the second solution, that is, they are not merged into a user group with other users, and the data processing and solution method are the same as above. The electric energy meter that is detected as abnormal in both solution results is confirmed as an electric energy meter with an abnormal metering state.

[0107] In one embodiment, the method for evaluating the metering status of an electric energy meter also includes: inputting the model regression coefficient, power-related parameters, and actual power consumption of each target user into a multivariate linear regression model of statistical line loss to obtain a statistical line loss prediction value; determining a statistical line loss average value based on the actual statistical line loss value and the number of samples in the target line loss sample; determining a regression sum of squares based on the statistical line loss prediction value and the statistical line loss average value; determining a total sum of squares based on the actual statistical line loss value and the statistical line loss average value; and determining a model evaluation value of the multivariate linear regression model of statistical line loss based on the regression sum of squares and the total sum of squares.

[0108] In one example, the power-related parameters include: actual statistical line loss value, technical line loss value and total power supply. The statistical line loss prediction value refers to the statistical line loss value obtained by using the multivariate linear regression model of statistical line loss, that is, the multivariate linear regression model of statistical line loss y=x b +u1*x1+u2*x2+u3*x3+…+u n *x n +S*x s +w t *x t It should be noted that at this time, y in the multivariate linear regression model of the statistical line loss is to be solved, and other parameters are known, that is, y at this time represents the predicted value of the statistical line loss; the actual value of the statistical line loss refers to the statistical line loss value obtained by using the difference between the total power supply and the actual power consumption of all users, that is, through The statistical line loss average refers to the ratio between the actual value of the statistical line loss and the number of samples. In one example, the difference between the actual value of the statistical line loss and the average value of the statistical line loss can be determined, and then the square of the difference between the two can be used as the total sum of squares. For example, assuming that the actual value of the statistical line loss is recorded as y i , the average value of statistical line loss is recorded as Total sum of squares In one example, the difference between the statistical line loss prediction value and the statistical line loss average value may be determined, and the square of the difference between the two may be used as the regression square sum. For example, assuming that the statistical line loss prediction value is recorded as The average value of the statistical line loss is recorded as The regression sum of squares In one example, the ratio of the regression sum of squares to the total sum of squares can be used as the model evaluation value of the multivariate linear regression model of statistical line loss. For example, assuming that the model evaluation value is recorded as R 2 , the calculation formula is R 2 =SSR / SST. If the model evaluation value does not reach the preset evaluation value, it indicates that the multivariate linear regression model of statistical line loss is inaccurate, and the output is low-power users whose average power is less than the set threshold, and the process ends. Among them, the inaccuracy of the model indicates that there are unobservable variables to a large extent, such as user electricity theft, power meter failure, etc., which generally cause the user's power to be at a low level. The model is inaccurate and the solution is unreliable, so the low-power users are output as target users and no further processing is performed.

[0109] It can be understood that the model evaluation value R 2 The conformity of the multivariate linear regression model of statistical line loss is verified. If the model evaluation value R 2 Reaching the preset evaluation value indicates that the conformity of the multivariate linear regression model of statistical line loss meets the requirements, and the model regression coefficient is reliable.

[0110] The technical solution of the embodiment of the present invention takes the power supply and use units such as the substation and the line as the statistical object of line loss, and constructs a multivariate linear regression model based on the components of line loss. The attributes and weights of each component in the line loss are analyzed to determine the electric energy meter with abnormal metering status. First, based on the line loss calculation method of voltage lossless conduction theory, the technical line loss is accurately calculated, which solves the problem that the technical line loss cannot be accurately measured and the linear variable or constant is used to approximate the evaluation, resulting in model inaccuracy. Second, according to the calculated technical line loss value, the line loss attribute of the statistical line loss is judged, and the initial line loss samples of the statistical line loss are classified to ensure the consistency of the line loss sample attributes. Third, the data outlier test method is used for the classified line loss samples to eliminate the data abnormal line loss samples and reduce the impact of the possible instability of the independent variables on the evaluation results. Fourth, the independent variables are combined and solved based on the principle of similar user power load levels and the highest correlation of load curves, so as to achieve the dimensionality reduction of the independent variables in the multivariate linear regression model; solve the problem of multicollinearity of independent variables; reduce the sample demand to solve the problem of difficulty in sample acquisition; reduce the possibility of inaccuracy of the model due to the large time span of the sample data and the change of the independent variable characteristics; and reduce the difficulty of solving the multivariate linear regression model. Fifth, the data anomaly detection criteria are used to test the significant regression coefficients in the solution results to determine the electric energy meters with abnormal metering status. The solution results of the independent variable regression coefficients of the multivariate linear regression model based on the line loss components cannot quantitatively evaluate whether the metering status of the electric energy meter is normal, but compared with the electric energy meter with normal metering status, the impact of abnormal electric energy meters on line loss should be significant, and its regression coefficient should be significantly different from that of electric energy meters with normal metering status. Sixth, when the independent variable merge group test is abnormal, it is decomposed into a single independent variable and re-solved for inspection and confirmation. The output is confirmed as an abnormal electric energy meter after re-inspection, which improves the reliability of the evaluation of the metering status of the electric energy meter.

[0111] In one embodiment, Figure 3 Schematic diagram of the structure of an electric energy meter measurement status evaluation device provided by an embodiment of the present invention. Figure 3 As shown, the device includes: a line loss value determination module 310, a classification and elimination module 320, a detection and elimination module 330, a user merging module 340, a coefficient determination module 350 and a state analysis module 360.

[0112] The line loss value determination module 310 is used to determine the technical line loss value of the object to be counted within the line loss statistical period;

[0113] A classification and elimination module 320 is used to classify and eliminate the initial line loss samples of the associated statistical line loss values ​​according to the technical line loss values ​​to obtain classified line loss samples;

[0114] A detection and elimination module 330 is used to eliminate abnormal sample data in the classified line loss samples by using a data outlier detection method to obtain a target line loss sample;

[0115] A user merging module 340 is used to merge users based on the power level and power curve correlation of each user in the target line loss sample to obtain a merged target user if the sample number of the target line loss sample does not meet the preset sample margin set with the number of users included in the statistical object;

[0116] A coefficient determination module 350 is used to determine a corresponding model regression coefficient using a pre-created multivariate linear regression model of statistical line loss and power-related parameters in the target line loss sample and actual power consumption of each target user;

[0117] The state analysis module 360 ​​is used to analyze the metering state of the electric energy meter corresponding to each user by using an abnormal detection criterion and a model regression coefficient to obtain an abnormal electric energy meter.

[0118] In one embodiment, the line loss value determination module 310 includes:

[0119] A parameter acquisition unit, used to acquire the actual active power, user voltage and gateway power supply voltage of each user in the statistical object in each line loss statistical sub-period;

[0120] Theoretical active power determination unit, used to determine the theoretical active power of the associated user in the line loss statistical sub-period according to the actual active power, the user voltage and the gateway power supply voltage;

[0121] A theoretical power determination unit, used to determine the user theoretical power of the associated user in the line loss statistical period according to the theoretical active power of each user in each line loss statistical sub-period;

[0122] Theoretical power loss determination unit, used to determine the theoretical power loss of each user in the online loss statistical period according to the user's theoretical power and the measured power in the online loss statistical period;

[0123] The line loss value determination unit is used to determine the technical line loss value of the object to be counted within the line loss statistical period according to the theoretical loss power of each user and the number of users included in the object to be counted.

[0124] In one embodiment, the classification and elimination module 320 includes:

[0125] A threshold determination unit, used to determine the upper and lower thresholds for the classification of initial line loss samples based on the technical line loss value, the power consumption of all electric energy meters in the statistical object, the power consumption of the associated equipment not included in the metering, and the accuracy level of the electric energy meter;

[0126] An abnormal sample screening unit, used for screening out high-loss abnormal samples from the initial line loss samples according to an upper threshold, and screening out low-loss abnormal samples from the initial line loss samples according to a lower threshold;

[0127] A classified line loss sample screening unit is used to remove the low-loss abnormal samples from the initial line loss samples and use the high-loss abnormal samples as classified line loss samples if the number of high-loss abnormal samples contained in the initial line loss samples is greater than the number of low-loss abnormal samples;

[0128] The classified line loss sample screening unit is also used to remove the high-loss abnormal samples from the initial line loss samples and use the low-loss abnormal samples as classified line loss samples if the number of high-loss abnormal samples contained in the initial line loss samples is less than the number of low-loss abnormal samples.

[0129] In one embodiment, the data anomaly detection method includes: quartile method; the detection and elimination module 330 includes:

[0130] A statistical line loss rate determination unit, used to determine a corresponding statistical line loss rate according to the statistical line loss power and the total power supply in the classified line loss sample;

[0131] A boundary value determination unit, used to determine a corresponding abnormal upper boundary value and an abnormal lower boundary value by using a quartile method and a statistical line loss rate;

[0132] The elimination unit is used to eliminate abnormal sample data whose statistical line loss rate is greater than the abnormal upper boundary value and abnormal sample data whose statistical line loss rate is less than the abnormal lower boundary value from the classified line loss samples to obtain the target line loss samples.

[0133] In one embodiment, the user merging module 340 includes:

[0134] A correlation determination unit, used to determine the correlation between the power curve of each user in the target line loss sample and the power curve of other single users;

[0135] A correlation coefficient search unit, used to search for another user with the highest correlation with a power curve of a user and the corresponding correlation coefficient;

[0136] A merging unit, used for merging two users whose correlation coefficients reach a correlation coefficient threshold into one user group;

[0137] The merging unit is further configured to merge users according to the power levels of the unmerged single users to obtain the merged target user if the number of users consisting of the user group and the unmerged single users does not meet the preset sample margin.

[0138] In one embodiment, the power-related parameters include: actual statistical line loss value, technical line loss value and total power supply; the coefficient determination module 350 includes:

[0139] An actual power consumption acquisition unit, used to acquire the total power supply in the target line loss sample and the actual power consumption of each target user;

[0140] A statistical line loss actual value determination unit, used to determine the statistical line loss actual value according to the total power supply and the actual power consumption of each target user;

[0141] The coefficient determination unit is used to input the actual value of statistical line loss, actual power consumption and technical line loss value into the pre-built statistical line loss multivariate linear regression model solution formula to obtain the corresponding model regression coefficient.

[0142] In one embodiment, the electric energy meter measurement state device further includes:

[0143] A prediction value determination module is used to input the model regression coefficient, power-related parameters and the actual power consumption of each target user into a multivariate linear regression model of statistical line loss to obtain a statistical line loss prediction value;

[0144] A statistical line loss average value determination module is used to determine the statistical line loss average value according to the actual statistical line loss value and the number of samples in the target line loss sample;

[0145] A regression square sum determination module, used to determine the regression square sum according to the statistical line loss prediction value and the statistical line loss average value;

[0146] A total square sum determination module, used for determining the total square sum according to the actual value of the statistical line loss and the average value of the statistical line loss;

[0147] The model evaluation value determination module is used to determine the model evaluation value of the multivariate linear regression model of statistical line loss according to the regression square sum and the total square sum.

[0148] In one embodiment, the model regression coefficient includes: a user regression coefficient; the state analysis module 360 ​​includes:

[0149] A regression coefficient classification unit is used to classify the user regression coefficient according to positive and negative values ​​to obtain the first type of user regression coefficient and the second type of user regression coefficient;

[0150] The regression coefficient screening unit is used to perform cyclic detection on a group of data consisting of the regression coefficients of the first type of users and a group of data consisting of the regression coefficients of the second type of users respectively by using the Laida criterion, and screen out the regression coefficients of abnormal users;

[0151] The abnormal electric energy meter determination unit is used to take the electric energy meter of the user associated with the abnormal user regression coefficient as the abnormal electric energy meter.

[0152] In one embodiment, when the abnormal user corresponding to the abnormal electric energy meter is an abnormal user group, the electric energy meter measurement state evaluation device further includes:

[0153] The judgment and analysis module is used to treat each user in the abnormal user group as a single user, and return to the step of determining the technical line loss value of the object to be counted within the line loss statistical period, and screen out abnormal users in the abnormal user group.

[0154] The electric energy meter measurement state evaluation device provided in the embodiment of the present invention can execute the electric energy meter measurement state evaluation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0155] In one embodiment, Figure 4 is a structural block diagram of an electronic device provided by an embodiment of the present invention, such as Figure 4 As shown, a schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0156] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0157] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0158] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the electric energy meter measurement state evaluation method.

[0159] In some embodiments, the electric energy meter measurement state evaluation method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the electric energy meter measurement state evaluation method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the electric energy meter measurement state evaluation method in any other appropriate manner (e.g., by means of firmware).

[0160] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0161] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0162] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0163] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0164] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0165] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0166] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, can implement the method for evaluating the metering status of an electric energy meter provided in any embodiment of the present application.

[0167] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present application, and the programming language includes an object-oriented programming language, such as Java, Smalltalk, C++, and also includes a conventional procedural programming language, such as "C" language or similar programming language. The program code can be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0168] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0169] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for evaluating the measurement status of an electric energy meter, characterized in that: include: Determine the technical line loss value of the object to be counted within the line loss statistical period; Classifying and eliminating initial line loss samples with associated statistical line loss values ​​according to the technical line loss values ​​to obtain classified line loss samples; Using a data outlier detection method to remove abnormal sample data from the classified line loss samples to obtain target line loss samples; If the number of samples of the target line loss sample does not satisfy the preset sample margin set for the number of users included in the statistical object, users are merged based on the power level and power curve correlation of each user in the target line loss sample to obtain the merged target user; The corresponding model regression coefficient is determined by using a pre-created multivariate linear regression model of statistical line loss, power-related parameters in the target line loss sample, and actual power consumption of each target user; The abnormal detection criterion and the model regression coefficient are used to analyze the metering state of the electric energy meter corresponding to each user to obtain the abnormal electric energy meter.

2. The method according to claim 1, characterized in that Determining the technical line loss value of the object to be counted within the line loss statistical period includes: Obtain the actual active power, user voltage and gateway power supply voltage of each user in the statistical object in each line loss statistical sub-period; Determine the theoretical active power of the associated user in the line loss statistics sub-period according to the actual active power, the user voltage and the gateway power supply voltage; Determine the theoretical power consumption of the associated user in the line loss statistical period according to the theoretical active power consumption of each user in each line loss statistical sub-period; Determine the theoretical power loss of each user during the online loss statistical period according to the theoretical power consumption and the measured power consumption of each user during the online loss statistical period; The technical line loss value of the object to be counted within the line loss statistical period is determined according to the theoretical power loss of each user and the number of users included in the object to be counted.

3. The method according to claim 1, characterized in that The step of classifying and eliminating the initial line loss samples of the associated statistical line loss values ​​according to the technical line loss values ​​to obtain the classified line loss samples includes: Determine the upper and lower thresholds for the classification of the initial line loss samples based on the technical line loss value, the power consumption of all electric energy meters in the statistical objects, the power consumption of the associated equipment not included in the metering, and the accuracy level of the electric energy meters; Screening the initial line loss samples to obtain high-loss abnormal samples according to the upper threshold, and screening the initial line loss samples to obtain low-loss abnormal samples according to the lower threshold; If the number of the high-loss abnormal samples included in the initial line loss samples is greater than the number of the low-loss abnormal samples, the low-loss abnormal samples are removed from the initial line loss samples, and the high-loss abnormal samples are used as classified line loss samples; If the number of the high-loss abnormal samples included in the initial line loss samples is less than the number of the low-loss abnormal samples, the high-loss abnormal samples are removed from the initial line loss samples, and the low-loss abnormal samples are used as classified line loss samples.

4. The method according to claim 1, characterized in that The data anomaly detection method includes: a quartile method; using a data outlier detection method to remove abnormal sample data in the classified line loss samples to obtain a target line loss sample, including: Determine the corresponding statistical line loss rate according to the statistical line loss power and the total power supply in the classified line loss sample; Using the quartile method and the statistical line loss rate to determine the corresponding abnormal upper boundary value and abnormal lower boundary value; The abnormal sample data whose statistical line loss rate is greater than the abnormal upper boundary value and the abnormal sample data whose statistical line loss rate is less than the abnormal lower boundary value are removed from the classified line loss samples to obtain the target line loss samples.

5. The method according to claim 1, characterized in that The user merging based on the power level of each user in the target line loss sample and the power curve correlation to obtain the merged target user includes: Determine the power curve correlation between the power curve of each user in the target line loss sample and the power curve of other single users; Find another user with the highest correlation with a user's power curve and the corresponding correlation coefficient; Merging two users whose correlation coefficients reach a correlation coefficient threshold into one user group; If the number of users consisting of the user group and the unmerged single users does not meet the preset sample margin, the users are merged according to the power levels of the unmerged single users to obtain the merged target users.

6. The method according to claim 1, characterized in that The power-related parameters include: actual statistical line loss value, technical line loss value and total power supply; the multivariate linear regression model using the pre-created statistical line loss and the power-related parameters in the target line loss sample and the actual power consumption of each target user determines the corresponding model regression coefficient, including: Obtain the total power supply in the target line loss sample and the actual power consumption of each target user; Determine the actual value of the statistical line loss according to the total power supply and the actual power consumption of each target user; The actual value of the statistical line loss, the actual power consumption and the technical line loss value are input into a pre-constructed multivariate linear regression model solution formula of the statistical line loss to obtain the corresponding model regression coefficient.

7. The method according to claim 6, characterized in that The method further comprises: Inputting the model regression coefficient, the power-related parameters and the actual power consumption of each target user into the multivariate linear regression model of the statistical line loss to obtain a statistical line loss prediction value; Determine the average statistical line loss value according to the actual statistical line loss value and the number of samples in the target line loss sample; Determine a regression sum of squares according to the statistical line loss prediction value and the statistical line loss average value; Determine a total sum of squares according to the actual value of the statistical line loss and the average value of the statistical line loss; A model evaluation value of the multivariate linear regression model of the statistical line loss is determined according to the regression sum of squares and the total sum of squares.

8. The method according to claim 1, characterized in that The model regression coefficient includes: a user regression coefficient; the abnormal detection criterion and the model regression coefficient are used to analyze the metering state of the electric energy meter corresponding to each user to obtain the abnormal electric energy meter, including: Classifying the user regression coefficients according to positive and negative values ​​to obtain first-category user regression coefficients and second-category user regression coefficients; Using the Laida criterion, a set of data consisting of the regression coefficients of the first type of users and a set of data consisting of the regression coefficients of the second type of users are respectively subjected to cyclic detection to screen out the regression coefficients of abnormal users; The electric energy meter of the user associated with the abnormal user regression coefficient is taken as the abnormal electric energy meter.

9. The method according to any one of claims 1 to 8, characterized in that: In the case where the abnormal user corresponding to the abnormal electric energy meter is an abnormal user group, the method further includes: Each user in the abnormal user group is regarded as a single user, and the step of determining the technical line loss value of the object to be counted within the line loss statistical period is returned to screen out the abnormal users in the abnormal user group.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the electric energy meter measurement status evaluation method according to any one of claims 1 to 9.

Images