A method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things

Abstract

The present invention relates to the field of data processing technology, and more specifically, to a method for monitoring safe electricity use of an electric energy meter based on the Internet of Things. The method comprises: obtaining a time series sequence of historical parameters of the electric energy meter, and determining an actual value of each data point in the time series sequence of the historical parameters; determining the degree of fluctuation of an initial time window; determining the size of a target time window, wherein the size of the target time window is negatively correlated with the degree of fluctuation of the initial time window; constructing an initial time window by taking the adjacent data point on the right side of the target time window as the first data point, and finally obtaining multiple target time windows in the time series sequence of the historical parameters, and obtaining a predicted value of each data point in the target time window by a weighted moving average algorithm; obtaining a monitoring result of safe electricity use of the electric energy meter based on the abnormality degree of the time series sequence of the historical parameters and a threshold comparison result, thereby effectively improving the accuracy of the monitoring result of safe electricity use of the electric energy meter.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology. More specifically, the present invention relates to a method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things. Background Art

[0002] The energy meter is an instrument used to measure the energy consumption of electricity. The use of energy meters runs through all aspects of power supply, including power generation, transmission, distribution, and finally to the user end. The accuracy and safety of energy meters are directly related to the stable operation of the power grid and the safety of users' electricity use. Therefore, in order to ensure the stable operation of the power grid and the safety of users' electricity use, it is necessary to monitor the safe operation of energy meters so that operation and maintenance personnel can handle them in time.

[0003] The patent application document with publication number CN109858785A discloses a method and system for evaluating the operating status of a smart electric energy meter. The method and system first construct a number of Bayesian networks based on historical data, then calculate the similarity between the Bayesian networks, and select a number of Bayesian network individuals with high precision and strong diversity. Finally, the selected Bayesian networks are integrated using the convex function evidence theory method to obtain the probability value of each operating status of the smart electric energy meter.

[0004] The above existing scheme mainly obtains the probability of each operating state of the smart electricity meter through historical data. However, in actual applications, electricity demand is affected by a variety of complex factors, such as weather changes, economic activities, user behavior, etc. The changes in these factors are often difficult to predict and have uncertainty. If this scheme is used to perform probability calculation for all possible operating states, the data processing volume is large and the accuracy is low.

[0005] Based on this, how to accurately obtain the monitoring results of safe electricity use by the electric energy meter is an urgent problem to be solved by technical personnel in this field. Summary of the invention

[0006] In order to solve the above technical problem of how to accurately obtain the monitoring result of safe electricity use of the electric energy meter, the present invention proposes a method for monitoring safe electricity use of the electric energy meter based on the Internet of Things, which method includes the following steps:

[0007] Obtaining each historical parameter time series sequence of the electric energy meter, and determining the actual value of each data point in the historical parameter time series sequence;

[0008] The size of the initial time window is preset, and the initial time window is constructed with the first data point in the historical parameter time series as the starting point, and the maximum actual value and the minimum actual value in the initial time window are obtained; the maximum value of the absolute value of the difference between the mean of the actual values ​​of the data points in the initial time window and the maximum actual value and the minimum actual value is used as the fluctuation index of the initial time window; the fluctuation degree of the initial time window is determined, and the fluctuation degree is positively correlated with the fluctuation index; the size of the target time window is determined, and the size of the target time window is negatively correlated with the fluctuation degree of the initial time window; the initial time window is adjusted based on the size of the target time window to obtain the target time window, and the initial time window is constructed with the adjacent data point on the right side of the target time window as the first data point, and finally multiple target time windows in the historical parameter time series are obtained, and the predicted value of each data point in the target time window is obtained by the weighted moving average algorithm; based on the abnormality degree of the historical parameter time series and the threshold comparison result, the safe electricity consumption monitoring result of the electric energy meter is obtained, wherein the abnormality degree is positively correlated with the absolute value of the difference between the predicted value and the actual value of the data point.

[0009] The present invention takes into account that when the safe electricity usage monitoring result of the electric energy meter is obtained by a prediction method, there may be fluctuations in the time window due to the continuous change of the power load. In order to accurately obtain the target time window, the present invention adjusts the initial time window by obtaining the degree of data fluctuation in the initial time window, which can effectively improve the accuracy of the predicted value of the data point in the historical parameter time series sequence, and obtains the degree of abnormality of the historical parameter time series sequence by analyzing the deviation between the predicted value and the actual value, thereby effectively improving the accuracy of the safe electricity usage monitoring result of the electric energy meter obtained based on this.

[0010] According to a method for monitoring safe electricity use of an electric energy meter based on the Internet of Things provided by the present invention, obtaining the time series sequence of each historical parameter of the electric energy meter includes: collecting historical data of each parameter generated during the operation of the electric energy meter through sensors in the Internet of Things, preprocessing the historical data, and obtaining the time series sequence of historical parameters corresponding to each of the parameters, wherein the time series sequence of historical parameters includes multiple data points.

[0011] The present invention takes into account that the overall quality of the raw data collected by sensors in the Internet of Things is low, and therefore processes the raw data through a preprocessing method to prepare for subsequent data analysis.

[0012] According to a method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things provided by the present invention, the fluctuation degree of the initial time window satisfies the relationship:

[0013] ;

[0014] In the formula, Indicates the degree of fluctuation in the initial time window, represents the size of the initial time window, represents the actual value of the i-th data point in the initial time window, represents the actual mean value of the data points in the initial time window, Represents the average value of the difference between the actual value and the mean of the actual value of each data point in the initial time window, Indicates the volatility indicator for the initial time window.

[0015] The present invention provides an accurate calculation formula for the fluctuation degree of the initial time window. By combining the fluctuation index of the initial time window and the standard deviation of the data, the accuracy of the correction value of the initial time window obtained based on this can be effectively improved, and preparations can be made for adjusting the size of the initial time window.

[0016] According to a method for monitoring safe electricity use of electric energy meters based on the Internet of Things provided by the present invention, the determining of the size of the target time window further includes: calculating the number of trend changes in the initial time window:

[0017] ;

[0018] In the formula, represents the number of trend changes in the initial time window, represents the size of the initial time window, It represents the actual value difference between the i+1th data point and the ith data point in the initial time window. Represents the actual value difference between the i+2th data point and the i+1th data point in the initial time window; determines the degree of trend change in the initial time window, and the degree of trend change is negatively correlated with the number of trend changes in the initial time window; determines the size of the target time window, and the size of the target time window is negatively correlated with the degree of trend change.

[0019] The present invention takes into account that the number of trend changes in the initial time window has an impact on the size of the initial time window. If the number of trend changes in the initial time window is large, it means that the overall trend change in the initial time window is large. In order to accurately obtain the predicted value of the data point, the initial time window needs to be reduced to capture the changes in each data.

[0020] According to a method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things provided by the present invention, the trend change degree of the initial time window satisfies the relationship:

[0021] ;

[0022] In the formula, Indicates the degree of trend change in the initial time window, represents the size of the initial time window, represents the actual value of the i-th data point in the initial time window, represents the actual mean value of the data points in the initial time window, represents the collection time of the i-th data point in the initial time window, represents the mean value of the data points collected in the initial time window, represents the number of trend changes in the initial time window, Represents an exponential function with base e.

[0023] According to a method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things provided by the present invention, the step of determining the size of the target time window includes:

[0024] Calculate the revised value of the initial time window:

[0025] ;

[0026] In the formula, represents the correction value of the initial time window, Indicates the degree of fluctuation in the initial time window, Indicates the degree of trend change in the initial time window, represents an inverse tangent function; the product of the correction value of the initial time window and the size of the initial time window is rounded up to obtain the size of the target time window.

[0027] According to a method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things provided by the present invention, the abnormality degree of the historical parameter time series satisfies the relationship:

[0028] ;

[0029] In the formula, Indicates the degree of abnormality of the historical parameter time series. Indicates the number of data points in the historical parameter time series. represents the actual value of the jth data point in the historical parameter time series, represents the predicted value of the jth data point in the historical parameter time series, represents the linear normalization function.

[0030] According to a method for monitoring safe electricity usage of an electric energy meter based on the Internet of Things provided by the present invention, the electric energy meter safe electricity usage monitoring result is obtained based on the abnormality degree of the historical parameter time series sequence and the threshold comparison result, including: presetting a threshold; obtaining the abnormality degree of each of the historical parameter time series sequences, and taking the maximum value of the abnormality degree of each of the historical parameter time series sequences as the safety assessment index of the electric energy meter; when the safety assessment index is greater than the threshold, the electric energy meter safe electricity usage monitoring result is abnormal; when the safety assessment index is less than or equal to the threshold, the electric energy meter safe electricity usage monitoring result is safe.

[0031] According to a method for monitoring safe electricity usage of an electric energy meter based on the Internet of Things provided by the present invention, the method of obtaining the monitoring result of safe electricity usage of the electric energy meter further includes: issuing an early warning when the monitoring result of safe electricity usage of the electric energy meter is abnormal.

[0032] The present invention takes into account that when an electric energy meter is abnormal, there may be potential safety hazards, and therefore issues an early warning for the abnormal monitoring results so that operation and maintenance personnel can handle them in a timely manner.

[0033] The present invention has the following beneficial effects:

[0034] Based on the above technical scheme, the present invention adjusts the initial time window by obtaining the fluctuation degree in the initial time window, thereby reducing the influence of data fluctuation in the initial time window on the predicted value, and can effectively improve the accuracy of the predicted value of the data point in the historical parameter time series sequence, thereby effectively improving the accuracy of the electric energy meter safety electricity consumption monitoring result obtained based on this; in addition, the present invention also calculates the trend change degree in the initial time window, analyzes the influence of the data change trend in the initial time window on the predicted value, combines the fluctuation degree and the trend change degree in the initial time window, and obtains the abnormality degree of the historical parameter time series sequence, thereby effectively improving the accuracy of the electric energy meter safety electricity consumption monitoring result obtained based on this. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0036] Figure 1 A schematic flow chart of a method for monitoring safe electricity consumption of an electric energy meter based on the Internet of Things is provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0038] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0039] It should be noted that in the actual operation of the power system, electricity is affected by many factors, such as weather changes, economic activities, user behavior, etc. The changes in these factors are often difficult to predict and are uncertain. Traditional prediction algorithms are difficult to capture these changes when determining the safe electricity use monitoring results of electricity meters, resulting in low accuracy of the monitoring results.

[0040] Based on this, the embodiment of the present invention provides a method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things. By analyzing the data fluctuations in the initial time window of each historical parameter time series of the electric energy meter, the initial time window is adjusted to obtain the target time window, which can effectively improve the accuracy of the predicted value of the obtained data point, so as to accurately obtain the safe electricity consumption monitoring result of the electric energy meter. Figure 1 As shown, Figure 1 A schematic flow chart of a method for monitoring safe electricity consumption of an electric energy meter based on the Internet of Things provided by an embodiment of the present invention, the method comprising the following steps:

[0041] S1: Obtain the time series of each historical parameter of the electric energy meter, and determine the actual value of each data point in the time series of the historical parameter.

[0042] It should be noted that there are many parameters generated during the operation of the electric energy meter. In order to obtain the operation status of the electric energy meter more comprehensively, the historical parameter time series of multiple parameters can be obtained for comprehensive analysis. When the predicted value of the data point is determined by the weighted moving average algorithm, the average value in the initial time window is obtained by assigning weights to the data points in the window based on the distance. Therefore, the size of the initial time window needs to be adjusted according to the changes in the data to improve the accuracy of data prediction.

[0043] For example, in an embodiment of the present invention, when obtaining the time series sequences of various historical parameters of the electric energy meter, the historical data of various parameters generated during the operation of the electric energy meter can be collected by sensors in the Internet of Things, and the historical data can be preprocessed to obtain the historical parameter time series sequences corresponding to each parameter, and the historical parameter time series sequences contain multiple data points.

[0044] The preprocessing method can be set according to actual needs.

[0045] For example, the parameter may be voltage, current, etc., which may be set according to actual needs.

[0046] It is understandable that each data point in the obtained historical parameter time series corresponds to an actual value obtained by collection and a sampling time corresponding to the actual value. For example, if the parameter is voltage, the actual value is the voltage value, and the sampling time of the data point is the sampling time of the voltage value.

[0047] For example, a voltage sensor may be used to collect voltage, and a current sensor may be used to collect current, etc. The collection method may be specifically set according to actual needs.

[0048] Specifically, when collecting historical data of parameters, the sensor can be set inside the electric energy meter, the collection frequency can be set according to actual needs, and the collected historical data of each parameter can be transmitted to the power monitoring database and classified and marked.

[0049] After preparing the data in the above way, you can continue with the following steps.

[0050] S2: Preset the size of the initial time window and determine the degree of fluctuation of the initial time window.

[0051] It should be noted that, considering the operation of the power grid, various factors such as load changes, equipment failures or user behavior may cause complex fluctuations in the data of the electric energy meter. And such fluctuations are irregular. In order to accurately capture these fluctuations and evaluate their impact on the monitoring results of the electric energy meter, the historical parameter time series can be divided, and the changes in the data in the local time window can be used to evaluate the fluctuation degree of the recent data.

[0052] By way of example, in an embodiment of the present invention, when determining the degree of fluctuation of an initial time window, the size of the initial time window can be preset, the initial time window can be constructed with the first data point in the historical parameter time series as the starting point, and the maximum actual value and the minimum actual value in the initial time window can be obtained; the maximum value of the absolute value of the difference between the mean of the actual values ​​of the data points in the initial time window and the maximum actual value and the minimum actual value is used as the fluctuation index of the initial time window; and the degree of fluctuation of the initial time window is determined, and the degree of fluctuation is positively correlated with the fluctuation index.

[0053] It is understandable that the size of the time window is the number of data points contained in the time window. For example, if the size of the time window is 30, it contains 30 data points.

[0054] Specifically, in an embodiment of the present invention, when determining the volatility index of the initial time window, the maximum value of the actual values ​​corresponding to all data points in the initial time window can be first obtained as the maximum actual value in the initial time window; the minimum value of the actual values ​​corresponding to all data points can be taken as the minimum actual value in the initial time window; the absolute value of the difference between the mean actual value of the data points in the initial time window and the maximum actual value, and the absolute value of the difference between the mean actual value of the data points in the initial time window and the minimum actual value are obtained respectively, and the absolute values ​​of the two differences are compared, and the absolute value of the larger difference is obtained as the maximum value, which is the volatility index of the initial time window.

[0055] It can be seen that the step of obtaining the volatility index of the initial time window is to determine the distance from the point farthest from the actual value mean to the actual value mean. The larger the value, the greater the extreme deviation of the actual value of the data point in the initial time window, and the greater the possibility that there are extreme values ​​far from the actual value mean in the data points, resulting in an increase in the overall dispersion of the data.

[0056] The volatility index of the initial time window can reflect the overall dispersion of the initial time window, but it is impossible to accurately obtain the specific volatility in the initial time window. Based on this, the embodiment of the present invention introduces the standard deviation, which can be used to evaluate the volatility of data, but the calculation of the standard deviation is easily affected by abnormal extreme values, resulting in low accuracy of the obtained volatility. Therefore, the standard deviation of the difference between each data point and the mean of the actual value of all data points can be obtained, and then combined with the volatility index, so as to accurately obtain the volatility in the initial time window.

[0057] Specifically, in the embodiment of the present invention, when determining the degree of fluctuation of the initial time window, the actual value mean of the data points in the initial time window can be obtained first. , determine the difference between the actual value and the mean of the actual value of each data point in the initial time window, and then get the average value of all the differences , and combined with the volatility indicator of the initial time window , and the fluctuation degree of the initial time window is obtained. For details, please refer to the following relationship:

[0058] ;

[0059] In the formula, Indicates the degree of fluctuation in the initial time window, represents the size of the initial time window, represents the actual value of the i-th data point in the initial time window, represents the actual mean value of the data points in the initial time window, Represents the average value of the difference between the actual value and the mean of the actual value of each data point in the initial time window, Indicates the volatility indicator for the initial time window.

[0060] In the above formula, It represents the standard deviation of the difference between each data point and the mean of the actual values ​​of all data points. The larger the value, the more dispersed the distribution of the data points around the mean in the initial time window; conversely, the smaller the value, the more concentrated the distribution of the data points around the mean in the initial time window.

[0061] After the fluctuation degree of the initial time window is obtained based on the above steps, the initial time window can be adjusted based on the fluctuation degree of the initial time window.

[0062] S3: Determine the size of the target time window.

[0063] It should be noted that due to factors such as the user's electricity demand, the scheduling of the power grid, and changes in the external environment, the load in the electricity consumption process is constantly changing, and the change in load will directly affect the performance of the electrical parameters monitored by the electric energy meter. The increase or decrease in load will cause changes in parameters such as equipment current, voltage, and power, but such changes are affected by multiple factors and have no regularity, resulting in multiple trend changes in parameter data. Based on this, when determining the size of the target time window, the embodiment of the present invention can also distinguish trend changes within the window to improve the accuracy of smooth prediction.

[0064] By way of example, in an embodiment of the present invention, when determining the size of the target time window, the number of trend changes in the initial time window can also be obtained to determine the degree of trend change in the initial time window, where the degree of trend change is negatively correlated with the number of trend changes in the initial time window; the size of the target time window is determined.

[0065] Specifically, when calculating the number of trend changes in the initial time window, you can first obtain the actual values ​​corresponding to the i-th data point, the i+1-th data point, and the i+2-th data point in the initial time window, and then calculate the actual value difference between the i+1-th data point and the i-th data point. , the actual value difference between the i+2th data point and the i+1th data point ; and based on the actual value difference Difference from actual value The ratio of , get the number of trend changes in the initial time window, for details, refer to the following relationship:

[0066] ;

[0067] In the formula, represents the number of trend changes in the initial time window, represents the size of the initial time window, It represents the actual value difference between the i+1th data point and the ith data point in the initial time window. Represents the actual value difference between the i+2th data point and the i+1th data point in the initial time window.

[0068] In the above formula, The value can reflect the number of data trend changes. If the ratio of the differences between three adjacent data points is less than 0, it means that the trend directions from the i-th data point to the i+1-th data point and the i+1-th data point to the i+2-th data point are opposite, which is recorded as a trend change; if the ratio is greater than or equal to 0, it means that the trend directions from the i-th data point to the i+1-th data point and the i+1-th data point to the i+2-th data point are the same, and the trend has not changed; by traversing all data points in the initial time window, the number of trend changes in the initial time window can be obtained.

[0069] Among them, since the i-th data point at the edge does not have the i+1-th data point and the i+2-th data point, it is not processed in the technical solution provided in the embodiment of the present invention.

[0070] Based on the above formula, the number of trend changes in the initial time window can be obtained. In order to avoid errors in the monitoring results caused by randomness or noise in the data in the initial time window, the covariance between the actual value of the data point in the initial time window and the corresponding collection time can also be obtained to reflect the consistency of the trend change. By combining the number of trend changes and the consistency of trend changes, the accuracy of the trend change degree in the initial time window can be improved.

[0071] By way of example, in an embodiment of the present invention, when determining the degree of trend change in an initial time window, the collection time corresponding to each data point can also be obtained; the trend indicator of the initial time window is obtained by the covariance of the actual value of the data point in the initial time window and the collection time; the degree of trend change in the initial time window is determined, and the degree of trend change is negatively correlated with the trend indicator.

[0072] Specifically, when determining the degree of trend change in the initial time window, the actual value mean of the data points in the initial time window can be obtained first. and the mean value at the time of collection , then based on the actual value mean , mean value at the time of collection And the number of trend changes, get the trend change degree of the initial time window, for details, please refer to the following relationship:

[0073] ;

[0074] In the formula, Indicates the degree of trend change in the initial time window, represents the size of the initial time window, represents the actual value of the i-th data point in the initial time window, represents the actual mean value of the data points in the initial time window, represents the collection time of the i-th data point in the initial time window, represents the mean value of the data points collected in the initial time window, represents the number of trend changes in the initial time window, Represents an exponential function with base e.

[0075] In the above formula, It represents the covariance between the actual value of the data point in the initial time window and the acquisition time. The covariance can represent the degree of joint change between the two variables. If the two variables in the initial time window increase or decrease at the same time, the covariance is positive, indicating that the actual value and the acquisition time have the same overall change trend; if one variable increases and the other decreases, the covariance is negative, indicating that the actual value and the acquisition time have inconsistent overall change trends.

[0076] It represents the product of the standard deviation of the two variables, the actual value in the initial time window and the value at the time of collection. In order to avoid excessive deviation of the results due to data dimension problems, the covariance between different variables can be compared on this scale, thereby accurately evaluating the consistency of the overall change trend between the variables.

[0077] Based on the above formula, it can be seen that if the trend of the data in the initial time window changes more frequently, and the overall change trend between the actual value of the data point in the initial time window and the sampling time is inconsistent, the trend of the data in the initial time window will change more.

[0078] After obtaining the trend change degree and fluctuation degree of the data in the initial time window based on the above steps, the size of the target time window can be obtained based on the trend change degree and fluctuation degree; wherein, the size of the target time window is negatively correlated with the fluctuation degree of the initial time window; the size of the target time window is negatively correlated with the trend change degree.

[0079] It should be noted that the size of the initial time window can be adjusted according to the changes in the data. If the time window is too large, some important short-term fluctuation information may be smoothed out, causing the prediction results to deviate from the actual changes; if the time window is too small, it may be too sensitive to random fluctuations, affecting the stability and accuracy of the prediction. Therefore, in order to improve the accuracy of the prediction, the size of the window can be adaptively adjusted based on the degree of fluctuation and trend change of the data in the initial window.

[0080] For example, in an embodiment of the present invention, when determining the size of the target time window, the correction value of the initial time window can be calculated first, and the product of the correction value of the initial time window and the size of the initial time window can be rounded up to obtain the size of the target time window.

[0081] Specifically, the correction value of the initial time window can be determined by referring to the following relationship:

[0082] ;

[0083] In the formula, represents the correction value of the initial time window, Indicates the degree of fluctuation in the initial time window, Indicates the degree of trend change in the initial time window, Represents the inverse tangent function.

[0084] In the above formula, It represents the product of the data fluctuation degree and trend change degree in the initial time window. The larger the value, the more drastic the data fluctuation and the more frequent the trend change in the initial time window. At this time, it is necessary to reduce the time window in order to more sensitively capture the data dynamics and respond to parameter changes in time. On the contrary, the smaller the product value, the more stable the data trend and the smaller the fluctuation. At this time, a larger time window is needed to smooth out short-term fluctuations and enhance the stability of the prediction.

[0085] Based on the above formula, it can be seen that the closer the final correction value is to 0.5, the smaller the initial time window is; the closer the correction value is to 1.5, the larger the initial time window is.

[0086] For example, in an embodiment of the present invention, the target time window may be obtained by adjusting the initial time window based on the size of the target time window. In order to obtain the size of the target time window more accurately, the size of the time window may be adjusted iteratively.

[0087] Specifically, the product of the correction value of the initial time window and the size of the initial time window is rounded up to obtain the size of the target time window, and then the size of the initial time window is adjusted to obtain the target time window; and based on the above steps again, the correction value of the adjusted target time window is obtained, and the above steps are repeated until the difference between the correction values ​​in the target time window before and after the adjustment is less than the error threshold, and the adjusted target time window is used as the final target time window.

[0088] The error threshold can be set to 0.03, which can be set according to actual needs.

[0089] After obtaining the first target time window in the historical parameter time series sequence based on the above steps, continue to perform the following steps.

[0090] S4: The adjacent data point on the right side of the target time window is used as the first data point to construct an initial time window, and finally multiple target time windows in the historical parameter time series are obtained. The predicted value of each data point in the target time window is obtained by the weighted moving average algorithm.

[0091] It should be noted that after obtaining the first target time window in the historical parameter time series sequence based on the above steps, other target time windows in the historical parameter time series sequence can be obtained based on the above steps. Obtaining the predicted value of each data point in the time window by the weighted moving average algorithm belongs to the existing solution, and the embodiment of the present invention will not be described in detail here.

[0092] Specifically, the adjacent data point on the right side of the target time window is used as the new first data point to construct a new initial time window, and the correction value of the new initial time window is obtained based on the above steps and then adjusted. Finally, the historical parameter time series sequence can be divided into multiple target time windows.

[0093] In this process, there may be insufficient number of the latest data points, which makes it impossible to build the initial time window. It is understandable that the historical parameter time series will increase with the increase of collected parameter data, and the changes of very few data points cannot be directly used as the basis for abnormal monitoring results. Therefore, for such data points, it is still necessary to collect enough data points before processing.

[0094] S5: Based on the abnormality degree and threshold comparison results of the historical parameter time series, the safe electricity consumption monitoring results of the electric energy meter are obtained.

[0095] It should be noted that after obtaining the target time window in the historical parameter time series based on the above steps, the predicted value of each data point in the target time window can be obtained by the weighted moving average algorithm. By analyzing the deviation between the predicted value and the actual value of each data point, the degree of abnormality of the historical parameter time series can be obtained; wherein, the degree of abnormality is positively correlated with the absolute value of the difference between the predicted value and the actual value of the data point.

[0096] For example, in the embodiment of the present invention, the abnormality degree of the historical parameter time series sequence is determined, and the specific details can be referred to the following relationship:

[0097] ;

[0098] In the formula, Indicates the degree of abnormality of the historical parameter time series. Indicates the number of data points in the historical parameter time series. represents the actual value of the jth data point in the historical parameter time series, represents the predicted value of the jth data point in the historical parameter time series, represents the linear normalization function.

[0099] In the above formula, It represents the sum of the absolute values ​​of the differences between the actual values ​​and the predicted values ​​of all data points in the historical parameter time series. The larger the value, the more actual values ​​of data points in the historical parameter time series deviate from the expected values, and the higher the possibility of abnormal data in the historical parameter time series.

[0100] It can be understood that, based on the above steps, the abnormality degree of each historical parameter time series sequence can be obtained, and the safe electricity consumption monitoring result of the electric energy meter can be obtained through the abnormality degree of the historical parameter time series sequence.

[0101] By way of example, in an embodiment of the present invention, when obtaining the safe electricity usage monitoring result of the electric energy meter based on the comparison result between the abnormality degree of the historical parameter time series sequence and the threshold, a threshold can be preset first; the abnormality degree of each historical parameter time series sequence is obtained, and the maximum value of the abnormality degree of each historical parameter time series sequence is used as the safety assessment index of the electric energy meter; when the safety assessment index is greater than the threshold, the safe electricity usage monitoring result of the electric energy meter is abnormal; when the safety assessment index is less than or equal to the threshold, the safe electricity usage monitoring result of the electric energy meter is safe.

[0102] The threshold value may be set according to actual needs, and the embodiment of the present invention does not impose too many limitations on this.

[0103] It can be seen that in the embodiment of the present invention, when determining the monitoring result of safe electricity consumption of the electric energy meter, the time series of each historical parameter of the electric energy meter is obtained, and the actual value of each data point in the time series of the historical parameter is determined; the size of the initial time window is preset, and the initial time window is constructed with the first data point in the time series of the historical parameter as the starting point, and the maximum actual value and the minimum actual value in the initial time window are obtained; the maximum value of the absolute value of the difference between the mean of the actual value of the data point in the initial time window and the maximum actual value and the minimum actual value is used as the fluctuation index of the initial time window; the degree of fluctuation of the initial time window is determined, and the degree of fluctuation is positively correlated with the fluctuation index; the size of the target time window is determined, and the size of the target time window is negatively correlated with the degree of fluctuation of the initial time window; the adjacent data point on the right side of the target time window is used as the first data point to construct the initial time window, and finally multiple target time windows in the time series of the historical parameter are obtained, and the predicted value of each data point in the target time window is obtained by the weighted moving average algorithm; based on the abnormality degree of the time series of the historical parameter and the threshold comparison result, the monitoring result of safe electricity consumption of the electric energy meter is obtained, and the abnormality degree is positively correlated with the absolute value of the difference between the predicted value and the actual value of the data point.

[0104] In this way, the present invention takes into account that when the safe electricity usage monitoring results of the electric energy meter are obtained through the prediction method, there may be fluctuations in the time window due to the continuous changes in the power load. In order to accurately obtain the target time window, the embodiment of the present invention adjusts the initial time window by obtaining the degree of data fluctuation in the initial time window, which can effectively improve the accuracy of the predicted values ​​of the data points in the historical parameter time series sequence, and thus effectively improve the accuracy of the safe electricity usage monitoring results of the electric energy meter obtained based on this.

[0105] After obtaining the safe electricity usage monitoring result of the electric energy meter based on the above, the abnormal safe electricity usage monitoring result can also be processed.

[0106] For example, in an embodiment of the present invention, after obtaining the safe electricity usage monitoring result of the electric energy meter, an early warning may be issued when the safe electricity usage monitoring result of the electric energy meter is abnormal.

[0107] In this way, by issuing an early warning, the operation and maintenance personnel can be reminded to deal with the electricity meter in a timely manner, thereby effectively reducing safety hazards.

[0108] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things, characterized in that: include: Obtaining each historical parameter time series sequence of the electric energy meter, and determining the actual value of each data point in the historical parameter time series sequence; Preset the size of the initial time window, construct the initial time window with the first data point in the historical parameter time series as the starting point, obtain the maximum actual value and the minimum actual value in the initial time window; take the maximum value of the absolute value of the difference between the mean of the actual values ​​of the data points in the initial time window and the maximum actual value and the minimum actual value as the volatility index of the initial time window; determine the volatility degree of the initial time window, and the volatility degree is positively correlated with the volatility index; Determine the degree of trend change for the initial time window: ； In the formula, Indicates the degree of trend change in the initial time window, represents the size of the initial time window, represents the actual value of the i-th data point in the initial time window, represents the actual mean value of the data points in the initial time window, represents the collection time of the i-th data point in the initial time window, represents the mean value of the data points collected in the initial time window, represents the number of trend changes in the initial time window, represents an exponential function with base e; Determining a size of a target time window, wherein the size of the target time window is negatively correlated with a degree of fluctuation of the initial time window; The target time window is obtained by adjusting the initial time window based on the size of the target time window, and the adjacent data point on the right side of the target time window is used as the first data point to construct the initial time window, and finally multiple target time windows in the historical parameter time series are obtained, and the predicted value of each data point in the target time window is obtained by the weighted moving average algorithm; Based on the abnormality degree and threshold comparison result of the historical parameter time series, the safe electricity consumption monitoring result of the electric energy meter is obtained, wherein the abnormality degree is positively correlated with the absolute value of the difference between the predicted value and the actual value of the data point.

2. According to the method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things in claim 1, it is characterized in that: The step of obtaining the time series of historical parameters of the electric energy meter includes: The historical data of various parameters generated during the operation of the electric energy meter are collected by sensors in the Internet of Things, and the historical data are preprocessed to obtain the historical parameter time series sequences corresponding to each of the parameters, wherein the historical parameter time series sequences include multiple data points.

3. The method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things according to claim 1 is characterized in that: The fluctuation degree of the initial time window satisfies the relationship: ； In the formula, Indicates the degree of fluctuation in the initial time window, represents the size of the initial time window, represents the actual value of the i-th data point in the initial time window, represents the actual mean value of the data points in the initial time window, Represents the average value of the difference between the actual value and the mean of the actual value of each data point in the initial time window, Indicates the volatility indicator for the initial time window.

4. The method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things according to claim 1 is characterized in that: Determining the size of the target time window also includes: Calculate the number of trend changes in the initial time window: ； In the formula, represents the number of trend changes in the initial time window, represents the size of the initial time window, It represents the actual value difference between the i+1th data point and the ith data point in the initial time window. Represents the actual value difference between the i+2th data point and the i+1th data point in the initial time window; determines the degree of trend change in the initial time window, and the degree of trend change is negatively correlated with the number of trend changes in the initial time window; determines the size of the target time window, and the size of the target time window is negatively correlated with the degree of trend change.

5. The method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things according to claim 4 is characterized in that: Determining the size of the target time window includes: Calculate the revised value of the initial time window: ； In the formula, represents the correction value of the initial time window, Indicates the degree of fluctuation in the initial time window, Indicates the degree of trend change in the initial time window, represents the inverse tangent function; The product of the correction value of the initial time window and the size of the initial time window is rounded up to obtain the size of the target time window.

6. The method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things according to claim 1 is characterized in that: The abnormal degree of the historical parameter time series satisfies the relationship: ； In the formula, Indicates the degree of abnormality of the historical parameter time series. Indicates the number of data points in the historical parameter time series. represents the actual value of the jth data point in the historical parameter time series, represents the predicted value of the jth data point in the historical parameter time series, represents the linear normalization function.

7. The method for monitoring safe electricity consumption of electric energy meters based on the Internet of Things according to claim 1 is characterized in that: The abnormal degree and threshold comparison result based on the historical parameter time series are used to obtain the safe electricity consumption monitoring result of the electric energy meter, including: Preset threshold value; Obtaining the abnormality degree of each of the historical parameter time series, and taking the maximum value of the abnormality degree of each of the historical parameter time series as the safety assessment index of the electric energy meter; When the safety assessment index is greater than the threshold, the safe electricity usage monitoring result of the electric energy meter is obtained to be abnormal; When the safety assessment index is less than or equal to the threshold, the safe electricity use monitoring result of the electric energy meter is obtained as safe.

8. The method for monitoring safe electricity use of electric energy meters based on the Internet of Things according to claim 1 or 7, characterized in that: The step of obtaining the safe electricity usage monitoring result of the electric energy meter further includes: When the safe electricity usage monitoring result of the electric energy meter is abnormal, an early warning is issued externally.

Images