A method for monitoring the operation status of a smart electric energy meter

Abstract

The present invention relates to the field of data processing technology, and more specifically, to a method for monitoring the operating status of a smart electric energy meter. The method comprises: collecting various data sequences of the smart electric energy meter; obtaining the slope of each local data of each data sequence, and obtaining the local trend index of each data sequence based on the slope; predicting various data of the smart electric energy meter using an ARIMA time series prediction algorithm and an exponential smoothing method to obtain a first prediction value and a second prediction value of each data; obtaining the trend consistency of the prediction values ​​of all data items based on the difference between the first prediction value and the second prediction value and the local trend index of each data sequence; obtaining the adjusted prediction value of each data item according to the local trend index of each data item and the trend consistency of the prediction value of all data items, and monitoring the operating status of the smart electric energy meter. The prediction value of each data obtained by the present invention is more accurate.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and more specifically, to a method for monitoring the operating status of a smart electric energy meter. Background Art

[0002] Smart energy meters can not only realize real-time measurement of electric energy, but also optimize electricity use and improve power supply quality through data collection and analysis. Through advanced data analysis algorithms, on the basis of real-time evaluation of the operating status and health level of the energy meter, early warning of operating status failures of smart energy meters can be given, which can not only timely discover and diagnose faults, improve the reliability and safety of the power system, but also help users optimize energy consumption and reduce energy costs. Therefore, an effective method for monitoring the operating status of smart energy meters not only improves the level of intelligence in power management, but also provides solid technical support for the sustainable development of energy. By obtaining various historical data of smart energy meters and predicting various data at the next moment, it helps staff to discover and diagnose faults in a timely manner.

[0003] The patent application document with the publication number CN117312926A currently discloses a method, device, electronic equipment and medium for transformer fault prediction, which obtains the unsteady time series of dissolved gas concentration in transformer oil; converts the unsteady time series into a steady time series based on a differential algorithm; constructs an ARIMA model based on the steady time series; verifies the parameters in the ARIMA model based on the information criterion to obtain the information criterion information amount, determines the optimal parameters of the ARIMA model based on the information criterion information amount, and obtains the optimal ARIMA model based on the optimal parameters; inputs the dissolved gas concentration in the target transformer oil into the optimal ARIMA model to obtain the fault type of the target transformer.

[0004] The ARIMA time series prediction algorithm in the above method takes into account all the data in each data series when predicting the predicted value of each data at the next moment based on the data series of the smart power meter, but is not sensitive enough to the changes in local data, resulting in that the changes in local data cannot be used as a reference for predicting the operating status data of the smart power meter, thereby reducing the accuracy of the predicted value. Summary of the invention

[0005] In order to solve the problem that the ARIMA time series prediction algorithm considers all data in each data sequence during prediction, but is not sensitive enough to the change of local data, resulting in that the change of local data cannot be used as a reference for predicting the data of the next moment, thereby reducing the accuracy of the predicted value, the present invention proposes a method for monitoring the operating status of a smart electric energy meter, the method comprising the following steps:

[0006] Collect various data sequences of smart electric energy meters;

[0007] Acquire abnormal data in each data sequence, and delete the abnormal data in each data sequence from each data sequence; acquire local data of each data sequence; acquire a local trend index of each data sequence, wherein the local trend index is obtained by weighted summation of slopes of each local data of each data sequence; use the ARIMA time series prediction algorithm and the exponential smoothing method to predict each data of the smart electric energy meter, and obtain a first prediction value and a second prediction value of each data;

[0008] Get the trend consistency of the predicted values ​​of all item data ; In the formula, I represents the number of data items of the smart energy meter; Represents the first predicted value of the i-th item of data; Represents the second predicted value of the i-th item of data; Represents the local trend index of the i-th data series; exp() represents an exponential function with a natural constant as the base; min() is the minimum function; max is the maximum function; e represents a natural constant; based on the local trend index of each data series and the trend consistency of the predicted values ​​of all data items, the adjusted predicted values ​​of each data item are obtained;

[0009] The operating status of the smart energy meter is monitored based on the predicted values ​​adjusted from various data.

[0010] The innovation of the present invention lies in analyzing the local data of each data series, obtaining the local trend index of each data series, and then obtaining the trend consistency of the predicted values ​​of all data items based on the trend consistency of the local trend index of each data series and the predicted values ​​of all data items; finally, adjusting the predicted values ​​of each data item based on the local trend index of each data series and the trend consistency of the predicted values ​​of all data items to obtain the adjusted predicted values ​​of each data item, taking into account the changing trend of the local data of each data series, so that the predicted values ​​of each data of the smart electric energy meter are more accurate.

[0011] Preferably, the obtaining of abnormal data in each data sequence includes:

[0012] The box plot algorithm is used to detect anomalies in the data of each data sequence and obtain the abnormal data in each data sequence.

[0013] Remove abnormal data from each data series to avoid affecting the prediction of subsequent data.

[0014] Preferably, the obtaining of the local trend index of each data series includes:

[0015] Obtain the slope of each local data of each data series;

[0016] ;

[0017] In the formula, Represents the local trend index of the i-th data series; n represents the number of local data; Represents the slope of the jth local data of the i-th data sequence.

[0018] This makes it easier to adjust the predicted values ​​of each data set based on the local trend index of each data series, so that the changes in local data can be taken into account, making the predicted values ​​of each data set more accurate.

[0019] Preferably, the obtaining of local data of each data sequence includes:

[0020] The number of local data n is preset, and n data in each data sequence are obtained in sequence from back to front as the local data of each data sequence.

[0021] This facilitates the subsequent analysis of the changing trends of local data of each data series.

[0022] Preferably, the step of obtaining the slope of each local data of each data sequence includes:

[0023] Get the j-th local data of the ith data sequence and the first data on its left and right sides in the ith data sequence for quadratic function fitting to obtain the fitted quadratic function of the j-th local data of the ith data sequence, and get the slope of the j-th local data of the ith data sequence in its fitted quadratic function, which is recorded as the slope of the j-th local data of the ith data sequence.

[0024] Preferably, obtaining the adjusted predicted values ​​of various data includes:

[0025] ;

[0026] In the formula, represents the adjusted predicted value of the i-th data; Represents the second predicted value of the i-th item of data; Represents the first predicted value of the i-th item of data; Represents the trend consistency of the predicted values ​​of all items of data; Represents the local trend index of the ith data series.

[0027] Taking into account the changing trends of local data of each data series, the predicted values ​​of each data are adjusted to make the adjusted predicted values ​​of each data more accurate.

[0028] Preferably, the monitoring of the operating state of the smart electric energy meter according to the adjusted predicted values ​​of various data includes:

[0029] If the adjusted predicted values ​​of various data exceed the normal value range of various data, the staff will be notified that there may be abnormal conditions in the smart electricity meter.

[0030] Preferably, the collecting of various data sequences of the smart electric energy meter includes:

[0031] The sampling time is preset to 1 minute / time. The voltage sensor and the current sensor are connected to the voltage measurement port and the current measurement port of the smart power meter, and a thermometer is placed near the smart power meter to collect voltage data, current data and temperature data in real time. The data is collected for a total of two hours. The voltage data, current data and temperature data are sorted in ascending order according to the sampling time to obtain each data sequence.

[0032] The present invention has the following beneficial effects: the present invention first analyzes the local data of each data sequence to obtain the local trend index of each data sequence, and then obtains the trend consistency of the predicted values ​​of all data items based on the trend consistency of the local trend index of each data sequence and the predicted values ​​of all data items; finally, the predicted values ​​of each data item are adjusted based on the local trend index of each data sequence and the trend consistency of the predicted values ​​of all data items to obtain the adjusted predicted values ​​of each data item, taking into account the changing trend of the local data of each data sequence, so that the predicted values ​​of each data of the smart electric energy meter are more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] 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:

[0034] Figure 1 The present invention is a flowchart of a method for monitoring the operating status of a smart electric energy meter according to an embodiment of the present invention. DETAILED DESCRIPTION

[0035] 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.

[0036] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0037] See also Figure 1, which shows a flowchart of a method for monitoring the operating status of a smart electric energy meter provided by an embodiment of the present invention, the method comprising the following steps:

[0038] S001. Collect various data sequences of smart electric energy meters.

[0039] It should be noted that it is known that the voltage, current, temperature and other data of the smart energy meter can reflect the operating status of the smart energy meter. Therefore, it is necessary to collect these data to conduct real-time status monitoring of the smart energy meter so that the operation and maintenance personnel can promptly discover the operating failure of the smart energy meter and take corresponding measures in time according to the type of failure.

[0040] In an embodiment of the present invention, the sampling time is preset to 1 minute / time, the voltage sensor and the current sensor are connected to the voltage measurement port and the current measurement port of the smart power meter, and a thermometer is arranged near the smart power meter to collect voltage data, current data and temperature data in real time for two hours in total. The voltage data, current data and temperature data are sorted in ascending order according to the sampling time to obtain each data sequence.

[0041] S002. Obtain abnormal data in each data sequence, and delete the abnormal data in each data sequence from each data sequence; obtain the slope of each local data of each data sequence, and obtain the local trend index of each data sequence based on the slope.

[0042] It should be noted that it is known that it is necessary to collect data from multiple aspects of the smart energy meter to predict the subsequent data of the smart energy meter. However, the method of collecting various data will inevitably cause some abnormal data to appear in various data sequences, and these abnormal data will have a greater impact on the data prediction, reducing the accuracy of the data prediction. Therefore, it is necessary to eliminate the abnormal data in each data sequence so that subsequent data prediction is based on normal data; and because the outlier degree of abnormal data is higher than that of normal data, the existence of abnormal data will make subsequent data predictions inaccurate, so the abnormal data can be detected according to the outlier degree of each data in each data sequence, and the abnormal data can be eliminated from the data.

[0043] In an embodiment of the present invention, a box plot algorithm is used to perform anomaly detection on data in each data sequence to obtain abnormal data in each data sequence, and the abnormal data in each data sequence is deleted from each data sequence. It should be noted that the box plot algorithm for anomaly monitoring is a well-known technology and will not be described in detail in the embodiment of the present invention.

[0044] It should be noted that when the known ARIMA time series prediction algorithm predicts the data of the next sampling moment of the smart energy meter, all the data in each data series are considered, which weakens the influence of the changing trend of the local data in each data series that is closest to the current time on the prediction of each data, making the prediction of each data inconsistent with the changing trend of the local data, resulting in inaccurate data prediction. Therefore, in the process of predicting each data, it is necessary to obtain the local data in each data series that is closest to the current time as the local data of each data series, and then analyze the changing trend of the local data of each data series to obtain the local trend index of each data series, and the changing trend of the local data of each data series is obtained based on the slope of each local data of each data series.

[0045] In an embodiment of the present invention, the number of local data n is preset, and n data in each data sequence are obtained from back to front in sequence as local data of each data sequence. In an embodiment of the present invention, the number of local data n is preset to 60. In other embodiments, implementers can preset the value of the number of local data n according to specific implementation circumstances.

[0046] In the embodiment of the present invention, the jth local data of the i-th data sequence and the first data on the left and right sides thereof in the i-th data sequence are obtained to perform quadratic function fitting, and the fitting quadratic function of the jth local data of the i-th data sequence is obtained, and the slope of the jth local data of the i-th data sequence in the fitting quadratic function is obtained, which is recorded as the slope of the jth local data of the i-th data sequence;

[0047] Get the local trend index for each data series:

[0048] ;

[0049] In the formula, Represents the local trend index of the i-th data series; n represents the number of local data; Represents the slope of the jth local data of the i-th data sequence; The weight representing the slope of the jth local data of the i-th data sequence, the denominator is the sum of the index values ​​of all local data of the i-th data sequence, and the numerator is the index value of the jth local data of the i-th data sequence, so that the slope of the local data closer to the current time has a greater impact on the local trend index of each data sequence; The larger the value is, the greater the change in the local data of each data series is.

[0050] S003. Use the ARIMA time series prediction algorithm and the exponential smoothing method to predict the data of the smart energy meter, and obtain the first prediction value and the second prediction value of each data; based on the difference between the first prediction value and the second prediction value and the local trend index of each data series, obtain the trend consistency of the prediction values ​​of all data items; according to the local trend index of each data item and the trend consistency of the prediction values ​​of all data items, adjust the difference between the first prediction value and the second prediction value of each data item to obtain the adjusted prediction value of each data item.

[0051] It should be noted that the known ARIMA time series prediction algorithm obtains the actual prediction value according to the changing trend of the data, while the exponential smoothing method obtains the prediction value by weighted average of the historical data, and therefore reflects the average level of the data over a period of time; therefore, the present invention uses the ARIMA time series prediction algorithm and the exponential smoothing method to predict the various data of the smart electric energy meter, respectively, to obtain the first prediction value and the second prediction value of each data, and the difference between the first prediction value of each data and the second prediction value reflects the actual trend of the prediction value of each data.

[0052] It should be further explained that it is known that the local trend index of each data series reflects the trend change of the local data of each data series. Therefore, when the actual trend of the predicted value of each data is closer to the trend change of the local data of each data series, it means that the predicted value of each data obtained using the ARIMA time series prediction algorithm is more consistent with the change trend of the local data of each data series, and the possibility of needing to adjust it is smaller. Therefore, in an embodiment of the present invention, the trend consistency of the predicted values ​​of all data items is obtained according to the difference between the first predicted value of each data and the second predicted value and the similarity of the local trend index of each data series.

[0053] In the embodiment of the present invention, based on each data sequence, the ARIMA time series prediction algorithm is used to predict each data of the smart electric energy meter to obtain a first prediction value of each data; based on each data sequence, the exponential smoothing method is used to predict each data of the smart electric energy meter to obtain a second prediction value of each data;

[0054] Get the trend consistency of the predicted values ​​of all item data:

[0055] ;

[0056] In the formula, Represents the trend consistency of the predicted values ​​of all items of data; I represents the number of data items of the smart energy meter; Represents the first predicted value of the i-th item of data; Represents the second predicted value of the i-th item of data; Represents the local trend index of the i-th data sequence; exp() represents an exponential function with a natural constant as the base; min() is the minimum function; max is the maximum function; e represents a natural constant; The difference between the first predicted value of the i-th data item and the second predicted value of the i-th data item represents the actual trend of the first predicted value of the i-th data item; The ratio of the actual trend of the first predicted value of the i-th data item to the minimum and maximum values ​​in the local trend index of the i-th data sequence characterizes the trend consistency of the predicted value of the i-th data item. The larger the value, the more similar the trend change of the first predicted value of the i-th data item is to the trend change of the local data of the data sequence. Representatives will The value is normalized to between 0 and 1; finally, the trend consistency of the predicted values ​​of each item of data is comprehensively considered to obtain the trend consistency of the predicted values ​​of all items of data.

[0057] It should be noted that the second predicted value of each data item obtained by the known exponential smoothing method reflects the average level of the data. Therefore, in the present invention, the gap between the first predicted value and the second predicted value of each data item is adjusted based on the second predicted value of each data item, that is, the gap between the first predicted value and the average level of each data item is adjusted. It is known that the larger the local trend index of each data sequence is, the greater the change of the local data of each data sequence is, and therefore the larger the value of each data item at the next sampling moment is, and therefore the larger the gap between the first predicted value and the average level of each data item needs to be adjusted. Moreover, since the trend consistency of the predicted values ​​of all data items indicates the degree of conformity between the change trend of the first predicted value of each data item and the change trend of the local data of each data item, the larger the value is, the higher the degree of conformity is, and the smaller the degree of adjustment of the gap between the first predicted value and the average level of each data item is, therefore, according to the local trend index of each data item and the trend consistency of the predicted values ​​of all data items, the difference between the first predicted value and the second predicted value of each data item is adjusted to obtain the adjusted predicted value of each data item.

[0058] Get the adjusted forecast value of each data:

[0059] ;

[0060] In the formula, represents the adjusted predicted value of the i-th data; Represents the second predicted value of the i-th item of data; Represents the first predicted value of the i-th item of data; Represents the trend consistency of the predicted values ​​of all items of data; Represents the local trend index of the ith data series.

[0061] S004. Monitor the operating status of the smart energy meter based on the adjusted forecast values ​​of various data.

[0062] In the embodiment of the present invention, if the adjusted predicted value of each data exceeds the normal value range of each data, the staff is notified that the smart energy meter may have an abnormal situation, so that the staff can take countermeasures in advance.

[0063] 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 the operating status of a smart electric energy meter, characterized in that: include: Collect various data sequences of smart electric energy meters; Obtaining abnormal data in each data sequence, and deleting the abnormal data in each data sequence from each data sequence; obtaining local data of each data sequence; obtaining local trend index of each data sequence, including: obtaining the slope of each local data of each data sequence; ； In the formula, Represents the local trend index of the i-th data series; n represents the number of local data; represents the slope of the jth local data of the i-th data sequence, wherein the local trend index is obtained by weighted summing the slopes of the local data of each data sequence; using the ARIMA time series prediction algorithm and the exponential smoothing method to predict each data of the smart electric energy meter, and obtaining the first prediction value and the second prediction value of each data; Get the trend consistency of the predicted values ​​of all item data ; In the formula, I represents the number of data items of the smart energy meter; Represents the first predicted value of the i-th item of data; represents the second predicted value of the i-th item of data; exp() represents an exponential function with a natural constant as the base; min() is a minimum function; max() is a maximum function; e represents a natural constant; according to the local trend index of each data sequence and the trend consistency of the predicted values ​​of all items of data, the adjusted predicted values ​​of each item of data are obtained, including: ; In the formula, represents the adjusted predicted value of the i-th data; Represents the trend consistency of the predicted values ​​of all items of data; The operating status of the smart energy meter is monitored based on the predicted values ​​adjusted from various data.

2. A method for monitoring the operating status of a smart electric energy meter according to claim 1, characterized in that: The obtaining of abnormal data in each data sequence includes: The box plot algorithm is used to detect anomalies in the data of each data sequence and obtain the abnormal data in each data sequence.

3. The method for monitoring the operating status of a smart electric energy meter according to claim 1, characterized in that: The obtaining of local data of each data sequence includes: The number of local data n is preset, and n data in each data sequence are obtained in sequence from back to front as the local data of each data sequence.

4. The method for monitoring the operating status of a smart electric energy meter according to claim 1, characterized in that: The step of obtaining the slope of each local data of each data sequence includes: Get the j-th local data of the ith data sequence and the first data on its left and right sides in the ith data sequence for quadratic function fitting to obtain the fitted quadratic function of the j-th local data of the ith data sequence, and get the slope of the j-th local data of the ith data sequence in its fitted quadratic function, which is recorded as the slope of the j-th local data of the ith data sequence.

5. The method for monitoring the operating status of a smart electric energy meter according to claim 1, characterized in that: The monitoring of the operating status of the smart electric energy meter according to the adjusted predicted values ​​of various data includes: If the adjusted predicted values ​​of various data exceed the normal value range of various data, the staff will be notified that there may be abnormalities in the smart electricity meter.

6. The method for monitoring the operating status of a smart electric energy meter according to claim 1, characterized in that: The data sequences collected from the smart electric energy meter include: The sampling time is preset to 1 minute / time. The voltage sensor and the current sensor are connected to the voltage measurement port and the current measurement port of the smart power meter, and a thermometer is placed near the smart power meter to collect voltage data, current data and temperature data in real time. The data is collected for a total of two hours. The voltage data, current data and temperature data are sorted in ascending order according to the sampling time to obtain each data sequence.

Images