Data collection method and system based on power collection terminal

By analyzing the stationary and discrete characteristics of power data points in the HPLC field fault diagnosis tester, calculating the filter coefficient for non-local mean filtering, the problem of noise interference in power data acquisition is solved, and the accuracy and quality of data acquisition are improved.

CN118656766BActive Publication Date: 2024-11-19PANJIN POWER SUPPLY COMPANY OF STATE GRID LIAONING ELECTRIC POWER SUPPLY +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411142653.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-20
Publication Date
2024-11-19
Estimated Expiration
2044-08-20

AI Technical Summary

Technical Problem

HPLC field fault diagnosis tester is susceptible to noise and interference during power data acquisition, resulting in data errors. The existing non-local mean filtering method is not effectively applicable to power data processing, and there is a lack of filtering and denoising method for power data.

Method used

By collecting current timing sequences on the power acquisition terminal, analyzing the stationary and discrete characteristics of the data points, combining these characteristics to calculate the filter coefficients, perform non-local mean filtering of the power data, and data processing is performed using the filter coefficients to improve the accuracy of the collected data. .

Benefits of technology

It effectively improves the overall quality of the power acquisition terminal data, avoids data distortion, and improves the accuracy of data acquisition.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118656766B_ABST
    Figure CN118656766B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of data processing technology, and in particular to a data collection method and system based on an electric power collection terminal. The method includes the following steps: collecting the current time series of an electric energy meter, presetting a search range of data points in the current time series; taking any data point as a target data point, and determining the stationary characteristics of the target data point; obtaining the difference between each data point in the search range of the target data point and the median of the search range, and determining the discrete characteristics of the target data point; taking the modulus of a two-dimensional vector composed of the stationary characteristics and discrete characteristics of the target data point as the change characteristics of the target data point in the search range, and determining the degree of expression of the target data point on the change characteristics; calculating the filter coefficient of the target data point and using the filter coefficient to perform non-local mean filtering on the target data point to obtain the final collected data, which can effectively improve the accuracy of the data collected by the electric power collection terminal.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a data collection method and system based on a power collection terminal. Background Art

[0002] As a power collection terminal, HPLC field fault diagnosis tester is widely used in the high-speed carrier meter reading system of the low-voltage area of ​​the State Grid, with multiple functions such as data collection and fault diagnosis. However, in the actual process of collecting power data, HPLC field fault diagnosis tester may be affected by various interferences and noises, such as measurement errors, environmental interference, etc., resulting in errors in the collected power data. Therefore, in the process of power data collection, it is necessary to filter the power data to improve the accuracy of the fault diagnosis results.

[0003] In the 3rd issue of Computer Applications and Software in 2012, an article entitled "A Non-local Mean Filter Based on Optimized Parameters" was published, which provided a filtering method. The core idea of ​​this method is to extract the edge information of the noisy image and establish an optimization model based on a large number of experiments. Finally, the non-local mean algorithm based on the optimization model is applied to the denoising process of multiple images, thereby improving the denoising performance of the traditional non-local mean algorithm.

[0004] Although the filtering method proposed in the above scheme is an improvement on the existing non-local mean algorithm, it mainly adjusts the filtering coefficient in the non-local mean algorithm through the edge information and texture quantity of the image, and is not suitable for the processing of power data.

[0005] Based on this, how to filter and denoise the power data collected by the HPLC field fault diagnosis tester based on the non-local means algorithm to improve the data collection quality and thus improve the data collection accuracy of the terminal is a technical problem that needs to be urgently solved by technical personnel in this field. Summary of the invention

[0006] In order to solve the technical problem of how to filter and denoise the power data collected by the HPLC field fault diagnosis tester based on the non-local means algorithm, improve the data collection quality, and thus improve the data collection accuracy of the terminal, the present invention provides a data collection method and system based on the power collection terminal.

[0007] In a first aspect, the present invention provides a data collection method based on a power collection terminal, which adopts the following technical solution:

[0008] The data collection method based on the power collection terminal comprises the following steps:

[0009] Collect the current time series sequence of the electric energy meter, the current time series sequence includes multiple data points, and the search range of the data points is preset in the current time series sequence; take any data point as the target data point, obtain the variance of the search range of the target data point, determine the stationary feature of the target data point, and the stationary feature is negatively correlated with the variance of the search range of the target data point; obtain the difference between each data point in the search range of the target data point and the median of the search range, determine the discrete feature of the target data point, and the discrete feature is positively correlated with the absolute value of the difference; take the modulus of the two-dimensional vector composed of the stationary feature and the discrete feature of the target data point as the change feature of the target data point in the search range, determine the performance degree of the target data point on the change feature, and the performance degree is negatively correlated with the change feature; calculate the filter coefficient of the target data point: ; In the formula, represents the filter coefficient of the target data point, Indicates the performance level of the target data point. Indicates the changing characteristics of the target data point in the search range, represents the hyperparameter, represents an exponential function with e as the base; and uses the filtering coefficient to perform non-local mean filtering on the target data points to obtain the final collected data.

[0010] The embodiment of the present invention takes into account that noise may exist when the power collection terminal collects data. Therefore, by filtering the current time series through non-local mean filtering, the overall quality of the collected data can be effectively improved, thereby effectively improving the accuracy of the data collected by the power collection terminal; further, the embodiment of the present invention also takes into account that a fixed filter coefficient may cause data distortion. Therefore, by analyzing the data change trend of the data points in the search range of the target data point, the filter coefficient of the target data point is obtained, which can effectively improve the accuracy of the data collected by the power collection terminal.

[0011] According to the data collection method based on the power collection terminal provided by the present invention, the stationary characteristics of the target data points satisfy the relationship:

[0012] ;

[0013] In the formula, represents the stationary characteristics of the target data point, Indicates the number of data points included in the search range of the target data point. Represents the first-order difference value of the jth data point in the search range of the target data point, Indicates the variance of the search range for the target data point.

[0014] The present invention provides a precise method for calculating the stationary characteristics of a target data point, which can accurately obtain the stationary characteristics of a search range of the target data point, and subsequently obtain the contribution of the target data point to the stationary characteristics based on this.

[0015] According to the data collection method based on the power collection terminal provided by the present invention, the discrete characteristics of the target data points satisfy the relationship:

[0016] ;

[0017] In the formula, Represents the discrete features of the target data points, Indicates the number of data points included in the search range of the target data point. represents the jth data point in the search range of the target data point, represents the median of the search range for the target data point, Represents the absolute value symbol.

[0018] The present invention provides an accurate method for calculating the discrete features of a target data point, which can accurately obtain the discrete features of a search range of the target data point, and subsequently accurately obtain the contribution of the target data point to the discrete features based on this.

[0019] According to the data collection method based on the power collection terminal provided by the present invention, the determining the degree of expression of the target data point to the change feature includes: determining the contribution of the target data point to the stationary feature, the contribution of the stationary feature is negatively correlated with the stationary feature of the target data point; determining the contribution of the target data point to the discrete feature, the contribution of the discrete feature is negatively correlated with the discrete feature of the target data point; calculating the degree of expression of the target data point to the change feature:

[0020] ;

[0021] In the formula, Indicates the performance level of the target data point. Indicates the contribution of the target data point to the stationary feature, Indicates the contribution of the target data point to the discrete feature, Represents the changing characteristics of the target data point in the search range.

[0022] The present invention provides an accurate method for calculating the degree of expression of a target data point on a changing feature. By combining the contribution of the target data point to the stationary feature and the discrete feature, the degree of expression of the target data point on the changing feature can be accurately obtained.

[0023] According to the data collection method based on the power collection terminal provided by the present invention, the contribution of the target data point to the stationary feature satisfies the relationship:

[0024] ;

[0025] In the formula, Indicates the contribution of the target data point to the stationary feature, represents the first-order difference value of the target data point, Represents the stationary characteristics of the target data point.

[0026] According to the data collection method based on the power collection terminal provided by the present invention, the contribution of the target data point to the discrete feature satisfies the relationship:

[0027] ;

[0028] In the formula, Indicates the contribution of the target data point to the discrete feature, represents the current value of the target data point, represents the median of the search range for the target data point, Represents the discrete features of the target data points, Represents the absolute value symbol.

[0029] According to the data collection method based on the power collection terminal provided by the present invention, the non-local mean filtering of the target data point using the filter coefficient includes: presetting the length of the similarity window of the target data point, and obtaining the similarity window of the target data point with the target data point as the end in the current time series; inputting the Euclidean distance between the target data point and the data point in the similarity window into a Gaussian function, and the Gaussian function transforms the Euclidean distance according to the filter coefficient of the target data point to obtain the similarity between the target data point and the data point in the similarity window, and the similarity is negatively correlated with the Euclidean distance; determining the filter value of the target data point, and the filter value is negatively correlated with the sum of the similarities of all data points in the similarity window.

[0030] The present invention analyzes the target data point and the data change trend within the search range, and determines a suitable filter coefficient for each target data point when using non-local mean filtering, so as to avoid data distortion caused by excessive smoothing of the data point, thereby improving the accuracy of data collected by the power collection terminal.

[0031] In a second aspect, the present invention provides a data acquisition system based on a power acquisition terminal, which adopts the following technical solution:

[0032] The data collection system based on the power collection terminal includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the data collection method based on the power collection terminal is implemented.

[0033] By adopting the above technical solution, the above data collection method based on the power collection terminal is generated into a computer program and stored in the memory to be loaded and executed by the processor, so that a terminal device is made according to the memory and the processor for easy use.

[0034] The present invention has the following technical effects:

[0035] Based on the above technical scheme, the present invention performs noise filtering on the current time series through non-local mean filtering, which can effectively improve the overall quality of the collected data, thereby effectively improving the accuracy of the data collected by the power collection terminal; in addition, the present invention also obtains the filtering coefficient of the target data point by analyzing the data change trend of the data point in the search range of the target data point, thereby avoiding data distortion, thereby effectively improving the accuracy of the data collected by the power collection terminal. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, 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-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0037] Figure 1 A schematic diagram of a flow chart of a data collection method based on a power collection terminal provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0038] 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.

[0039] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0040] As a power collection terminal, the HPLC field fault diagnosis tester has multiple functions such as data collection and fault diagnosis. In the actual process of collecting power data, it may be affected by various interferences and noises, such as measurement errors, environmental interference, etc., resulting in errors in the collected power data. Therefore, in the process of power data collection, it is necessary to filter the power data to improve the accuracy of the data collected by the power collection terminal, so that accurate fault diagnosis results can be obtained later.

[0041] The non-local mean algorithm is a filtering method with good denoising effect. However, power data generally has its own variation characteristics. When the power data obtained from the power collection terminal is processed by this algorithm, if a fixed filter coefficient is used for all data, some data may be over-filtered or under-filtered, resulting in excessive smoothing of the characteristics of the data itself and causing data loss.

[0042] Based on this, the embodiment of the present invention discloses a data collection method based on a power collection terminal, which can effectively improve the accuracy of data collected by the power collection terminal. Figure 1 As shown, Figure 1 A schematic flow chart of a data collection method based on a power collection terminal provided in an embodiment of the present invention includes the following steps S1 to S5:

[0043] S1: Collect the current time series of the electric energy meter, the current time series includes multiple data points, and a search range of the data points is preset in the current time series.

[0044] It should be noted that there are many types of power data, including current, voltage, and power. The embodiment of the present invention takes the current data processing as an example for explanation, but it does not mean that the embodiment of the present invention is limited to this. Adjacent data points in the current time series are affected by the same grid structure and load distribution, and there will be spatial correlation in their data performance. When collecting current data, the embodiment of the present invention determines the filter coefficient of each data point by analyzing the data change characteristics of the data point in its search range.

[0045] For example, when obtaining the search range of a data point, an embodiment of the present invention may use the current data point as the end of the search range, construct the search range of the data point in the historical data points before the data point, and the obtained search range of the data point includes the data point itself.

[0046] The size of the search range of data points can be set according to actual needs.

[0047] It is understandable that the search range for obtaining data points is based on historical data points that have been collected before the data point. This can avoid the situation where subsequent data points need to be collected before filtering the current data point, thereby effectively reducing the impact of data filtering delay on subsequent data processing.

[0048] After obtaining the search range of each data point based on the above method, we can continue to analyze the degree to which the data point expresses the data change characteristics in the search range. If the data change characteristics in the search range of the data point are more obvious, the degree to which the data point expresses the data change characteristics is also higher. Correspondingly, the filtering parameter of the data point should also be smaller to reduce the possibility of over-smoothing. Among them, the data change characteristics in the search range of the data point can be reflected by the stable characteristics and discrete characteristics of the data point.

[0049] S2: Take any data point as the target data point, obtain the variance of the search range of the target data point, and determine the stationary characteristics of the target data point.

[0050] It should be noted that the stationary characteristics of the target data point can represent the trend and periodic changes of the data points in the search range of the target data point. The lower the stationary characteristics of the target data point, the more obvious the change trend of the search range of the target data point is.

[0051] For example, the stationary feature of the target data point is negatively correlated with the variance of the search range of the target data point; the stationary feature of the target data point can be specifically determined by referring to the following relationship:

[0052] ;

[0053] In the formula, represents the stationary characteristics of the target data point, Indicates the number of data points included in the search range of the target data point. Represents the first-order difference value of the jth data point in the search range of the target data point, Indicates the variance of the search range for the target data point.

[0054] In the above formula, the variance The larger it is, the greater the fluctuation of the search range of the target data point, the worse the data stationarity in the search range, and the worse the stationary characteristics of the target data point.

[0055] The first-order difference value of a data point can reflect the changing trend or speed of the data point. By accumulating the first-order difference values ​​of each data point in the search range, the cumulative change in the search range can be obtained. The cumulative change can reflect the cumulative effect of the overall change in the search range of the data point. The smaller the cumulative change, the less obvious the data change trend, that is, the data change in the search range is relatively stable, and the better the stability characteristics of the target data point.

[0056] S3: Obtain the difference between each data point in the search range of the target data point and the median of the search range to determine the discrete features of the target data point.

[0057] It should be noted that the median of the search range of data points can accurately reflect the central trend of the data in the search range. When the data distribution is uneven or there is a large volatility, the median can usually better represent the central trend of the data, and the difference between each data point in the search range and the median can reflect the degree to which each data point deviates from the central trend.

[0058] For example, the absolute value of the difference between each data point in the target data point search range and the median of the search range is positively correlated with the discrete feature of the target data point; the discrete feature of the target data point can be specifically determined by referring to the following relationship:

[0059] ;

[0060] In the formula, Represents the discrete features of the target data points, Indicates the number of data points included in the search range of the target data point. represents the jth data point in the search range of the target data point, represents the median of the search range for the target data point, Represents the absolute value symbol.

[0061] In the above formula, Represents the cumulative sum of the differences between each data point and the median in the search range of the target data point. The larger the value, the more dispersed the data points in the search range; the smaller the value, the more concentrated the data points in the search range.

[0062] After obtaining the stationary characteristics and discrete characteristics of the data points based on the above steps, the data change characteristics in the search range of the data points can be reflected based on the stationary characteristics and discrete characteristics of the data points, and the degree of expression of the target data points on the change characteristics can be determined:

[0063] S4: The modulus of the two-dimensional vector composed of the stationary feature and the discrete feature of the target data point is used as the change feature of the target data point in the search range to determine the degree to which the target data point expresses the change feature.

[0064] It should be noted that the larger the modulus of the two-dimensional vector composed of the stationary characteristics and discrete characteristics of the target data point, the more significant the comprehensive changes in the two dimensions of the stationary characteristics and discrete characteristics of the data point in the search range of the target data point, and the more obvious the change characteristics of the data point in the search range. However, the stationary characteristics and discrete characteristics of the target data point are obtained based on the change trend of the data point in the search range of the target data point. Therefore, when determining the degree of expression of the target data point to the change characteristics, it is necessary to analyze the contribution of the target data point to the overall stationary characteristics and discrete characteristics of the data points in the search range. The higher the contribution of the target data point, the higher the degree of expression of the target data point to the change characteristics.

[0065] For example, in an embodiment of the present invention, when determining the degree to which a target data point expresses a changing feature, the contribution of the target data point to the stationary feature can be determined first, and the contribution of the stationary feature is negatively correlated with the stationary feature of the target data point; then the contribution of the target data point to the discrete feature is determined, and the contribution of the discrete feature is negatively correlated with the discrete feature of the target data point; finally, the degree to which the target data point expresses the changing feature is calculated.

[0066] For example, the contribution of the target data point to the stationary feature can be determined by referring to the following relationship:

[0067] ;

[0068] In the formula, Indicates the contribution of the target data point to the stationary feature, represents the first-order difference value of the target data point, Represents the stationary characteristics of the target data point.

[0069] For example, the contribution of the target data point to the discrete feature can be determined by referring to the following relationship:

[0070] ;

[0071] In the formula, Indicates the contribution of the target data point to the discrete feature, represents the current value of the target data point, represents the median of the search range for the target data point, Represents the discrete features of the target data points, Represents the absolute value symbol.

[0072] After obtaining the contribution of the target data point to the stationary feature and the contribution to the discrete feature based on the above formula, the degree of expression of the target data point to the variable feature can be obtained based on the contribution of the target data point to the stationary feature and the contribution to the discrete feature.

[0073] For example, the degree to which the target data point expresses the change feature is negatively correlated with the change feature in the search range of the target data point. To determine the degree to which the target data point expresses the change feature, please refer to the following relationship:

[0074] ;

[0075] In the formula, Indicates the performance level of the target data point. Indicates the contribution of the target data point to the stationary feature, Indicates the contribution of the target data point to the discrete feature, Represents the changing characteristics of the target data point in the search range.

[0076] In the above formula, the greater the degree to which the target data point expresses the changing characteristics, the greater the possibility that the target data point contains important information, and the smaller the filtering coefficient of the target data point when it is processed by non-local mean filtering, so as to avoid information loss caused by excessive smoothing of the target data point.

[0077] S5: Calculate the filter coefficient of the target data point, and use the filter coefficient to perform non-local mean filtering on the target data point to obtain the final collected data.

[0078] For example, in the embodiment of the present invention, the filter coefficient of the target data point can be determined by referring to the following relationship:

[0079] ;

[0080] In the formula, represents the filter coefficient of the target data point, Indicates the performance level of the target data point. Indicates the changing characteristics of the target data point in the search range, represents the hyperparameter, Represents an exponential function with base e.

[0081] Among them, the hyperparameters It can be 10, which can be set according to actual needs.

[0082] After determining the filter coefficient of the target data point, the target data point can be processed based on the filter coefficient:

[0083] For example, in an embodiment of the present invention, when a non-local mean filter is performed on a target data point using a filter coefficient, the length of a similarity window of the target data point can be preset, and a similarity window of the target data point is obtained with the target data point as the end in the current time series; the Euclidean distance between the target data point and the data point in the similarity window is input into a Gaussian function, and the Gaussian function transforms the Euclidean distance according to the filter coefficient of the target data point to obtain the similarity between the target data point and the data point in the similarity window, and the similarity is negatively correlated with the Euclidean distance; and the filter value of the target data point is determined, and the filter value is negatively correlated with the sum of the similarities of all data points in the similarity window.

[0084] The length of the similarity window of the target data point is the number of data points contained in the similarity window of the target data point; the length of the similarity window can be set according to actual needs.

[0085] For example, the filter value of the target data point is determined, and the specific formula is as follows:

[0086] ;

[0087] In the formula, represents the filtered value of the target data point, represents the length of the similarity window of the target data point, Indicates the target data point and the The similarity between data points, The similarity window of the target data point The current value of each data point.

[0088] It can be seen that in the embodiment of the present invention, when collecting data based on the power collection terminal, the current time series sequence of the electric energy meter can be obtained, the current time series sequence includes multiple data points, and the search range of the data points is preset in the current time series sequence; any data point is taken as the target data point, the variance of the search range of the target data point is obtained, and the stationary feature of the target data point is determined, and the stationary feature is negatively correlated with the variance of the search range of the target data point; the difference between each data point in the search range of the target data point and the median of the search range is obtained to determine the discrete feature of the target data point, and the discrete feature is positively correlated with the absolute value of the difference; the modulus of the two-dimensional vector composed of the stationary feature and the discrete feature of the target data point is taken as the change feature of the target data point in the search range, and the expression degree of the target data point to the change feature is determined, and the expression degree is negatively correlated with the change feature; the filter coefficient of the target data point is calculated, and the filter coefficient is used to perform non-local mean filtering on the target data point to realize data collection.

[0089] In this way, the embodiment of the present invention takes into account that there may be noise when the power collection terminal collects data, which affects the use of subsequent data. Based on this, the embodiment of the present invention performs noise filtering on the current timing sequence through non-local mean filtering, which can effectively improve the overall quality of the collected data, thereby effectively improving the accuracy of the data collected by the power collection terminal; further, the embodiment of the present invention takes into account that a fixed filter coefficient may cause data distortion. Based on this, the embodiment of the present invention analyzes the data change trend of the data points in the search range of the target data points to obtain the filter coefficient of the target data point, which can effectively improve the accuracy of the data collected by the power collection terminal.

[0090] An embodiment of the present invention further discloses a data acquisition system based on a power acquisition terminal, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a data acquisition method based on a power acquisition terminal according to the present invention is implemented.

[0091] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0092] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM, a dynamic random access memory DRAM, a static random access memory SRAM, an enhanced dynamic random access memory EDRAM, a high bandwidth memory HBM, a hybrid memory cube HMC, etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0093] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0094] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A data collection method based on a power collection terminal, characterized in that: include: Collecting a current time series sequence of an electric energy meter, wherein the current time series sequence includes a plurality of data points, and a search range of the data points is preset in the current time series sequence; Taking any data point as a target data point, obtaining the variance of the search range of the target data point, and determining a stationary feature of the target data point, wherein the stationary feature is negatively correlated with the variance of the search range of the target data point; Obtaining the difference between each data point in the search range of the target data point and the median of the search range, determining the discrete feature of the target data point, wherein the discrete feature is positively correlated with the absolute value of the difference; using the modulus of the two-dimensional vector composed of the stationary feature and the discrete feature of the target data point as the change feature of the target data point in the search range, and determining the degree of expression of the target data point on the change feature, including: Determine the contribution of the target data point to the stationary feature, satisfying the relationship: ; In the formula, Indicates the contribution of the target data point to the stationary feature, represents the first-order difference value of the target data point, Represents the stationary characteristics of the target data point; Determine the contribution of the target data point to the discrete feature, satisfying the relationship: ; In the formula, Indicates the contribution of the target data point to the discrete feature, represents the current value of the target data point, represents the median of the search range for the target data point, Represents the discrete features of the target data points, Indicates the absolute value symbol; Calculate the degree to which the target data point represents the change feature: ; In the formula, Indicates the performance level of the target data point. Indicates the changing characteristics of the target data point in the search range; Calculate the filter coefficient of the target data point: ; In the formula, represents the filter coefficient of the target data point, represents the hyperparameter, represents an exponential function with e as the base; and uses the filtering coefficient to perform non-local mean filtering on the target data points to obtain the final collected data.

2. The data collection method based on the power collection terminal according to claim 1 is characterized in that: The stationary characteristics of the target data points satisfy the relationship: ; In the formula, represents the stationary characteristics of the target data point, Indicates the number of data points included in the search range of the target data point. Represents the first-order difference value of the jth data point in the search range of the target data point, Indicates the variance of the search range for the target data point.

3. The data collection method based on the power collection terminal according to claim 1 is characterized in that: The discrete features of the target data points satisfy the relationship: ; In the formula, Represents the discrete features of the target data points, Indicates the number of data points included in the search range of the target data point. represents the jth data point in the search range of the target data point, represents the median of the search range for the target data point, Represents the absolute value symbol.

4. The data collection method based on the power collection terminal according to claim 1 is characterized in that: The using the filter coefficient to perform non-local mean filtering on the target data point includes: Presetting the length of the similarity window of the target data point, and obtaining the similarity window of the target data point with the target data point as the end in the current time series; The Euclidean distance between the target data point and the data point in the similar window is input into a Gaussian function, and the Gaussian function transforms the Euclidean distance according to the filter coefficient of the target data point to obtain the similarity between the target data point and the data point in the similar window, wherein the similarity is negatively correlated with the Euclidean distance; A filter value of the target data point is determined, wherein the filter value is negatively correlated with the sum of similarities of all data points in the similarity window.

5. The data acquisition system based on the power acquisition terminal is characterized by: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the data collection method based on the power collection terminal according to any one of claims 1 to 4 is implemented.