A method and system for processing high current test data of explosion-proof electrical appliances

Abstract

The present invention relates to the field of data processing, and in particular to a method and system for processing large current test data of explosion-proof electrical appliances. The method comprises the steps of: obtaining a time series sequence of current data of the explosion-proof electrical appliance and temperature data of the same period; filtering each current data in the time series sequence of the current data using a Wiener filtering algorithm; obtaining the filter fitting error obtained by each iterative learning during the filtering process; obtaining reference data with the current data as the center, and calculating the degree of associated abnormality according to the reference data and the corresponding temperature data; calculating an adjustment coefficient according to the degree of associated abnormality; adjusting the filter fitting error using the adjustment coefficient; and performing filtering control using the adjusted filter fitting error to complete the current data filtering process to assist in current data abnormality detection. The present invention adaptively adjusts the filter fitting error to control the degree of denoising, thereby more accurately removing noise information in the current data.

Description

Technical Field

[0001] The invention relates to the field of data processing, and in particular to a method and system for processing high current test data of explosion-proof electrical appliances. Background Art

[0002] For the safety of electricity use, it is necessary to ensure that the circuit of the electrical appliance is cut off in time when there is too much current passing through, to prevent the excessive current from burning out the circuit in the electrical appliance, or too much current from causing the electrical appliance to explode. The current detection and analysis device in the explosion-proof electrical appliance will detect the current situation and cut off the circuit in time when there is too much current.

[0003] Explosion-proof electrical appliances need to collect current data for current detection. However, due to the defects of the sensor itself and other signal interference from the surrounding environment, there is noise interference in the collected current data. Noise information will affect the results of current detection analysis. In order to improve the accuracy of current detection analysis, the collected current data needs to be denoised.

[0004] As a commonly used denoising algorithm, the Wiener filter algorithm obtains the denoised data of each data by linearly fitting the surrounding data of each data. Compared with other filtering algorithms, the corresponding filtering weight parameters are not artificially pre-set, but are obtained by fitting according to the reference signal of the signal, so the filtering weight parameters of the filtering algorithm are more in line with the actual characteristics of the signal. However, in general, it is impossible to obtain a suitable reference signal, for example, it is impossible to obtain a suitable reference signal for current data. Therefore, only the actual current data collected can be used as a reference signal. When the actual current data is used as a reference signal, there is noise information in the reference signal itself. When the surrounding data is linearly fitted based on the reference signal, if the fitting is excessive, the noise information will also be fitted, resulting in the existence of noise information in the denoised data obtained by linear fitting, and the noise information cannot be removed more comprehensively. If the fitting is insufficient, the current data characteristics cannot be mined, and accurate denoised data cannot be obtained. Therefore, how to control the degree of fitting becomes the research focus of the present invention.

[0005] The patent application document with publication number CN116415118A discloses a signal denoising method based on wavelet threshold transform and Wiener filtering. The method in the document achieves information denoising by combining wavelet threshold transform and Wiener filtering algorithm. The method in the document does not involve the problem of fitting degree control. Summary of the invention

[0006] In order to solve the problem of how to control the degree of fitting, the present invention provides a method and system for processing high current test data of explosion-proof electrical appliances.

[0007] In a first aspect, the present invention provides a method for processing high current test data of explosion-proof electrical appliances, which adopts the following technical solution:

[0008] A method for processing high current test data of explosion-proof electrical appliances, comprising the steps of:

[0009] Obtain the current data time series of explosion-proof electrical appliances and the temperature data of the same period;

[0010] Using the Wiener filtering algorithm to filter each current data in the current data time series; obtaining the filter fitting error obtained by each iterative learning during the filtering process; obtaining reference data with the current data as the center, and calculating the degree of correlation anomaly, wherein the degree of correlation anomaly represents the degree of correlation deviation between the reference data and the temperature data at the same time;

[0011] Calculate the adjustment factor , the weight of each reference data in the fitting process is set by using the degree of associated anomaly, and all reference data are fitted to obtain polynomials with various numbers of terms, Indicates The number of fitting errors of all reference data calculated by the polynomial with the number of terms is lower than the fitting error of the current data, Indicates The value of the number of items, The number of possible values ​​indicating the number of items. Indicates the degree of correlation deviation, Indicates normalization processing;

[0012] The filter fitting error is adjusted by using the adjustment coefficient. The adjusted filter fitting error is positively correlated with the filter fitting error and negatively correlated with the adjustment coefficient. The adjusted filter fitting error is used for filtering control to complete the current data filtering process to assist in current data abnormality detection.

[0013] The present invention accurately adjusts the filter fitting error by accurately analyzing the situation where the current data at the filter position is interfered by noise, thereby filtering and removing the current data at the filter position more accurately, and more comprehensively removing the noise information in the current data while reducing the loss of useful information. Further, when analyzing the situation where the current data at the filter position is interfered by noise, the feature that the noise information will cause the correlation between the temperature data and the current data to decrease is taken into account, so as to accurately measure the situation where each current data is interfered by noise by using the correlation deviation between the temperature data and the current data; further, when analyzing the situation where the current data at the filter position is interfered by noise, it is also taken into account that the noise information has local detail characteristics, and by analyzing the fitting deviation of the current data at each filter position under polynomials with different numbers of terms, it is possible to accurately reflect the situation where the change characteristics of the current data at the filter position conform to the local detail characteristics, and more accurately evaluate the situation where the current data at the filter position is interfered by noise.

[0014] Preferably, the method for obtaining the degree of association abnormality includes:

[0015] Taking a reference data as the center, obtaining data in a preset window, calculating the correlation between the data in the preset window of the reference data and the temperature data in the same period, and recording it as the correlation of the reference data;

[0016] The difference between the mean value of the correlation of all reference data and the correlation of the reference data is divided by the standard deviation of the correlation of all reference data and then normalized to obtain the degree of association anomaly of the reference data.

[0017] The present invention accurately reflects the abnormal correlation situation at each current data by analyzing the correlation between the current data and the temperature data at each current data and the correlation deviation at other current data, thereby providing a basis for subsequent accurate adjustment of the filtering fitting error.

[0018] Preferably, the adjusting the filter fitting error by using the adjustment coefficient includes:

[0019] The filter fitting error is divided by the adjustment coefficient to obtain the adjusted filter fitting error.

[0020] The present invention uses an adjustment coefficient reflecting the noise content to adjust the filter fitting error, thereby providing a basis for subsequent accurate control of the noise removal amount.

[0021] Preferably, the weight of each reference data in the fitting process is set by using the degree of associated anomaly, and polynomials with multiple numbers of terms are obtained by fitting all reference data, including:

[0022] A number of items is preset, and all reference data are fitted into a polynomial of the number of items; in the iterative fitting process, the data obtained by taking the inverse of the sum of the correlation abnormality degree of each reference data and 1 is used as the weight of the fitting error of each reference data, and the weighted sum of the fitting errors of all reference data is performed to obtain the comprehensive fitting error; in response to the convergence of the comprehensive fitting error, the fitting iteration is terminated to obtain the polynomial of the number of items.

[0023] In the process of polynomial fitting, the present invention uses the degree of associated abnormality characterizing the noise content as a weight, thereby eliminating the interference of noise information in other data on the polynomial fitting, thereby more accurately fitting the current variation law and providing a basis for subsequent accurate noise content analysis.

[0024] Preferably, the use of the adjusted filter fitting error to perform filter control to complete the current data filtering process includes:

[0025] In response to the adjusted filter fitting error being less than a preset fitting threshold, the iterative learning ends, a filter fitting value is obtained, and the filter fitting value is used as filtered data of the current data.

[0026] Preferably, the auxiliary current data abnormality detection includes:

[0027] In response to the current data being greater than a preset first current threshold, or the number of continuous current data exceeding the second current threshold being greater than a preset number threshold, it is determined that the explosion-proof electrical appliance has a current abnormality and requires power-off protection; otherwise, it is determined that the explosion-proof electrical appliance does not have a current abnormality and does not require power-off protection.

[0028] The present invention uses a threshold comparison method to perform abnormality detection, and its detection method is simple to operate and has high detection efficiency.

[0029] Preferably, the obtaining of the filter fitting error obtained by each iterative learning in the filtering process includes:

[0030] The Q data before the current data are used as filter window data, where Q represents a preset order. In each iterative learning process of the filtering process, the weight coefficient of the filter window data is obtained, and the fitting value fitted by the weight coefficient and the filter window data is used as the filter fitting value of the current data. The absolute value of the difference between the filter fitting value and the current data is used as the filter fitting error of the current data.

[0031] Preferably, the obtaining of reference data centered on the current data includes:

[0032] Taking the current data as the center, a preset number of continuous current data are acquired as reference data of the current data.

[0033] In a second aspect, the present invention provides a high current test data processing system for explosion-proof electrical appliances, which adopts the following technical solution:

[0034] A high current test data processing system for explosion-proof electrical appliances comprises: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the high current test data processing method for explosion-proof electrical appliances is implemented.

[0035] By adopting the above technical solution, the above-mentioned large current test data processing method of an explosion-proof electrical appliance is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a terminal device based on the memory and the processor for easy use.

[0036] The present invention has the following technical effects:

[0037] The present invention accurately adjusts the filter fitting error by accurately analyzing the noise interference of the current data at the filter position, thereby more accurately filtering and removing the current data at the filter position, more comprehensively removing the noise information in the current data and reducing the loss of useful information.

[0038] Furthermore, when analyzing the situation where the current data at the filtering position is interfered by noise, the characteristic that the noise information will cause the correlation between the temperature data and the current data to decrease is taken into account, so that the correlation deviation between the temperature data and the current data is used to accurately measure the situation where each current data is interfered by noise;

[0039] Furthermore, when analyzing the situation where the current data at the filtering position is interfered by noise, the local detail characteristics of the noise information are also taken into account. By analyzing the fitting deviation of the current data at each filtering position under polynomials with different numbers of terms, the variation characteristics of the current data at the filtering position are accurately reflected to be consistent with the local detail characteristics, and the situation where the current data at the filtering position is interfered by noise can be more accurately evaluated. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] 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.

[0041] Figure 1 It is a flow chart of a method in a method for processing high current test data of an explosion-proof electrical appliance according to an embodiment of the present invention;

[0042] Figure 2 It is a schematic diagram of polynomial fitting curves with different numbers of terms provided by the present invention. DETAILED DESCRIPTION

[0043] 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.

[0044] 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.

[0045] The embodiment of the present invention discloses a method for processing high current test data of explosion-proof electrical appliances, referring to Figure 1 , comprising steps S1 to S4:

[0046] S1: Obtain the current data time series of explosion-proof electrical appliances and the temperature data of the same period.

[0047] Specifically, the sensor in the explosion-proof electrical appliance collects current data once every n seconds, collects data N times, and arranges all the collected current data in time series to obtain a current data time series sequence, n represents a preset collection interval, N represents a preset number of times, and this embodiment is described by taking n as 0.5 and N as 200 as an example. Other embodiments may take other values, and this embodiment does not make specific restrictions.

[0048] The temperature data of the same period as the current data time series is collected at the same time interval using the sensors in the explosion-proof electrical appliances.

[0049] S2: Filter each current data in the current data time series sequence using the Wiener filtering algorithm; obtain the filter fitting error obtained by each iterative learning during the filtering process; obtain reference data centered on the current data, and calculate the degree of associated abnormality, wherein the degree of associated abnormality represents the degree of deviation of the correlation between the reference data and the temperature data at the same moment.

[0050] S20: Filter each current data in the current data time series using the Wiener filtering algorithm; and obtain the filtering fitting error obtained by each iterative learning during the filtering process.

[0051] It should be noted that, since the degree of noise interference at each current data is different, the fitting situation at each current data can be determined by analyzing the degree of noise interference at each current data. Among them, at the current data with less noise interference, the fitting degree needs to be high, and at the current data with high noise interference, the fitting degree can be appropriately reduced to prevent the noise data information from being fitted. The Wiener filter algorithm controls the fitting situation based on the filter fitting error, and when the filter fitting error is less than the preset threshold, the fitting iteration will be stopped. When the current data is subject to more noise interference, when the filter fitting error is less than the preset threshold, overfitting has occurred, and the noise information is also fitted. Therefore, in order to prevent this situation, the filter fitting error needs to be adjusted according to the situation of noise at each current data. First, the filter fitting error is obtained.

[0052] Optionally, as an example, each current data in the current data time series is filtered using a Wiener filtering algorithm; and a filtering fitting error obtained by each iterative learning during the filtering process is obtained, including:

[0053] The Wiener filtering algorithm is used to filter each current data in the current data time series, and the Q data before the current data are used as the filter window data, Q represents the preset order, and in each iterative learning process of the filtering process, the weight coefficient of the filter window data is obtained, and the weight coefficient is used as the weight to weight and sum the filter fitting value of the current data for all the filter window data, and the absolute value of the difference between the filter fitting value and the current data is used as the filter fitting error of the current data. This embodiment is described by taking Q as 20 as an example, and other embodiments can take other values, and this embodiment does not make specific restrictions.

[0054] S21: Acquire reference data with the current data as the center, and calculate the correlation abnormality degree, wherein the correlation abnormality degree represents the correlation deviation degree between the reference data and the temperature data at the same time.

[0055] It should be noted that when the current data is abnormal or when the current data is interfered by noise, the data will be reflected as being too large. Therefore, in order to prevent the corresponding information when the current data is abnormal from being removed during the denoising process, it is necessary to analyze the noise content on the basis of eliminating the abnormal interference of the current data.

[0056] It should be further explained that since the current is related to the temperature, the noise information is not the real current information, and thus the correlation between the noise information and the temperature is poor. Therefore, the real noise content can be analyzed based on this feature.

[0057] Preferably, as an example, obtaining reference data with the current data as the center and calculating the degree of associated abnormality include:

[0058] Taking the current data as the center, a preset number of continuous current data is obtained as reference data of the current data. This embodiment is described by taking the preset number as 21 as an example, and other embodiments may take other values, which are not specifically limited in this embodiment.

[0059] Taking a reference data as the center, obtaining data in a preset window, calculating the correlation between the data in the preset window of the reference data and the temperature data in the same period, and recording it as the correlation of the reference data;

[0060] The difference between the correlation mean of all reference data and the correlation of the reference data is divided by the correlation standard deviation of all reference data and then normalized to obtain the correlation abnormality degree of the reference data. In this embodiment, the maximum and minimum normalization method is used to normalize the above data. Other embodiments may use other normalization methods, which are not specifically limited in this embodiment. At the same time, the preset window in this embodiment is a window with a length of 5 and a height of 1. Other embodiments can be set according to actual conditions, which are not specifically limited in this embodiment.

[0061] It is understandable that when the data is not disturbed, the correlation between the current data in different regions and the temperature in the corresponding time period is similar. Therefore, the correlation between the current data and the temperature data at the corresponding moment and the difference in the value of the correlation between all the current data and the corresponding temperature data can reflect the noise interference at each current data.

[0062] S3: Calculate the adjustment coefficient.

[0063] It should be noted that since there is noise information in the temperature data, the noise information in the temperature data will also cause the correlation between the current data and the temperature data to become lower. Therefore, simply correlating the abnormality degree cannot accurately reflect the situation where the current data contains noise information.

[0064] It should be further explained that, since the temporal regularity of current data is relatively large, the temporal regularity of noise data is relatively small. At the same time, when fitting the temporal regularity of data, the structural regularity information is generally fitted first, and then the detailed regularity information is fitted, and then the noise information can be fitted. Therefore, the noise content at each position can be judged by analyzing the deviation of each current data under different degrees of fitting.

[0065] Preferably, as an example, calculating the adjustment coefficient includes:

[0066] Preset multiple numbers of terms, set the number of terms of the polynomial to one of the numbers of terms, and perform polynomial fitting processing on all reference data; in the fitting iteration process, take the data obtained by taking the inverse of the sum of the correlation abnormality degree of each reference data and 1 as the weight of the fitting error of each reference data, and perform weighted summation on the fitting errors of all reference data to obtain a comprehensive fitting error; in response to the convergence of the comprehensive fitting error, the fitting iteration ends to obtain a polynomial with the number of terms.

[0067] It should be noted that, generally, the number of terms of the polynomial is set between 3 and 15. Therefore, in this embodiment, the number of terms is taken as an integer between 3 and 15. Other embodiments may take other values, and this embodiment does not impose any specific restrictions.

[0068] It can be understood that when fitting the current data law, in order to reduce the interference degree of the noise data, the weight of the fitting error is set by using the degree of associated abnormality, so that a lower weight is set for the fitting error of the current data with a large noise content, and a higher weight is set for the fitting error of the current data with a small noise content. The weighted fitting error is thus used to constrain the fitting of the current data law, so that the fitted polynomial can better reflect the real current data law information.

[0069] The adjustment coefficient satisfies the relationship:

[0070]

[0071] in, Indicates The number of fitting errors of all reference data calculated by the polynomial with the number of terms is lower than the fitting error of the current data. The larger the value is, the more current data have smaller fitting errors than the fitting error of the current data, indicating that the current data is more affected by noise. Indicates The value of the number of items, the larger the value, the larger the number of items, the larger the number of items, the more fitting the detailed variation characteristics of all reference data, and the noise information is random, the noise information is mainly manifested in the detailed variation characteristics, so As The weight of is used to analyze the noise interference of the current data. The number of possible values ​​indicating the number of items. Indicates the degree of correlation deviation. The larger the value, the less consistent the correlation between current and temperature at the current data is with the correlation at other locations. Therefore, the current data is more affected by noise. Represents the adjustment factor. Indicates normalization processing. This embodiment uses The function normalizes the data to the range of -1 to 1. Other embodiments may use other normalization methods, which are not specifically limited in this embodiment. represents the hyperbolic tangent function.

[0072] It is understandable that if Figure 2 As shown in the schematic diagram of the polynomial fitting curve with different numbers of terms, as the number of polynomial terms increases, the degree of detail description is greater. However, the randomness of noise information is relatively large, and noise information will cause the regularity of data to deteriorate, so the regularity of noise information is relatively poor. Noise information is generally hidden in the detailed information with greater personalization, so the greater the degree of detail, the greater the possibility of noise information, so the number of polynomial terms is used as a weight to analyze the situation where the current data is interfered by noise.

[0073] The above embodiments involve fitting errors, and the method for determining the fitting errors is described below:

[0074] Use the The polynomial with the number of terms is used to fit the fitting value of each reference data, and the square of the difference between the reference data and the fitting value is used as the fitting error of the reference data. The fitting error of the current data is also obtained.

[0075] S4: adjusting the filter fitting error by using the adjustment coefficient, wherein the adjusted filter fitting error is positively correlated with the filter fitting error and negatively correlated with the adjustment coefficient; and performing filtering control by using the adjusted filter fitting error to complete the current data filtering process to assist in current data abnormality detection.

[0076] S40: adjusting the filter fitting error by using the adjustment coefficient, wherein the adjusted filter fitting error is positively correlated with the filter fitting error and negatively correlated with the adjustment coefficient.

[0077] It should be noted that, during the filtering process, when the filter fitting is more accurate, the filter fitting error reflects the noise information. Since the adjustment coefficient can characterize the degree to which the current data is interfered with by noise, the larger the adjustment coefficient, the greater the degree to which the current data is interfered with by noise. When the current data is interfered with by noise to a greater extent, the filter fitting error needs to be larger, thereby ensuring that the noise information is removed more comprehensively. On the contrary, when the noise coefficient is smaller, the current data is less interfered with by noise to a lesser extent. When the current data is interfered with by noise to a lesser extent, the filter fitting error needs to be smaller, thereby preventing excessive denoising and removing useful information.

[0078] Preferably, as an example, adjusting the filter fitting error by using the adjustment coefficient includes:

[0079] The filter fitting error is divided by the adjustment coefficient to obtain the adjusted filter fitting error.

[0080] S41: Perform filtering control using the adjusted filtering fitting error to complete the current data filtering process.

[0081] Preferably, as an example, the current data filtering process is completed by using the adjusted filter fitting error for filtering control, including:

[0082] In response to the adjusted filter fitting error being less than a preset fitting threshold, the iterative learning ends, a filter fitting value is obtained, and the filter fitting value is used as filtered data of the current data.

[0083] This embodiment is described by taking the preset fitting threshold value of 0.01 as an example. Other embodiments may take other values, and this embodiment does not make any specific limitation.

[0084] S42: To assist in current data abnormality detection.

[0085] It should be noted that when the current data exceeds the normal range by a large amount, it is easy to cause damage to the electrical appliances, and the circuit needs to be cut off in time. At the same time, the current data exceeds the normal range by a small amount, but lasts for a long time, which will cause the circuit temperature to be high and cause damage to the electrical appliances. In this case, the circuit also needs to be cut off.

[0086] Preferably, as an example, to assist in current data anomaly detection, the method includes:

[0087] In response to the current data being greater than a preset first current threshold, or the number of continuous current data exceeding the second current threshold being greater than a preset number threshold, it is determined that the explosion-proof electrical appliance has a current abnormality and requires power-off protection; otherwise, it is determined that the explosion-proof electrical appliance does not have a current abnormality and does not require power-off protection.

[0088] This embodiment is described by taking the preset first current threshold as 15, the preset second current threshold as 8, and the preset quantity threshold as 15 as an example. Other embodiments may take other values, and this embodiment does not make specific limitations.

[0089] The embodiment of the present invention also discloses a high current test data processing system for explosion-proof electrical appliances, 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 high current test data processing method for explosion-proof electrical appliances according to the present invention is implemented.

[0090] 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.

[0091] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory, a dynamic random access memory, a static random access memory, an enhanced dynamic random access memory, a high bandwidth memory, a hybrid storage cube, 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 the device or accessible or connectable to the device.

[0092] 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.

[0093] 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 method for processing high current test data of explosion-proof electrical appliances, characterized in that: Includes steps: Obtain the current data time series of explosion-proof electrical appliances and the temperature data of the same period; Using the Wiener filtering algorithm to filter each current data in the current data time series; Obtain the filter fitting error obtained by each iterative learning during the filtering process; Taking the current data as the center, obtaining reference data and calculating the degree of associated abnormality include: taking a reference data as the center, obtaining data in a preset window, calculating the correlation between the data in the preset window of the reference data and the temperature data in the same period, and recording it as the correlation of the reference data; The difference between the correlation mean of all reference data and the correlation of the reference data is divided by the correlation standard deviation of all reference data and normalized to obtain the correlation abnormality degree of the reference data; Calculate the adjustment factor , a number of items is preset, and all reference data are fitted into a polynomial of the number of items; in the fitting iteration process, the data obtained by taking the inverse of the cumulative sum of the correlation abnormality degree of each reference data and 1 is used as the weight of the fitting error of each reference data, and the weighted sum of the fitting errors of all reference data is obtained to obtain the comprehensive fitting error; in response to the convergence of the comprehensive fitting error, the fitting iteration ends, and the polynomial of the number of items is obtained, Indicates The number of fitting errors of all reference data calculated by the polynomial with the number of terms is lower than the fitting error of the current data, Indicates The value of the number of items, The number of possible values ​​indicating the number of items. Indicates the degree of correlation deviation, Indicates normalization processing; The filter fitting error is adjusted by using the adjustment coefficient. The adjusted filter fitting error is positively correlated with the filter fitting error and negatively correlated with the adjustment coefficient. The adjusted filter fitting error is used for filtering control to complete the current data filtering process to assist in current data abnormality detection.

2. A method for processing high current test data of explosion-proof electrical appliances according to claim 1, characterized in that: The adjusting the filter fitting error by using the adjustment coefficient comprises: The filter fitting error is divided by the adjustment coefficient to obtain the adjusted filter fitting error.

3. A method for processing high current test data of explosion-proof electrical appliances according to claim 1, characterized in that: The method of using the adjusted filter fitting error to perform filter control to complete the current data filtering process includes: In response to the adjusted filter fitting error being less than a preset fitting threshold, the iterative learning ends, a filter fitting value is obtained, and the filter fitting value is used as filtered data of the current data.

4. The method for processing high current test data of explosion-proof electrical appliances according to claim 1, characterized in that: The method of assisting in current data abnormality detection includes: In response to the current data being greater than a preset first current threshold, or the number of continuous current data exceeding the second current threshold being greater than a preset number threshold, it is determined that the explosion-proof electrical appliance has a current abnormality and requires power-off protection; otherwise, it is determined that the explosion-proof electrical appliance does not have a current abnormality and does not require power-off protection.

5. The method for processing high current test data of explosion-proof electrical appliances according to claim 1, characterized in that: The filtering fitting error obtained by each iterative learning in the filtering process includes: The Q data before the current data are used as filter window data, where Q represents a preset order. In each iterative learning process of the filtering process, the weight coefficient of the filter window data is obtained, and the fitting value fitted by the weight coefficient and the filter window data is used as the filter fitting value of the current data. The absolute value of the difference between the filter fitting value and the current data is used as the filter fitting error of the current data.

6. A method for processing high current test data of explosion-proof electrical appliances according to claim 1, characterized in that: The obtaining of reference data centered on the current data includes: Taking the current data as the center, a preset number of continuous current data are acquired as reference data of the current data.

7. A high current test data processing system for explosion-proof electrical appliances, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a method for processing high current test data of an explosion-proof electrical appliance according to any one of claims 1 to 6 is implemented.

Images