Piezoresistor failure life threshold determination method and device

By using PCA and isolated forest algorithms for dimensionality reduction and outlier detection, the problem of determining the failure threshold in varistor lifetime prediction is solved, enabling safer use of varistors.

CN118966004AInactive Publication Date: 2024-11-15HUBEI UNIV OF TECH
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411121042.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-11-15
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing methods for predicting the lifespan of varistors are insufficient to accurately determine the failure lifespan threshold, leading to varistor failure before reaching its lifespan limit, which poses a safety hazard.

Method used

The PCA algorithm is used to reduce the dimensionality of the varistor test dataset to two dimensions. Anomaly test data is identified using the Isolation Forest algorithm. Based on this data, the failure lifetime threshold of the varistor is determined.

Benefits of technology

This improves the safety of varistor use by reducing data processing complexity and accurately determining the failure life threshold, thus preventing varistor failure before its service life reaches its limit.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118966004A_ABST
    Figure CN118966004A_ABST
Patent Text Reader

Abstract

The invention relates to a piezoresistor failure life threshold determination method and device, and belongs to the technical field of equipment test.The piezoresistor failure life threshold determination method comprises the steps that the dimension of a piezoresistor test data set is reduced to two dimensions based on a PCA algorithm, and a first data matrix is obtained; determining all abnormal test data in the first data matrix based on an isolated forest algorithm; and determining a failure life threshold value of the piezoresistor based on all the abnormal test data in the first data matrix. According to the method, the threshold value of the failure life of the piezoresistor is determined, so that the safety of the piezoresistor in use is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of equipment testing, and in particular to a method and device for determining a failure life threshold value of a varistor. Background Art

[0002] Surge protection devices (SPDs) can suppress lightning surges and transient overvoltages on the line, effectively absorb, disperse or suppress lightning electromagnetic pulses, reduce damage to internal equipment, and thus ensure the safety and stable operation of the equipment. In practical applications, voltage-limiting SPDs (usually using varistors) are one of the most common and most commonly used types. Voltage-limiting SPDs have the characteristics of fast response and rapid overvoltage guidance, can effectively protect electrical equipment from damage caused by surge voltages, and are easy to install and maintain.

[0003] Accurately predicting the remaining service life of an SPD is of great significance to the operation and maintenance of the SPD, and can greatly reduce manpower and material resources. The internal protection device of a voltage-limiting SPD is generally a zinc oxide (ZnO) varistor. The SPD gradually deteriorates after being subjected to surges for a long time. The fundamental reason is that the internal varistor device has aged. As a key protection component in an SPD, the study of the degradation of the ZnO varistor is the key to SPD status assessment and life prediction. Although the existing varistor life prediction method can predict the life of the varistor, the varistor often fails before its service life reaches its limit. Continuing to use the varistor after failure poses a great safety hazard. Summary of the invention

[0004] In view of this, it is necessary to provide a method and device for determining the failure life threshold of a varistor, so as to solve the problem that the existing varistor life prediction method is difficult to determine the failure life threshold of the varistor.

[0005] In order to solve the above problems, the present invention provides a method for determining a failure life threshold of a varistor, comprising: Based on the PCA algorithm, the dimension of the varistor test data set is reduced to two dimensions to obtain the first data matrix; Determine all abnormal test data in the first data matrix based on an isolation forest algorithm; Based on all abnormal test data in the first data matrix, a failure life threshold of the varistor is determined.

[0006] In a possible implementation, the PCA algorithm is used to reduce the dimension of the varistor test data set to two dimensions to obtain a first data matrix, including: Standardizing the second data matrix to obtain a third data matrix, where the second data matrix is ​​a data matrix corresponding to the varistor test data set, the number of rows of the second data matrix corresponds to the number of data in the varistor test data set, and the number of columns of the second data matrix corresponds to the number of test data variables in the varistor test data set; constructing a correlation coefficient matrix based on the third data matrix, and determining a contribution rate of each principal component based on an eigenvalue of the correlation coefficient matrix; Based on the column vectors corresponding to the two principal components with the highest contribution rates, a first data matrix is ​​determined.

[0007] In a possible implementation, the correlation coefficient matrix is ​​determined based on the following formula:

[0008] in, Indicates the third data matrix Line The elements of the column, Indicates the first Line The elements of the column, Indicates the first The average value of the column elements, Indicates the first The variance of the column elements, is the number of rows of the second data matrix, represents the correlation coefficient matrix, represents the transpose of the third data matrix, represents the third data matrix.

[0009] In a possible implementation, the contribution rate of each principal component is determined based on the following formula:

[0010] in, Indicates The contribution rate of the principal components, The correlation coefficient matrix is eigenvalues, is the order of the correlation coefficient matrix.

[0011] In a possible implementation, determining all abnormal test data in the first data matrix based on the isolation forest algorithm includes: Randomly selecting a preset number of rows of data from the first data matrix each time to construct multiple subtrees; A feature is randomly selected in each subtree, and a split threshold is randomly selected within the value range of the selected feature for binary splitting until each subtree reaches a preset height or each leaf node of each subtree has only one data; Determine the abnormal value corresponding to each row of data in the first data matrix in the isolation forest, and determine all abnormal test data in the first data matrix based on the abnormal value corresponding to each row of data in the isolation forest.

[0012] In a possible implementation, determining the failure life threshold of the varistor based on all abnormal test data in the first data matrix includes: Determine the positions of all abnormal test data of the varistor test data set based on the positions of all abnormal test data in the first data matrix; The failure life threshold of the varistor is determined based on the average value of the varistor life corresponding to all abnormal test data in the varistor test data set.

[0013] In a possible implementation, the test data variables in the varistor test data set include: Varistor voltage, nonlinear coefficient, leakage current, shunt resistance and shunt capacitance.

[0014] The present invention also provides a device for determining a failure life threshold of a varistor, comprising: A dimension reduction module, used for reducing the dimension of the varistor test data set to two dimensions based on the PCA algorithm to obtain a first data matrix; A first determination module, used to determine all abnormal test data in the first data matrix based on an isolation forest algorithm; The second determination module is used to determine the failure life threshold of the varistor based on all abnormal test data in the first data matrix.

[0015] The present invention also provides an electronic device, including a memory and a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for determining the failure life threshold of a varistor as described above is implemented.

[0016] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for determining a failure life threshold value of a varistor as described above is implemented.

[0017] The beneficial effects of the present invention are as follows: the method and device for determining the failure life threshold of a varistor provided by the present invention first reduce the dimension of a varistor test data set to two dimensions through a PCA algorithm to obtain a first data matrix, thereby reducing data processing complexity, and then determine all abnormal test data in the first data matrix according to an isolation forest algorithm. Since the abnormal test data corresponds to the data when the varistor fails, finally, the failure life threshold of the varistor can be determined based on all the abnormal test data in the first data matrix, thereby realizing the determination of the failure life threshold of the varistor, and then improving the safety of the varistor when in use. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic flow chart of an embodiment of a method for determining a varistor failure life threshold provided by the present invention; Figure 2 A schematic diagram of a flow chart of an embodiment of outlier detection based on an island forest algorithm provided by the present invention; Figure 3 A schematic structural diagram of an embodiment of a device for determining a failure life threshold value of a varistor provided by the present invention; Figure 4 A schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0020] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0021] In the description of the present invention, reference to an "embodiment" means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive of other embodiments. It is explicitly and implicitly understood by those skilled in the art that the described embodiments may be combined with other embodiments.

[0022] SPD can suppress lightning surges and transient overvoltages on the line, effectively absorb, disperse or suppress lightning electromagnetic pulses, reduce damage to internal equipment, and thus ensure the safety and stable operation of the equipment. In practical applications, voltage-limiting SPD (usually using varistors) is one of the most common and most commonly used types. Voltage-limiting SPD has the characteristics of fast response and rapid overvoltage guidance, can effectively protect electrical equipment from damage by surge voltage, and is easy to install and maintain.

[0023] Accurately predicting the remaining service life of an SPD is of great significance to the operation and maintenance of the SPD, and can greatly reduce manpower and material resources. The internal protection device of a voltage-limiting SPD is generally a zinc oxide (ZnO) varistor. The SPD gradually deteriorates after being subjected to surges for a long time. The fundamental reason is that the internal varistor device has aged. As a key protection component in an SPD, the study of the degradation of the ZnO varistor is the key to SPD status assessment and life prediction. Although the existing varistor life prediction method can predict the life of the varistor, the varistor often fails before its service life reaches its limit. Continuing to use the varistor after failure poses a great safety hazard.

[0024] In order to solve the above problems, the present invention provides a method for determining a failure life threshold of a varistor.

[0025] The specific embodiments are described in detail below: A specific embodiment of the present invention discloses a method for determining a failure life threshold of a varistor, combining Figure 1 Come and see, Figure 1 The flowchart of an embodiment of a method for determining a failure life threshold of a varistor provided by the present invention includes steps S101 to S103, wherein: In step S101, the dimension of the varistor test data set is reduced to two dimensions based on the PCA algorithm to obtain a first data matrix; In step S102, all abnormal test data in the first data matrix are determined based on the isolation forest algorithm; In step S103, based on all abnormal test data in the first data matrix, a failure life threshold of the varistor is determined.

[0026] During implementation, the dimension of the varistor test data set can first be reduced to two dimensions according to the principal component analysis (PCA) algorithm to reduce the complexity of data processing and obtain a first data matrix.

[0027] Then, all abnormal test data in the first data matrix may be determined according to the isolation forest algorithm, and then the failure life threshold of the varistor may be determined through all the abnormal test data in the first data matrix.

[0028] The method for determining the failure life threshold of a varistor provided by the present invention can be applied to detection scenarios of surge protectors, and can also be applied to detection scenarios of other components based on varistor operation, and the present invention does not make specific limitations on this.

[0029] Compared with the prior art, the method for determining the failure life threshold of the varistor provided in this embodiment first reduces the dimension of the varistor test data set to two dimensions through the PCA algorithm to obtain a first data matrix, thereby reducing the complexity of data processing, and then determines all abnormal test data in the first data matrix according to the isolation forest algorithm. Since the abnormal test data corresponds to the data when the varistor fails, finally, the failure life threshold of the varistor can be determined based on all the abnormal test data in the first data matrix, thereby realizing the determination of the failure life threshold of the varistor, and then improving the safety of the varistor when using it.

[0030] Exemplarily, the PCA algorithm is used to reduce the dimension of the varistor test data set to two dimensions to obtain a first data matrix, including: Standardizing the second data matrix to obtain a third data matrix, where the second data matrix is ​​a data matrix corresponding to the varistor test data set, the number of rows of the second data matrix corresponds to the number of data in the varistor test data set, and the number of columns of the second data matrix corresponds to the number of test data variables in the varistor test data set; constructing a correlation coefficient matrix based on the third data matrix, and determining a contribution rate of each principal component based on an eigenvalue of the correlation coefficient matrix; Based on the column vectors corresponding to the two principal components with the highest contribution rates, a first data matrix is ​​determined.

[0031] Specifically, when the dimension of the varistor test data set is reduced to two dimensions according to the PCA algorithm, the data matrix corresponding to the varistor test data set (ie, the second data matrix) may first be standardized to obtain a third data matrix.

[0032] After the third data matrix is ​​obtained, a correlation coefficient matrix can be constructed according to the third data matrix, and then the contribution rate of each principal component in the correlation coefficient matrix can be determined according to the eigenvalues ​​of the correlation coefficient matrix.

[0033] After determining the contribution rate of each principal component in the correlation coefficient matrix, the first data matrix can be determined according to the column vectors corresponding to the two principal components with the highest contribution rates. For example, the column vectors corresponding to the two principal components with the highest contribution rates can be added to obtain the first data matrix.

[0034] Exemplarily, the correlation coefficient matrix is ​​determined based on the following formula:

[0035] in, Indicates the third data matrix Line The elements of the column, Indicates the first Line The elements of the column, Indicates the first The average value of the column elements, Indicates the first The variance of the column elements, is the number of rows of the second data matrix, represents the correlation coefficient matrix, represents the transpose of the third data matrix, represents the third data matrix.

[0036] Specifically, when constructing the correlation coefficient matrix according to the third data matrix, the correlation coefficient matrix can be constructed according to the above formula.

[0037] Exemplarily, the contribution rate of each principal component is determined based on the following formula:

[0038] in, Indicates The contribution rate of the principal components, The correlation coefficient matrix is eigenvalues, is the order of the correlation coefficient matrix.

[0039] Specifically, the contribution rate of each principal component in the correlation coefficient matrix can be determined according to the above formula.

[0040] Exemplarily, the determining all abnormal test data in the first data matrix based on the isolation forest algorithm includes: Randomly selecting a preset number of rows of data from the first data matrix each time to construct multiple subtrees; A feature is randomly selected in each subtree, and a split threshold is randomly selected within the value range of the selected feature for binary splitting until each subtree reaches a preset height or each leaf node of each subtree has only one data; Determine the abnormal value corresponding to each row of data in the first data matrix in the isolation forest, and determine all abnormal test data in the first data matrix based on the abnormal value corresponding to each row of data in the isolation forest.

[0041] Specifically, when all abnormal test data in the first data matrix are determined according to the isolation forest algorithm, a preset number of rows of data may be randomly selected (with replacement) multiple times in the first data matrix to construct multiple subtrees.

[0042] After completing the construction of multiple subtrees, a feature can be randomly selected from each subtree, and a split threshold can be randomly selected within the value range of the selected feature for binary splitting until each subtree reaches a preset height or each leaf node of each subtree has only one data, thereby completing the construction of the isolation forest.

[0043] Finally, the outlier value corresponding to each row of data in the first data matrix in the isolation forest can be determined, and then according to the outlier value corresponding to each row of data in the first data matrix in the isolation forest, the row data with outlier values ​​close to 1 is determined as the abnormal test data in the first data matrix.

[0044] Exemplarily, determining the failure life threshold of the varistor based on all abnormal test data in the first data matrix includes: Determine the positions of all abnormal test data of the varistor test data set based on the positions of all abnormal test data in the first data matrix; The failure life threshold of the varistor is determined based on the average value of the varistor life corresponding to all abnormal test data in the varistor test data set.

[0045] Specifically, when determining the failure life threshold of the varistor based on all abnormal test data in the first data matrix, the positions of all abnormal test data in the varistor test data set can be first determined based on the positions of all abnormal test data in the first data matrix.

[0046] Then, the failure life threshold of the varistor can be determined according to the average value of the varistor life corresponding to all abnormal test data in the varistor test data set. The method for determining the varistor life value can be an existing determination method or other determination method, which is not limited by the present invention.

[0047] Exemplarily, the test data variables in the varistor test data set include: Varistor voltage, nonlinear coefficient, leakage current, shunt resistance and shunt capacitance.

[0048] Specifically, the test data variables in the varistor test data set may include varistor voltage, nonlinear coefficient, leakage current, parallel resistance and parallel capacitance. The above five test data variables are test data variables that are determined based on experimental results and have a strong correlation with the life of the varistor.

[0049] The technical solution of the present invention is better described below with reference to a specific embodiment: Combination Figure 2 Come and see, Figure 2 This is a flow chart of an embodiment of outlier detection based on the island forest algorithm provided by the present invention. The specific steps of the process are as follows: 1. First, the ZnO varistor dataset is reduced to 2D data and then isolated tree training is performed on it.

[0050] 2. Randomly select n sample points in the training set to form a subset, and then build a tree on the subset.

[0051] 3. In the subsample, randomly select a feature and randomly select a threshold within the value range of the feature for binary splitting. The threshold is between the maximum and minimum values ​​of the feature.

[0052] 4. Use recursion to repeat step 1) to build a tree until the tree reaches a specified height or there is only one data point in each leaf node.

[0053] 5. After building the model, calculate the anomaly score. The calculation process is as follows: set up is the sample point of the ZnO varistor dataset, using data points The path length in the entire isolated forest is , the formula for standardization is:

[0054] in Indicates that the given number of samples is When , the average path length of the binary tree constructed by the data is used to standardize the sample path length ; Harmonic series , is the Euler constant, approximately equal to 0.5772; is the number of training samples for a single binary tree.

[0055] The calculation formula for the anomaly score is:

[0056] in Representative data The mathematical expectation of the path length in the entire isolation forest is, The anomaly score for each sampled data instance. The value range is 0-1. If a point is an abnormal sample, then The closer s is to 1; if s is close to 0, the point is considered a normal sample. When it is around 0.5, it is considered that the data does not have obvious abnormal characteristics.

[0057] The embodiment of the present invention also provides a device for determining a failure life threshold value of a varistor, Figure 3 Come and see, Figure 3 This is a schematic diagram of the structure of an embodiment of a device for determining a failure life threshold of a varistor provided by the present invention. The device 300 for determining a failure life threshold of a varistor includes: A dimension reduction module 301 is used to reduce the dimension of the varistor test data set to two dimensions based on the PCA algorithm to obtain a first data matrix; A first determination module 302, configured to determine all abnormal test data in a first data matrix based on an isolation forest algorithm; The second determination module 303 is used to determine the failure life threshold of the varistor based on all abnormal test data in the first data matrix.

[0058] The specific implementation methods of each module of the device for determining the failure life threshold of a varistor can refer to the description of the above-mentioned method for determining the failure life threshold of a varistor, and have similar beneficial effects, which will not be repeated here.

[0059] It should be noted that the device for determining the failure life threshold of a varistor may be provided on a power system monitoring device or on a varistor testing device, and the present invention does not make any specific limitation on this.

[0060] The embodiment of the present invention further provides an electronic device, Figure 4 Come and see, Figure 4 This is a structural diagram of an embodiment of an electronic device provided by the present invention. The electronic device 400 includes a processor 401, a memory 402, and a computer program stored in the memory 402 and executable on the processor 401. When the processor 401 executes the program, the method for determining the failure life threshold of the varistor as described above is implemented.

[0061] As a preferred embodiment, the electronic device 400 further includes a display 403 for displaying that the processor 401 executes the method for determining the failure life threshold of the varistor as described above.

[0062] Exemplarily, the computer program may be divided into one or more modules / units, one or more modules / units are stored in the memory 402, and are executed by the processor 401 to complete the present invention. One or more modules / units may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device 400. For example, the computer program may be divided into the dimensionality reduction module 301, the first determination module 302, and the second determination module 303 in the above embodiment, and the specific functions of each module are as described above, and are not described one by one here.

[0063] The electronic device 400 may be a desktop computer, a notebook, a PDA or a smart phone with an adjustable camera module.

[0064] Among them, the processor 401 may be an integrated circuit chip with signal processing capabilities. The above-mentioned processor 401 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0065] The memory 402 may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electric erasable programmable read-only memory (EEPROM), etc. The memory 402 is used to store programs, and the processor 401 executes the program after receiving the execution instruction. The process definition method disclosed in any of the embodiments of the present invention may be applied to the processor 401 or implemented by the processor 401.

[0066] The display 403 may be an LCD display screen or an LED display screen, for example, a display screen on a mobile phone.

[0067] Understandably, Figure 4 The structure shown is only a schematic diagram of a structure of the electronic device 400. The electronic device 400 may also include Figure 4 More or fewer components as shown. Figure 4 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0068] The electronic device provided according to the above-mentioned embodiment of the present invention can be implemented with reference to the specific description of the method for determining the failure life threshold of a varistor according to the present invention, and has similar beneficial effects as the method for determining the failure life threshold of a varistor as described above, which will not be repeated here.

[0069] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method for determining a failure life threshold of a varistor as described above is implemented.

[0070] Generally speaking, the computer instructions for implementing the method of the present invention can be carried by any combination of one or more computer-readable storage media. Non-transitory computer-readable storage media can include any computer-readable media except for the signal itself that is temporarily propagating.

[0071] Computer-readable storage media may be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium that contains or stores a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0072] Computer program codes for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages, in particular, Python suitable for neural network computing and platform frameworks based on TensorFlow, PyTorch, etc. may be used. The program code may be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0073] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0074] The present invention discloses a method and device for determining a failure life threshold of a varistor. First, the dimension of a varistor test data set is reduced to two dimensions by a PCA algorithm to obtain a first data matrix, thereby reducing data processing complexity. Then, all abnormal test data in the first data matrix are determined according to an isolation forest algorithm. Since the abnormal test data corresponds to data when the varistor fails, the failure life threshold of the varistor can be determined according to all the abnormal test data in the first data matrix, thereby realizing the determination of the failure life threshold of the varistor, and further improving the safety of the varistor when in use.

[0075] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for determining a failure life threshold of a varistor, characterized in that: include: Based on the PCA algorithm, the dimension of the varistor test data set is reduced to two dimensions to obtain the first data matrix; Determine all abnormal test data in the first data matrix based on an isolation forest algorithm; Determine a failure life threshold of the varistor based on all abnormal test data in the first data matrix; The test data variables in the varistor test data set include: Varistor voltage, nonlinear coefficient, leakage current, shunt resistance and shunt capacitance; The PCA algorithm is used to reduce the dimension of the varistor test data set to two dimensions, and a first data matrix is ​​obtained, including: Standardizing the second data matrix to obtain a third data matrix, where the second data matrix is ​​a data matrix corresponding to the varistor test data set, the number of rows of the second data matrix corresponds to the number of data in the varistor test data set, and the number of columns of the second data matrix corresponds to the number of test data variables in the varistor test data set; constructing a correlation coefficient matrix based on the third data matrix, and determining a contribution rate of each principal component based on an eigenvalue of the correlation coefficient matrix; Determine a first data matrix based on the column vectors corresponding to the two principal components with the highest contribution rates; The correlation coefficient matrix is ​​determined based on the following formula: in, Indicates the third data matrix Line The elements of the column, Indicates the first Line The elements of the column, Indicates the first The average value of the column elements, Indicates the first The variance of the column elements, is the number of rows of the second data matrix, represents the correlation coefficient matrix, represents the transpose of the third data matrix, represents the third data matrix.

2. The method for determining the failure life threshold of a varistor according to claim 1, characterized in that: The contribution rate of each principal component is determined based on the following formula: in, Indicates The contribution rate of the principal components, The correlation coefficient matrix is eigenvalues, is the order of the correlation coefficient matrix.

3. The method for determining the failure life threshold of a varistor according to claim 1, characterized in that: The method of determining all abnormal test data in the first data matrix based on the isolation forest algorithm includes: Randomly selecting a preset number of rows of data from the first data matrix each time to construct multiple subtrees; A feature is randomly selected in each subtree, and a split threshold is randomly selected within the value range of the selected feature for binary splitting until each subtree reaches a preset height or each leaf node of each subtree has only one data; Determine the abnormal value corresponding to each row of data in the first data matrix in the isolation forest, and determine all abnormal test data in the first data matrix based on the abnormal value corresponding to each row of data in the isolation forest.

4. The method for determining the failure life threshold of a varistor according to any one of claims 1 to 3, characterized in that: Based on all abnormal test data in the first data matrix, determining the failure life threshold of the varistor includes: Determine the positions of all abnormal test data of the varistor test data set based on the positions of all abnormal test data in the first data matrix; The failure life threshold of the varistor is determined based on the average value of the varistor life corresponding to all abnormal test data in the varistor test data set.

5. A device for determining a failure life threshold of a varistor, characterized in that: include: A dimension reduction module, used for reducing the dimension of the varistor test data set to two dimensions based on the PCA algorithm to obtain a first data matrix; A first determination module, used to determine all abnormal test data in the first data matrix based on an isolation forest algorithm; A second determination module is used to determine a failure life threshold of the varistor based on all abnormal test data in the first data matrix; The test data variables in the varistor test data set include: Varistor voltage, nonlinear coefficient, leakage current, shunt resistance and shunt capacitance; The PCA algorithm is used to reduce the dimension of the varistor test data set to two dimensions, and a first data matrix is ​​obtained, including: Standardizing the second data matrix to obtain a third data matrix, where the second data matrix is ​​a data matrix corresponding to the varistor test data set, the number of rows of the second data matrix corresponds to the number of data in the varistor test data set, and the number of columns of the second data matrix corresponds to the number of test data variables in the varistor test data set; constructing a correlation coefficient matrix based on the third data matrix, and determining a contribution rate of each principal component based on an eigenvalue of the correlation coefficient matrix; Determine a first data matrix based on the column vectors corresponding to the two principal components with the highest contribution rates; The correlation coefficient matrix is ​​determined based on the following formula: in, Indicates the third data matrix Line The elements of the column, Indicates the first Line The elements of the column, Indicates the first The average value of the column elements, Indicates the first The variance of the column elements, is the number of rows of the second data matrix, represents the correlation coefficient matrix, represents the transpose of the third data matrix, represents the third data matrix.

6. An electronic device, characterized in that: The invention comprises a memory and a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for determining the failure life threshold of a varistor according to any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the method for determining the failure life threshold of a varistor according to any one of claims 1 to 4 is implemented.

Citation Information

Patent Citations

  • A method for detecting abnormal behavior of electricity consumption of consumers based on isolated forests

    CN109308306A

  • Method for carrying out anomaly detection on wind tunnel test data

    CN112362292A

  • PSO-SVR-based bearing residual service life prediction method

    CN112699502A

  • Abnormity detection method and system for subway air conditioner compressor based on isolated forest

    CN116181635A

  • PCA-VMD-based water-turbine generator set feature extraction and fault early warning method

    CN117131368A