Fault diagnosis method, device and equipment and computer readable storage medium

Abstract

The invention relates to a fault diagnosis method, device and equipment and a computer readable storage medium. The method comprises the following steps: acquiring an arc cosine value and an arc sine value corresponding to each piece of power parameter information in a power parameter information set of power electronic equipment; determining a triangular difference characteristic value of each electric power parameter information according to the arc cosine value and the arc sine value corresponding to each electric power parameter information; according to each triangular difference characteristic value, constructing a power parameter characteristic graph of the power electronic equipment; and processing the power parameter characteristic pattern by using a parameter fault diagnosis model to obtain a fault diagnosis result of the power electronic equipment. By adopting the method, the accuracy of the fault diagnosis result of the power electronic equipment can be improved.

Description

Technical Field

[0001] The present application relates to the field of power electronics technology, and in particular to a fault diagnosis method, device, equipment and computer-readable storage medium. Background Art

[0002] Power electronic equipment is widely used in production and life, and its stability is of great significance to the safety of people's lives and property as well as industrial production. In actual operation, the working environment of power electronic equipment is very complex, which causes various devices in power electronic equipment to often experience huge operating pressure, making power electronic devices very prone to various failures. Timely monitoring of the status of power electronic devices and pre-maintenance can reduce the possibility of equipment failure and downtime.

[0003] Taking open circuit faults and short circuit faults in power electronic equipment as an example, open circuit faults and short circuit faults will cause obvious changes in the voltage parameters of power electronic devices in the equipment. Especially when power electronic devices have aging faults, the parameters of the devices will drift, causing parametric failures in the equipment.

[0004] However, the related art lacks a solution that can accurately diagnose parametric faults of power electronic equipment. Summary of the invention

[0005] Based on this, it is necessary to provide a fault diagnosis method, apparatus, fault diagnosis equipment, computer-readable storage medium and computer program product that can improve the accuracy of fault diagnosis results of power electronic equipment in response to the above technical problems.

[0006] In a first aspect, the present application provides a fault diagnosis method, comprising:

[0007] Obtaining an arccosine value and an arcsine value corresponding to each power parameter information in a power parameter information set of a power electronic device;

[0008] Determine the trigonometric differential eigenvalue of each power parameter information according to the arccosine value and arcsine value corresponding to each power parameter information;

[0009] According to each triangular differential eigenvalue, a power parameter characteristic diagram of the power electronic equipment is constructed;

[0010] The parameter fault diagnosis model is used to process the power parameter characteristic diagram to obtain the fault diagnosis result of the power electronic equipment.

[0011] In one embodiment, obtaining an arccosine value and an arcsine value corresponding to each power parameter information in a power parameter information set of a power electronic device includes:

[0012] Normalizing each power parameter information to obtain each normalized result;

[0013] The arccosine values ​​and the arcsine values ​​of the normalized results are determined.

[0014] In one embodiment, the trigonometric differential eigenvalues ​​of each power parameter information are determined according to the arccosine value and the arcsine value corresponding to each power parameter information, including:

[0015] Subtract each arccosine value from each arcsine value to obtain each target result;

[0016] Multiply each normalized result with each target result to obtain each specified result;

[0017] According to each specified result, the triangular difference eigenvalue of each power parameter information is determined.

[0018] In one embodiment, determining the triangular differential eigenvalue of each power parameter information according to each specified result includes:

[0019] Each normalized result is first subjected to an arcsine transform and then a cosine transform, or first subjected to an arccosine transform and then a sine transform, to obtain each preset result;

[0020] Each preset result is multiplied with each designated result correspondingly to obtain the triangular differential eigenvalue of each power parameter information.

[0021] In one embodiment, a parameter fault diagnosis model is used to process a power parameter characteristic diagram to obtain a fault diagnosis result of a power electronic device, including:

[0022] Perform image switching on the power parameter feature map to obtain multiple image blocks, and map each image block to an embedding vector of each fixed dimension;

[0023] Incorporate the positional encoding into each embedding vector and add the image category vector to obtain each target vector;

[0024] Calculate the attention between each target vector to obtain the encoding of each image block;

[0025] The fully connected layer is used to process the encoding of each image block to obtain the fault diagnosis results of the power electronic equipment.

[0026] In one embodiment, constructing a power parameter characteristic diagram of a power electronic device according to each triangular differential eigenvalue includes:

[0027] Determine a preset value set according to the accuracy level of the power electronic device; wherein the preset value set includes a plurality of values ​​varying from a minimum preset value to a preset maximum value with a preset variation range;

[0028] Mapping each triangular differential eigenvalue into a preset value set to obtain each preset value;

[0029] Arrange the preset values ​​according to the sampling point order and the preset pixel size to obtain a specified image;

[0030] Perform color rendering on the specified image to obtain the power parameter characteristic diagram of the power electronic equipment.

[0031] In one of the embodiments, the power parameter information includes at least two types of parameter information, and correspondingly, the power parameter characteristic diagram of the power electronic device includes at least two characteristic diagrams;

[0032] The parameter fault diagnosis model is used to process the power parameter characteristic diagram to obtain the fault diagnosis results of the power electronic equipment, including:

[0033] Using a parameter fault diagnosis model to process at least two characteristic graphs respectively, to obtain at least two fault diagnosis results;

[0034] In the case that at least one fault result exists in the at least two fault diagnosis results, it is determined that the fault diagnosis result of the power electronic device is a fault result.

[0035] In a second aspect, the present application also provides a fault diagnosis device, the device comprising:

[0036] An acquisition unit, used to acquire an arccosine value and an arcsine value corresponding to each power parameter information in a power parameter information set of a power electronic device;

[0037] A determination unit, configured to determine a trigonometric differential eigenvalue of each power parameter information according to an arccosine value and an arcsine value corresponding to each power parameter information;

[0038] A construction unit, used for constructing a power parameter characteristic diagram of the power electronic device according to each triangular differential eigenvalue;

[0039] The processing unit is used to process the power parameter characteristic diagram using a parameter fault diagnosis model to obtain a fault diagnosis result of the power electronic equipment.

[0040] In a third aspect, the present application further provides a fault diagnosis device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0041] In a fourth aspect, the present application also provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when the computer program is executed by a processor.

[0042] In the embodiment of the present application, the trigonometric differential eigenvalues ​​of each power parameter information are determined according to the arccosine value and arcsine value corresponding to each power parameter information. The arcsine function and the arccosine function show two opposite change trends in [-1,1]. That is, in [-1,1], the dependent variable of the arcsine function increases with the increase of the independent variable, and the dependent variable of the arccosine function decreases with the increase of the independent variable. The closer the arcsine function and the arccosine function are to 0, the smaller the rate of change is, and the closer they are to -1 and 1, the larger the rate of change is. The trigonometric differential eigenvalues ​​of each power parameter information determined in this way do not change significantly when they are closer to 0, and change significantly when they are closer to -1 and 1. In this way, the power closer to the median of the power parameter information The parameter information, the corresponding triangular differential eigenvalue changes less, and the power parameter information that is farther away from the median of the power parameter information, the corresponding triangular differential eigenvalue changes more. Therefore, the fault characteristics are highlighted and the noise signal is suppressed, so that the accuracy of the fault diagnosis result of the determined power electronic device can be improved according to each triangular differential eigenvalue; in addition, according to each triangular differential eigenvalue, a power parameter characteristic diagram of the power electronic device is constructed, and the power parameter characteristic diagram is processed by a parameter fault diagnosis model to obtain the fault diagnosis result of the power electronic device, so that through the parameter fault diagnosis model, a more accurate diagnosis of the power electronic device can be achieved, further improving the accuracy of the fault diagnosis result of the determined power electronic device. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0044] FIG1 is an application environment diagram of a fault diagnosis method in one embodiment;

[0045] FIG2 is a schematic flow chart of a fault diagnosis method provided in the first embodiment;

[0046] FIG3 is a schematic flow chart of a fault diagnosis method provided by the second embodiment;

[0047] FIG4 is a schematic flow chart of a fault diagnosis method provided in the third embodiment;

[0048] FIG5 is a schematic flow chart of a fault diagnosis method provided in a fourth embodiment;

[0049] FIG6 is a schematic flow chart of a fault diagnosis method provided in the fifth embodiment;

[0050] FIG7 is a schematic diagram of the structure of a parameter fault diagnosis model provided by some embodiments;

[0051] FIG8 is a schematic flow chart of a fault diagnosis method provided in the sixth embodiment;

[0052] FIG9 is a schematic diagram of the structure of a fault diagnosis device provided in some embodiments;

[0053] FIG. 10 is a schematic diagram of the structure of a fault diagnosis device provided in some embodiments. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] The fault diagnosis method provided in the embodiments of the present application can be applied to a fault diagnosis system. FIG1 is a schematic diagram of the structure of a fault diagnosis system provided in some embodiments. As shown in FIG1 , the fault diagnosis system 100 includes a power electronic device 101, a fault diagnosis device 102, and a display device 103. Among them, the power electronic device 101 communicates with the fault diagnosis device 102, and the fault diagnosis device 102 communicates with the display device 103. The fault diagnosis device 102 can output the obtained fault diagnosis result to the display device 103 so that the display device 103 displays the fault diagnosis result. In some embodiments, when the fault diagnosis device 102 determines that the fault diagnosis result is fault information, it can output an alarm message to the display device 103, or it can output an alarm sound.

[0056] The power electronic device in any embodiment of the present application may include a power electronic device. Exemplarily, the power electronic device may include at least one of the following: a transformer, a rectifier, an inverter, a circuit breaker, a motor, a power cable, etc. In some embodiments, the fault diagnosis device may include a combination of one or at least two of the following: a computer, a monitoring device, a server, an Internet of Things device, a console, etc. The Internet of Things device may be a smart speaker, a smart TV, a smart air conditioner, a smart car device, a projection device, etc. The portable wearable device may be a smart watch, a smart bracelet, a head-mounted device, etc. The server may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0057] FIG. 2 is a flow chart of a fault diagnosis method provided in the first embodiment. As shown in FIG. 2 , the method is applied to a fault diagnosis device. The method includes:

[0058] S201: Obtain an arccosine value and an arcsine value corresponding to each power parameter information in a power parameter information set of a power electronic device.

[0059] The power electronic device may include at least one component to be detected, and the power parameter information set of the power electronic device may be: the power parameter information set of each component to be detected in the at least one component to be detected. Exemplarily, the power parameter information set of the power electronic device may be: the power parameter information set of a rectifier. Also exemplary, the power parameter information set of the power electronic device may be: the power parameter information set of a rectifier and the power parameter information set of an inverter.

[0060] The power parameter information may include at least one type of parameter information. Exemplarily, the at least one type of parameter information may include at least one of the following: a current information set, a voltage information set, a temperature information set, a vibration amplitude set, a power information set, a frequency information set, and the like.

[0061] Each parameter information of each component to be detected corresponds to an arccosine value and an arcsine value. Exemplarily, the component to be detected is a rectifier, the power parameter information includes current information, and each current information of the rectifier collected corresponds to an arccosine value and an arcsine value.

[0062] Each power parameter information of the power electronic device may be information collected by the fault diagnosis device, or may be information transmitted by the fault diagnosis device receiving the sensor. Exemplarily, the power parameters of the component to be detected may be collected once every preset time length to obtain the corresponding parameter information. For example, the preset time length may be 1 second, 2 seconds, 5 seconds, 1 minute or 5 minutes, etc. Exemplarily, the preset time length may be determined based on the ambient temperature information and / or the ambient humidity information. When the ambient temperature information is not within the preset temperature range, the preset time length is the first time length. When the ambient temperature information is within the preset temperature range, the preset time length is the second time length. The first time length is less than the second time length.

[0063] The power parameter information set of the power electronic device may be a set of power parameter information of the power electronic device collected at every preset step length. The preset step length may be equal to the number of power parameter information in the power parameter information set, or the preset step length may be less than the number of power parameter information in the power parameter information set. Exemplarily, the power parameter information collected for the 1st to 1500th times is one power parameter information set, and the power parameter information collected for the 1501st to 3000th times is another power parameter information set. In this case, the preset step length is the duration corresponding to 1500 collections, for example, 1500 seconds. Exemplarily, the power parameter information collected for the 1st to 1500th times is one power parameter information set, and the power parameter information collected for the 500th to 2000th times is another power parameter information set. In this case, the preset step length is the duration corresponding to 500 collections, for example, 500 seconds.

[0064] S202: Determine the trigonometric differential eigenvalue of each power parameter information according to the arccosine value and arcsine value corresponding to each power parameter information.

[0065] The units of the arccosine and arcsine are radians.

[0066] Exemplarily, the trigonometric differential eigenvalues ​​of each power parameter information are determined according to the corresponding difference between the arccosine value and the arcsine value corresponding to each power parameter information. For example, the corresponding difference between the arccosine value and the arcsine value corresponding to each power parameter information is determined as the trigonometric differential eigenvalue of each power parameter information.

[0067] In another exemplary embodiment, the trigonometric differential eigenvalue of each power parameter information is determined according to the corresponding ratio of the arccosine value and the arcsine value corresponding to each power parameter information. For example, the corresponding ratio of the arccosine value and the arcsine value corresponding to each power parameter information is determined as the trigonometric differential eigenvalue of each power parameter information.

[0068] In another exemplary embodiment, the trigonometric differential eigenvalue of each power parameter information is determined according to the corresponding difference between the square of the arccosine value and the square of the arcsine value corresponding to each power parameter information. For example, the corresponding difference between the square of the arccosine value and the square of the arcsine value corresponding to each power parameter information is determined as the trigonometric differential eigenvalue of each power parameter information.

[0069] In another exemplary embodiment, the trigonometric differential eigenvalue of each power parameter information is determined according to the corresponding ratio between the square of the arccosine value and the square of the arcsine value corresponding to each power parameter information. For example, the corresponding ratio between the square of the arccosine value and the square of the arcsine value corresponding to each power parameter information is determined as the trigonometric differential eigenvalue of each power parameter information.

[0070] S203: construct a power parameter characteristic diagram of the power electronic device according to each triangular differential eigenvalue.

[0071] Exemplarily, the power parameter characteristic map may include a grayscale map, a heat map, or other types of maps. The following takes a grayscale map as an example to illustrate the implementation steps of S203: map each triangular differential eigenvalue to an integer range of 0-255 to obtain each pixel value, and arrange each pixel value into M×N pixel values ​​in the order of sampling points to obtain a grayscale map. M and N are both integers greater than or equal to 2. Among them, each triangular differential eigenvalue is mapped to an integer range of 0-255 to obtain each pixel value, which can be determined by the following formula: p i = ceil((L i -L min ) / (L max -L min )*255), or p i =floor((L i -L min ) / (L max -L min )*255), or p i = round((L i -L min ) / (L max -L min )*255). Among them, p i Indicates the pixel value, ceil means rounding up; floor means rounding down, round means rounding up, L i represents the triangular differential eigenvalues, L max Indicates the maximum value of each triangular differential eigenvalue, L min Represents the minimum value of each triangular difference eigenvalue.

[0072] S204: Use a parameter fault diagnosis model to process the power parameter characteristic diagram to obtain a fault diagnosis result of the power electronic equipment.

[0073] The fault diagnosis result of the power electronic device may include at least one of the following: fault state information, alarm state information, and normal state information.

[0074] In some embodiments, the parameter fault diagnosis model may include Vision Transformer (also known as ViT). In other embodiments, the parameter fault diagnosis model may include other models, such as Swin Transformer.

[0075] In the embodiment of the present application, the trigonometric differential eigenvalues ​​of each power parameter information are determined according to the arccosine value and arcsine value corresponding to each power parameter information. The arcsine function and the arccosine function show two opposite change trends in [-1,1]. In [-1,1], the dependent variable of the arcsine function increases with the increase of the independent variable, and the dependent variable of the arccosine function decreases with the increase of the independent variable. The closer the arcsine function and the arccosine function are to 0, the smaller the rate of change is, and the closer they are to -1 and 1, the larger the rate of change is. The trigonometric differential eigenvalues ​​of each power parameter information determined in this way do not change significantly when they are closer to 0, and change significantly when they are closer to -1 and 1. In this way, the power closer to the median of the power parameter information The parameter information, the corresponding triangular differential eigenvalue changes less, and the power parameter information that is farther away from the median of the power parameter information, the corresponding triangular differential eigenvalue changes more. Therefore, the fault characteristics are highlighted and the noise signal is suppressed, so that the accuracy of the fault diagnosis result of the determined power electronic device can be improved according to each triangular differential eigenvalue; in addition, according to each triangular differential eigenvalue, a power parameter characteristic diagram of the power electronic device is constructed, and the power parameter characteristic diagram is processed by a parameter fault diagnosis model to obtain the fault diagnosis result of the power electronic device, so that through the parameter fault diagnosis model, a more accurate diagnosis of the power electronic device can be achieved, further improving the accuracy of the fault diagnosis result of the determined power electronic device.

[0076] FIG3 is a flow chart of a fault diagnosis method provided in the second embodiment. As shown in FIG3 , the method is applied to a fault diagnosis device. The difference between the method in the embodiment of FIG2 is that S201 includes the following steps:

[0077] S2011 . Normalize the acquired power parameter information of the power electronic equipment to obtain normalized results.

[0078] In some embodiments, S2011 can be implemented in the following manner: determine the target interval [A, B], set the power parameter information less than A to A, set the power parameter information greater than B to B, and keep other values ​​unchanged to obtain each adjustment parameter information, so that the value range of each adjustment parameter information is [A, B], and further normalize each adjustment parameter information to obtain each normalized result. Among them, normalizing each adjustment parameter information can be normalizing each adjustment parameter information to [-1, 1], and the determination formula is: q i = (M i -M mean ) / (M max -M min ), where q i Represents each normalized result, M i Indicates the information of each adjustment parameter, Mmean Represents the average value of each adjustment parameter information, M max Indicates the maximum value of each adjustment parameter information, M min Indicates the minimum value of each adjustment parameter information.

[0079] In some embodiments, before S201, the power parameter information of the target quantity of the power electronic device during normal operation can be obtained in advance, and the target interval [A, B] can be determined based on the maximum and minimum values ​​of the power parameter information of the target quantity. For example, the maximum value plus the first value is added to obtain B, and the minimum value is subtracted from the first value to obtain A. In some embodiments, the first value can be obtained according to the difference between the maximum value and the minimum value (hereinafter referred to as the target difference). Exemplarily, the first value can be a preset positive integer multiple of the target difference, or the first value can be the target difference divided by a preset positive integer multiple. For example, the first value can be the target difference, or the first value can be half of the target difference, or the first value can be twice the target difference.

[0080] In some other embodiments, the normalization process of each power parameter information in S2011 may be to normalize each power parameter information to [-1, 1], and the determination formula is: q i = (N i -N mean ) / (N max -N min ), where q i Represents each normalized result, N i Indicates various power parameter information, N mean Indicates the average value of each power parameter information, N max Indicates the maximum value of each power parameter information, N min Indicates the minimum value of each power parameter information.

[0081] S2012. Determine each arccosine value and each arcsine value of each normalized result.

[0082] For example, the normalized result q i The arc cosine value is q i The arc sine value is q i .

[0083] FIG4 is a flow chart of a fault diagnosis method provided in the third embodiment. As shown in FIG4 , the method is applied to a fault diagnosis device. The difference between the method in the embodiment of FIG2 is that S202 includes the following steps:

[0084] S2021. Subtract each arccosine value from each arcsine value to obtain each target result.

[0085] S2022. Multiply each normalized result with each target result correspondingly to obtain each designated result.

[0086] Among them, each normalized result is obtained in the following manner: each power parameter information is normalized to obtain each normalized result.

[0087] S2023. Determine the triangular differential eigenvalues ​​of each power parameter information according to each specified result.

[0088] In some other embodiments other than FIG. 4 , the trigonometric differential eigenvalues ​​of each power parameter information may also be determined in the following manner, for example, by subtracting each arccosine value from each arcsine value to obtain each target result, and determining the trigonometric differential eigenvalues ​​of each power parameter information according to each target result. Exemplarily, each normalized result is first subjected to an arcsine transform and then a cosine transform, or first subjected to an arccosine transform and then a sine transform, to obtain each preset result; each preset result is correspondingly multiplied with each target result to obtain the trigonometric differential eigenvalues ​​of each power parameter information. Again exemplarily, each target result is determined as the trigonometric differential eigenvalue of each power parameter information.

[0089] In some embodiments, S2023 may include: performing an inverse sine transform and then a cosine transform on each normalized result, or performing an inverse cosine transform and then a sine transform on each normalized result, to obtain each preset result; multiplying each preset result with each specified result to obtain a trigonometric differential eigenvalue of each power parameter information.

[0090] In some other embodiments, S2023 may include: determining each designated result as a triangular differential eigenvalue of each power parameter information.

[0091] FIG5 is a flow chart of a fault diagnosis method provided in the fourth embodiment. As shown in FIG5 , the method is applied to a fault diagnosis device. The difference between the method in the embodiment of FIG2 is that S204 includes the following steps:

[0092] S2041. Perform image switching on the power parameter feature map to obtain multiple image blocks, and map each image block to an embedding vector of a fixed dimension.

[0093] S2042. Incorporate the position code into each embedding vector and add the image category vector to obtain each target vector.

[0094] S2043. Calculate the attention between each target vector to obtain the encoding of each image block.

[0095] S2044. Use a fully connected layer to process the encoding of each image block to obtain a fault diagnosis result of the power electronic equipment.

[0096] FIG6 is a flowchart of a fault diagnosis method provided in the fifth embodiment. As shown in FIG6 , the method is applied to a fault diagnosis device. The difference between the method in the embodiment of FIG2 is that S203 includes the following steps:

[0097] S2031. Determine a preset value set according to the accuracy level of the power electronic equipment.

[0098] The preset value set includes a plurality of values ​​varying from a minimum preset value to a preset maximum value with a preset variation range.

[0099] In some embodiments, the minimum preset value and the maximum preset value may be preset. For example, the minimum preset value and the maximum preset value may be 0 and 255, respectively, or the minimum preset value and the maximum preset value may be the minimum pixel value and the maximum pixel value that the fault diagnosis device can render in color, respectively.

[0100] The accuracy level of the power electronic device may be preconfigured. In some embodiments, when the accuracy level is higher, there are more different power parameter information in the corresponding power parameter information set, and accordingly, the preset change amplitude may be set smaller, so that the features of the determined power parameter characteristic diagram are richer, and the fault diagnosis result of the power electronic device determined by the power parameter characteristic diagram is more accurate.

[0101] In some embodiments, the preset change range can also be determined based on the acquisition accuracy of the sensor, which is used to collect power parameter information of the power electronic device. In some embodiments, the preset change range can also be determined based on the computing power of the acquisition fault diagnosis device.

[0102] S2032. Map each triangular differential eigenvalue to a preset value set to obtain each preset value.

[0103] In some embodiments, S2032 may be determined by the following formula: i = ceil((S i -S min ) / (S max -S min )*E*F), or r i =floor((S i -S min ) / (S max -S min )*E*F), or r i = round((S i -S min ) / (S max -S min )*E*F). Among them, r iIndicates the preset values. Ceil means rounding up; floor means rounding down; round means rounding up to the nearest integer. i represents the eigenvalues ​​of each triangular difference, S max Indicates the maximum value of the preset value set, S min represents the minimum value of the preset value set, E represents the difference between the maximum value and the minimum value of the preset value set, and F represents the preset change range.

[0104] S2033. Arrange the preset values ​​according to the sampling point sequence and the preset pixel size to obtain a designated image.

[0105] S2034: Perform color rendering on the designated image to obtain a power parameter characteristic diagram of the power electronic device.

[0106] In some embodiments, the power parameter information includes at least two types of parameter information, and correspondingly, the power parameter characteristic diagram of the power electronic device includes at least two characteristic diagrams; the power parameter characteristic diagram is processed by a parameter fault diagnosis model to obtain a fault diagnosis result of the power electronic device, including: using the parameter fault diagnosis model to process the at least two characteristic diagrams respectively to obtain at least two fault diagnosis results; when there is at least one fault result in the at least two fault diagnosis results, determining that the fault diagnosis result of the power electronic device is a fault result.

[0107] For example, the power parameter information includes at least two types of parameter information, namely current value and voltage value. Then, the current value set corresponds to the first characteristic graph, the voltage value set corresponds to the second characteristic graph, the parameter fault diagnosis model is used to process the first characteristic graph, and the fault diagnosis result of the power electronic device is obtained as a fault result. The parameter fault diagnosis model is used to process the first characteristic graph, and the fault diagnosis result of the power electronic device is obtained as a normal result, and it is determined that the fault diagnosis result of the power electronic device is a fault result.

[0108] In the embodiments of the present application, technicians have discovered that the differences between the fault characteristics of parametric faults are small, which makes it difficult for general diagnostic methods to accurately diagnose parametric faults of power electronic equipment.

[0109] In order to accurately diagnose parametric faults in power electronic equipment, commonly used diagnostic methods include feature extraction methods and diagnostic models. In the existing field of power electronic circuit fault diagnosis, traditional time-frequency domain feature analysis methods such as Fourier transform, EMD decomposition, wavelet analysis, Hilbert-Huang transform, etc. only perform simple digital feature analysis and calculation on the output voltage or current analog quantity of the circuit. In recent years, with the development of artificial intelligence and machine vision technology, machine vision detection based on artificial intelligence algorithms has gradually gained attention in the field of fault diagnosis. However, currently in the field of machine vision fault diagnosis, most of them are still aimed at some appearance defects and abnormal equipment or component failures. For power electronic circuits with weak or complex features or unclear fault phenomena, the fault diagnosis method is inadequate. Therefore, the fault diagnosis method provided in the embodiment of the present application adopts a weighted angular difference feature extraction (WADFE) scheme. WADFE is an arccosine value and an arcsine value obtained by using the current signal of the power electronic device. The difference value is obtained by using the trigonometric difference formula to achieve the purpose of extracting important features of the current signal, which can highlight the fault features and suppress the noise signal. According to each trigonometric difference feature value, the accuracy of the fault diagnosis result of the determined power electronic device can be improved.

[0110] In view of the problem that the parametric fault diagnosis method of power electronic equipment cannot effectively extract fault features and the diagnostic model is difficult to establish a mapping relationship between characteristic signals and parametric fault types, an embodiment of the present application proposes a fault prediction and health management (PHM) method for parametric faults of power electronic equipment based on WADFE and a parameter fault diagnosis model. The method collects power parameter information (such as current signals) of power electronic equipment, obtains the arccosine value and arcsine value corresponding to each sampling point after normalization, and obtains the eigenvalue through the trigonometric function value difference formula, that is, after the eigenvalue conversion by WADFE, the fault features in the current signal are highlighted and the noise signal is suppressed. Then, the trigonometric differential eigenvalue after the eigenvalue conversion by WADFE is converted into a power parameter characteristic map (such as a grayscale map or a thermal map) as the input feature image of the parameter fault diagnosis model, so as to achieve accurate diagnosis of parametric faults of power electronic equipment.

[0111] Aiming at the problems that the characteristic changes between different degrees of aging of power electronic devices are not obvious, and conventional methods are difficult to effectively extract degradation characteristics, resulting in inaccurate diagnosis of parametric faults of power electronic equipment and low identification accuracy, a PHM method for parametric faults of power electronic equipment based on WADFE and parameter fault diagnosis model is proposed.

[0112] In the embodiment of the present application, after the current signal of the power electronic device is collected, the characteristic signal is first normalized by a normalization formula: ,in represents the original current signal, represents the normalized eigenvalue, and then the arc cosine value of the normalized eigenvalue is calculated: ,in, Represents the arccosine value corresponding to the eigenvalue, and finds the arcsine value ,in, Represents the arccosine value corresponding to the eigenvalue, and then constructs the trigonometric difference feature extraction formula through the trigonometric difference formula: , the arccosine value and arcsine value corresponding to the eigenvalue are substituted into the trigonometric difference feature extraction formula to output the trigonometric difference eigenvalue, which is the trigonometric difference eigenvalue of each power parameter information.

[0113] In some embodiments, equal , equal ,but Equivalent to .

[0114] because The values ​​are the same, so Also equivalent to ( )or ( ).

[0115] Finally, the triangular differential eigenvalues ​​are directly converted into grayscale images as the input feature images of the parametric fault diagnosis model.

[0116] Assume that the length, width and height of the input image are H, W and C respectively, P represents the size of the image block, C represents the number of image channels (the number of channels in the embodiment of the present application may be 1), FIG7 is a schematic diagram of the structure of a parameter fault diagnosis model provided by some embodiments, and the specific steps are as follows:

[0117] Step 1: Image segmentation. First, segment the image evenly into image blocks (e.g., image block a value i), each two-dimensional image block is flattened into a one-dimensional vector by linear projection after flattening. The dimension of each vector is .

[0118] Step 2: Image block encoding. Each flattened image block is mapped to a D-dimensional (i.e., the fixed dimension) vector by linear projection after flattening, and a position code (e.g., position 1 to 9) is added to each image block to indicate its original position in the image, which is used to calculate the attention between image blocks. In addition, a block numbered 0 is added separately to calculate the image category.

[0119] Step 3: Self-attention encoding. Using the self-attention mechanism in Transformer, we use multi-head self-attention (Multi-Head Attention) and multi-layer perceptron (MLP) and layer normalization (Norm) to calculate the attention between image blocks and obtain the encoding of each image block.

[0120] Step 4: Calculate the image category. The image block encoding obtained by the Transformer encoder is input into the fully connected layer. After dimension conversion, the Softmax function is used to obtain the probability vector of the image belonging to each category, and the maximum value is taken to obtain the image category (that is, the classification result).

[0121] The Transformer multi-head attention mechanism is a sequence-based network architecture. Given a query element query and a set of key elements, the multi-head attention module weighs the attention weights to adaptively aggregate the compatibility of key and query key pairs. Let q represent the query element, and its features are recorded as , Represents the key element, whose characteristics are recorded as , where C is the feature dimension, and Represent the collection of query elements and key elements respectively. The formula of the multi-head attention mechanism is: ,in: is the number of heads of multi-head attention, All are learnable parameters, by default .

[0122] Among them, the attention weight , normalized to ,in and , which is a learnable parameter.

[0123] The embodiment of the present application constructs a PHM method for parametric fault diagnosis of power electronic equipment through WADFE, grayscale conversion and parameter fault diagnosis model. For example, the current signal of the power electronic equipment is converted into a grayscale image after feature extraction by WADFE, which is used as the input feature image of the parameter fault diagnosis model to achieve accurate diagnosis of parametric faults of power electronic equipment. The embodiment of the present application has the following advantages: the original current signal is extracted with time series features by WADFE, and the fault features in the fault signal are highlighted and the noise is suppressed; the feature signal after WADFE will be directly converted into a grayscale image as the input feature image of the diagnosis model; the feature extraction method WADFE is combined with the parameter fault diagnosis model to maximize the mining of fault features in the feature image that can represent different types of parametric faults.

[0124] FIG8 is a flow chart of a fault diagnosis method provided in the sixth embodiment. As shown in FIG8 , the method is applied to a fault diagnosis device. The method includes:

[0125] S801. Acquire multiple current signals of power electronic equipment.

[0126] S802: Normalize the current signal to obtain a normalized result.

[0127] S803, obtaining an arccosine value of the normalized result, and obtaining an arcsine value of the normalized result.

[0128] S804: Determine trigonometric differential feature extraction according to the arccosine value and the arcsine value to obtain trigonometric differential feature values.

[0129] S805: Convert the triangular differential eigenvalues ​​into a power parameter characteristic diagram.

[0130] S806: Use the parameter fault diagnosis model to process the power parameter characteristic diagram to obtain a fault diagnosis result of the power electronic equipment.

[0131] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0132] Based on the same inventive concept, the embodiment of the present application also provides a fault diagnosis device for implementing the fault diagnosis method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more fault diagnosis device embodiments provided below can refer to the limitations on the fault diagnosis method above, and will not be repeated here.

[0133] In an exemplary embodiment, FIG9 is a schematic diagram of the structure of a fault diagnosis device provided in some embodiments. As shown in FIG9 , the fault diagnosis device 900 includes:

[0134] An acquisition unit 901 is used to acquire an arccosine value and an arcsine value corresponding to each power parameter information in a power parameter information set of a power electronic device;

[0135] A determination unit 902, configured to determine a trigonometric differential eigenvalue of each power parameter information according to an arccosine value and an arcsine value corresponding to each power parameter information;

[0136] A construction unit 903 is used to construct a power parameter characteristic diagram of the power electronic device according to each triangular differential eigenvalue;

[0137] The processing unit 904 is used to process the power parameter characteristic diagram using the parameter fault diagnosis model to obtain a fault diagnosis result of the power electronic device.

[0138] In some embodiments, the acquisition unit 901 is further used to: perform normalization processing on each of the power parameter information to obtain each normalization result; and determine each arccosine value and each arcsine value of each of the normalization results.

[0139] In some embodiments, the determination unit 902 is also used to: subtract each arccosine value from each arcsine value to obtain each target result; multiply each normalized result by each target result to obtain each specified result; and determine the trigonometric differential eigenvalue of each power parameter information according to each specified result.

[0140] In some embodiments, the determination unit 902 is also used to: perform an inverse sine transform and then a cosine transform on each normalized result, or perform an inverse cosine transform and then a sine transform on each normalized result, to obtain each preset result; and multiply each preset result with each specified result to obtain a trigonometric differential eigenvalue of each power parameter information.

[0141] In some embodiments, the processing unit 904 is further used to: perform image switching on the power parameter feature map to obtain multiple image blocks, and map each image block to an embedding vector of a fixed dimension; incorporate a position code into each embedding vector, and add an image category vector to obtain each target vector; calculate the attention between each target vector to obtain the code of each image block; and use a fully connected layer to process the code of each image block to obtain a fault diagnosis result of the power electronic device.

[0142] In some embodiments, the construction unit 903 is also used to: determine a preset value set according to the accuracy level of the power electronic device; wherein the preset value set includes multiple values ​​that change from a minimum preset value to a preset maximum value with a preset change amplitude; map each triangular differential eigenvalue to the preset value set to obtain each preset value; arrange each preset value according to the sampling point order and the preset pixel size to obtain a specified image; perform color rendering on the specified image to obtain a power parameter characteristic diagram of the power electronic device.

[0143] In some embodiments, the power parameter information includes at least two types of parameter information, and correspondingly, the power parameter characteristic diagram of the power electronic device includes at least two characteristic diagrams; the processing unit 904 is also used to: use a parameter fault diagnosis model to process the at least two characteristic diagrams respectively to obtain at least two fault diagnosis results; when there is at least one fault result in the at least two fault diagnosis results, determine that the fault diagnosis result of the power electronic device is a fault result.

[0144] Each module in the above-mentioned fault diagnosis device can be implemented in whole or in part by software, hardware and a combination thereof. Each module can be embedded in or independent of the processor in the fault diagnosis device in the form of hardware, or can be stored in the memory in the fault diagnosis device in the form of software, so that the processor can call and execute the operations corresponding to each module above.

[0145] In an exemplary embodiment, a fault diagnosis device is provided. FIG10 is a schematic diagram of the structure of a fault diagnosis device provided by some embodiments. The fault diagnosis device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the fault diagnosis device is used to provide computing and control capabilities. The memory of the fault diagnosis device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the fault diagnosis device is used to exchange information between the processor and an external device. The communication interface of the fault diagnosis device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a fault diagnosis method is implemented. The display unit of the fault diagnosis device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the fault diagnosis equipment can be a touch layer covering the display screen, or a button, trackball or touchpad set on the shell of the fault diagnosis equipment, or an external keyboard, touchpad or mouse.

[0146] Those skilled in the art will understand that the structure shown in FIG. 10 is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the fault diagnosis device to which the scheme of the present application is applied. The specific fault diagnosis device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0147] In an exemplary embodiment, a fault diagnosis device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the following steps when executing the computer program: obtaining an arccosine value and an arcsine value corresponding to each power parameter information in a power parameter information set of the power electronic device; determining a trigonometric differential eigenvalue of each power parameter information according to the arccosine value and the arcsine value corresponding to each power parameter information; constructing a power parameter characteristic diagram of the power electronic device according to each trigonometric differential eigenvalue; and processing the power parameter characteristic diagram using a parameter fault diagnosis model to obtain a fault diagnosis result of the power electronic device.

[0148] In one embodiment, when the processor executes the computer program, the following steps are also implemented: normalizing each power parameter information to obtain each normalized result; determining each arccosine value and each arcsine value of each normalized result; and determining the trigonometric differential eigenvalue of each power parameter information based on each arccosine value and each arcsine value.

[0149] In one embodiment, when the processor executes the computer program, the following steps are also implemented: subtracting each arccosine value from each arcsine value to obtain each target result; multiplying each normalized result by each target result to obtain each specified result; and determining the trigonometric differential eigenvalue of each power parameter information based on each specified result.

[0150] In one embodiment, when the processor executes the computer program, the following steps are also implemented: each normalized result is first subjected to an arcsine transformation and then a cosine transformation, or first subjected to an arccosine transformation and then a sine transformation, to obtain each preset result; each preset result is multiplied by each designated result to obtain a trigonometric differential eigenvalue of each power parameter information.

[0151] In one embodiment, when the processor executes the computer program, the following steps are also implemented: performing image switching on the power parameter characteristic diagram to obtain multiple image blocks, and mapping each image block to an embedding vector of a fixed dimension; incorporating a position code into each embedding vector, and adding an image category vector to obtain each target vector; calculating the attention between each target vector to obtain the code of each image block; and using a fully connected layer to process the code of each image block to obtain a fault diagnosis result of the power electronic device.

[0152] In one embodiment, when the processor executes the computer program, the following steps are also implemented: determining a preset value set according to the accuracy level of the power electronic device; wherein the preset value set includes multiple values ​​that change from a minimum preset value to a preset maximum value with a preset change amplitude; mapping each triangular differential eigenvalue to the preset value set to obtain each preset value; arranging each preset value according to the sampling point order and the preset pixel size to obtain a specified image; and color rendering the specified image to obtain a power parameter characteristic diagram of the power electronic device.

[0153] In one embodiment, the power parameter information includes at least two types of parameter information, and correspondingly, the power parameter characteristic diagram of the power electronic device includes at least two characteristic diagrams; when the processor executes the computer program, the following steps are also implemented: using a parameter fault diagnosis model to process the at least two characteristic diagrams respectively to obtain at least two fault diagnosis results; when there is at least one fault result in the at least two fault diagnosis results, determining that the fault diagnosis result of the power electronic device is a fault result.

[0154] In one embodiment, a computer-readable storage medium is provided, and when a computer program is executed by a processor, the steps of the method provided in any of the above embodiments are implemented.

[0155] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: obtaining an arccosine value and an arcsine value corresponding to each power parameter information in a power parameter information set of a power electronic device; determining a trigonometric differential eigenvalue of each power parameter information according to the arccosine value and the arcsine value corresponding to each power parameter information; constructing a power parameter characteristic diagram of the power electronic device according to each trigonometric differential eigenvalue; and processing the power parameter characteristic diagram using a parameter fault diagnosis model to obtain a fault diagnosis result of the power electronic device.

[0156] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: normalizing each power parameter information to obtain each normalized result; determining each arccosine value and each arcsine value of each normalized result; and determining the trigonometric differential eigenvalue of each power parameter information based on each arccosine value and each arcsine value.

[0157] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: subtracting each arccosine value from each arcsine value to obtain each target result; multiplying each normalized result by each target result to obtain each specified result; and determining the trigonometric differential eigenvalue of each power parameter information based on each specified result.

[0158] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: each normalized result is first subjected to an arcsine transformation and then a cosine transformation, or first subjected to an arccosine transformation and then a sine transformation, to obtain each preset result; each preset result is multiplied by each designated result to obtain a trigonometric differential eigenvalue of each power parameter information.

[0159] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: performing image switching on the power parameter feature map to obtain multiple image blocks, and mapping each image block to an embedding vector of a fixed dimension; incorporating a position code into each embedding vector, and adding an image category vector to obtain each target vector; calculating the attention between each target vector to obtain the code of each image block; and using a fully connected layer to process the code of each image block to obtain a fault diagnosis result of the power electronic device.

[0160] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: determining a preset value set according to the accuracy level of the power electronic device; wherein the preset value set includes multiple values ​​that change from a minimum preset value to a preset maximum value with a preset change amplitude; mapping each triangular differential eigenvalue to the preset value set to obtain each preset value; arranging each preset value according to the sampling point order and the preset pixel size to obtain a specified image; and color rendering the specified image to obtain a power parameter characteristic diagram of the power electronic device.

[0161] In one embodiment, the power parameter information includes at least two types of parameter information, and correspondingly, the power parameter characteristic diagram of the power electronic device includes at least two characteristic diagrams; when the computer program is executed by the processor, the following steps are also implemented: using a parameter fault diagnosis model to process the at least two characteristic diagrams respectively to obtain at least two fault diagnosis results; when there is at least one fault result in the at least two fault diagnosis results, determining that the fault diagnosis result of the power electronic device is a fault result.

[0162] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0163] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0164] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A fault diagnosis method, characterized in that: The method comprises: Obtaining an arccosine value and an arcsine value corresponding to each power parameter information in a power parameter information set of a power electronic device; Determine the trigonometric differential eigenvalue of each of the power parameter information according to the arccosine value and the arcsine value corresponding to each of the power parameter information; Constructing a power parameter characteristic diagram of the power electronic device according to each of the triangular differential eigenvalues; The power parameter characteristic diagram is processed using a parameter fault diagnosis model to obtain a fault diagnosis result of the power electronic device.

2. The method according to claim 1, characterized in that The step of obtaining the arccosine value and the arcsine value corresponding to each power parameter information in the power parameter information set of the power electronic device includes: Normalizing each of the power parameter information to obtain each normalized result; The arccosine values ​​and the arcsine values ​​of the normalized results are determined.

3. The method according to claim 2, characterized in that Determining the trigonometric differential eigenvalues ​​of each of the power parameter information according to the arccosine value and the arcsine value corresponding to each of the power parameter information includes: Subtracting each arccosine value from each arcsine value to obtain each target result; Multiplying each of the normalized results with each of the target results correspondingly to obtain each designated result; According to each of the specified results, a triangular differential eigenvalue of each of the power parameter information is determined.

4. The method according to claim 3, characterized in that: Determining the triangular differential characteristic value of each of the power parameter information according to each of the specified results includes: Performing an arcsine transform and then a cosine transform on each of the normalized results, or performing an arccosine transform and then a sine transform on each of the normalized results, to obtain each preset result; Each of the preset results is multiplied by each of the designated results correspondingly to obtain a triangular differential eigenvalue of each of the power parameter information.

5. The method according to any one of claims 1 to 4, characterized in that: The adopting the parameter fault diagnosis model to process the power parameter characteristic diagram to obtain the fault diagnosis result of the power electronic device includes: Performing image switching on the power parameter characteristic graph to obtain a plurality of image blocks, and mapping each of the image blocks to an embedding vector of a fixed dimension; Incorporating a position code into each of the embedding vectors and adding an image category vector to obtain each target vector; Calculating the attention between the target vectors to obtain the encoding of the image blocks; The encoding of each image block is processed using a fully connected layer to obtain a fault diagnosis result of the power electronic device.

6. The method according to any one of claims 1 to 4, characterized in that: The step of constructing a power parameter characteristic diagram of the power electronic device according to each of the triangular differential characteristic values ​​includes: Determine a preset value set according to the accuracy level of the power electronic device; wherein the preset value set includes a plurality of values ​​varying from a minimum preset value to a preset maximum value with a preset variation range; Mapping each of the triangular difference eigenvalues ​​into the preset value set to obtain each preset value; Arrange the preset values ​​according to the sampling point order and the preset pixel size to obtain a specified image; The designated image is rendered in color to obtain a power parameter characteristic diagram of the power electronic device.

7. The method according to any one of claims 1 to 4, characterized in that: The power parameter information includes at least two types of parameter information, and correspondingly, the power parameter characteristic diagram of the power electronic device includes at least two characteristic diagrams; The adopting the parameter fault diagnosis model to process the power parameter characteristic diagram to obtain the fault diagnosis result of the power electronic device includes: Using the parameter fault diagnosis model to process the at least two characteristic graphs respectively to obtain at least two fault diagnosis results; In the case that at least one fault result exists in the at least two fault diagnosis results, it is determined that the fault diagnosis result of the power electronic device is a fault result.

8. A fault diagnosis device, characterized in that: The device comprises: An acquisition unit, used to acquire an arccosine value and an arcsine value corresponding to each power parameter information in a power parameter information set of a power electronic device; A determination unit, configured to determine a trigonometric differential eigenvalue of each of the power parameter information according to an arccosine value and an arcsine value corresponding to each of the power parameter information; A construction unit, configured to construct a power parameter characteristic diagram of the power electronic device according to each of the triangular differential characteristic values; The processing unit is used to process the power parameter characteristic diagram using a parameter fault diagnosis model to obtain a fault diagnosis result of the power electronic device.

9. A fault diagnosis device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images