A bearing fault diagnosis method, system, storage medium and electronic device

Abstract

The present invention provides a bearing fault diagnosis method, system, storage medium and electronic device. The method includes obtaining source domain data and target domain data; denoising the source domain data and the target domain data; extracting features from the source domain data and the target domain data to obtain initial feature vectors; performing initialization processing on the initial feature vectors; iteratively updating nodes of the source domain data and the target domain data to obtain a source domain graph and a target domain graph, and calculating a difference value; calculating the similarity loss between the two, calculating the true distribution probability and the predicted distribution probability of the bearing category to obtain a cross-entropy loss function, obtaining a maximum mean difference function, and calculating a total loss function; calculating gradients of the metric function, the source domain graph and the target domain graph, nodes, and parameters of the graph kernel model; supervising the results, and jointly optimizing the graph kernel model in combination with the graph kernel loss function. The present invention reduces the morphological feature differences between the source domain graph and the target domain graph, and improves the generalization ability of the recognition model.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent fault diagnosis, and particularly relates to a bearing fault diagnosis method, system, storage medium and electronic device. Background Art

[0002] The axle box bearing is one of the most important rotating components in a rail train, and its health status directly affects the performance and service life of the equipment. Early diagnosis of bearing faults can effectively improve the accuracy and timeliness of fault detection, reduce operation and maintenance costs, and improve production efficiency.

[0003] In the prior art, the geometric features between the vibration signals of the axle box bearing are not effectively utilized. Traditional domain adaptation methods mostly rely on distribution metrics in a statistical sense (such as MMD, LMMD, etc.). Such explicit metric functions have limited ability to characterize the non-linear and non-stationary characteristics of vibration signals under complex working conditions. Especially when there are significant modal differences between the source domain and the target domain, it is difficult to effectively capture deep domain-invariant features. Secondly, existing feature alignment strategies mostly focus on shallow statistical matching and lack the ability to model the internal physical correlations of vibration signals, resulting in being easily interfered by local pseudo-correlated features when migrating across devices and working conditions. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a bearing fault diagnosis method, aiming to solve the technical problems mentioned in the background art.

[0005] To achieve the above purpose, the present invention is implemented by the following technical solutions:

[0006] A bearing fault diagnosis method includes the following steps:

[0007] Obtain the signal data of the bearing in a known environment, and use the signal data in the known environment as source domain data. Obtain the signal data of the bearing in other environments, and use the signal data in other environments as target domain data;

[0008] Perform denoising processing on the source domain data and the target domain data;

[0009] Extract features from the denoised source domain data and target domain data through a deep residual neural network combined with a self-attention mechanism to obtain an initial feature vector;

[0010] Regard the initial feature vector as a node, and perform weighted processing on the node in the time channel and the frequency channel to initialize the label embedding of the node;

[0011] Introduce a graph kernel model to iteratively update the nodes after initializing the label embeddings in the source domain data and the target domain data respectively, so as to obtain a source domain graph and a target domain graph, and calculate the difference value between the source domain graph and the target domain graph through a graph morphology difference metric function;

[0012] Calculate the similarity loss between the source domain graph and the target domain graph according to the difference value. Based on the cross-entropy loss function, obtain the true distribution probability of the bearing category according to the source domain data and the target domain data, and through the fully connected layer in the deep residual neural network, obtain the predicted distribution probability of the bearing category during the process of iteratively updating the nodes in the source domain data and the target domain data, so as to calculate the distribution difference loss between the true distribution probability of the bearing category and the predicted distribution probability of the bearing category. Introduce a kernel function, and calculate the distance of the mean embeddings of the source domain data and the target domain data in space through the maximum mean discrepancy function, so as to obtain the mean discrepancy loss, and calculate the total loss function by combining the similarity loss, the distribution difference loss and the mean discrepancy loss;

[0013] Based on the total loss function, calculate the gradients of the graph morphology difference metric function, the source domain graph and the target domain graph, the nodes, and the parameters of the graph kernel model in sequence;

[0014] Use the cross-entropy loss function to perform supervised learning on the classification results during the process of iteratively updating the nodes in the source domain data and the target domain data, and combine the graph morphology difference metric function and the maximum mean discrepancy function to jointly optimize the graph kernel model.

[0015] According to one aspect of the above technical solution, the feature extraction of the denoised source domain data and target domain data through the deep residual neural network combined with the self-attention mechanism to obtain the initial feature vector specifically includes:

[0016] Introduce a deep residual neural network combined with the self-attention mechanism, where the deep residual neural network includes a multi-scale residual feature extraction module, a channel-temporal dual-path attention mechanism module, a normalization layer, a global average pooling layer, and a global maximum pooling layer;

[0017] Use the multi-scale residual feature extraction module to extract the time-domain and frequency-domain features in the denoised source domain data and target domain data;

[0018] Normalize the denoised source domain data and target domain data through the normalization layer;

[0019] Use the global average pooling layer and the global maximum pooling layer to reflect continuous fault features and capture transient shock features;

[0020] Perform global average calculation on the inner channels of the input features in the deep residual neural network through the global average pooling layer to obtain an initial feature vector.

[0021] According to one aspect of the above technical solution, regarding the initial feature vector as a node, performing weighted processing on the node in the time channel and frequency channel to initialize the label embedding of the node, specifically including:

[0022] Regarding the initial feature vector as a node, based on the channel-temporal dual-path attention mechanism module, calculate the time-channel attention feature weight value of the node;

[0023] ;

[0024] ;

[0025] where; represents the node v of the source domain data in the deep residual neural network, represents the node i of the target domain data in the deep residual neural network, W1 and W2 are the full connection layer parameters in the deep residual neural network, represents the activation function, GAP represents the global average pooling operation, and ReLU represents the non-linear activation function;

[0026] Based on the channel-temporal dual-path attention mechanism module, calculate the frequency-channel attention feature weight value of the node;

[0027] ;

[0028] ;

[0029] where W3 is the full connection layer parameter, and GRU is the gated recurrent unit in the deep residual neural network;

[0030] Fuse the time-channel attention feature weight value and the frequency-channel attention feature weight value to obtain a fused feature weight value;

[0031] ;

[0032] ;

[0033] Calculate the internal learning parameter γ in the deep residual neural network;

[0034] ;

[0035] ;

[0036] Among them, GlobalPool(·) is the global average pooling operation, N ( v ) represents the node neighborhood feature; Wr represents the weight matrix, Wr ∈ R (d×2d) , and d represents the dimension of the output feature;

[0037] Initialize the label embedding of the nodes in the source domain data and the target domain data;

[0038] ;

[0039] .

[0040] According to one aspect of the above technical solution, the introduced graph kernel model iteratively updates the nodes after label embedding initialization in the source domain data and the target domain data respectively to obtain a source domain graph and a target domain graph, and calculates the difference value between the source domain graph and the target domain graph through a graph morphological difference metric function, specifically including:

[0041] Take the nodes after label embedding initialization as the initial embedding values, and perform neighbor aggregation calculations layer by layer on the initial embedding values in the source domain data and the target domain data respectively through the graph kernel model;

[0042] ; ;

[0043] Among them, k represents the number of neighbor points in the source domain data, N represents the total number of nodes in the source domain data, represents the embedding value corresponding to node u in the (k - 1)-th layer in the source domain data, represents the neighbor aggregation feature of node v in the k-th layer in the source domain data, j represents the number of neighbor points in the target domain data, C represents the total number of nodes in the target domain data, represents the embedding value corresponding to node f in the (j - 1)-th layer in the target domain data, represents the neighbor aggregation feature of node i in the j-th layer in the target domain data;

[0044] Based on the results of neighbor aggregation calculation, update the nodes with initial embedding in the source domain data and the target domain data respectively;

[0045] ; ;

[0046] Among them, represents the activation function, W represents the weight, b represents the bias, represents the updated node v in the k-th layer in the source domain data, Denote the updated node i in the j-th layer of the target domain data;

[0047] By aggregating the updated nodes in the source domain data and the target domain data respectively, construct the embedding representation of the graph to obtain the source domain graph and the target domain graph respectively;

[0048] ; ;

[0049] wherein, Denote the updated node v in the last layer K of the source domain data, V G Denote the total number of updated nodes in the source domain data, Denote the source domain graph, Denote the updated node i in the last layer J of the target domain data, I H Denote the total number of updated nodes in the target domain data, Denote the target domain graph;

[0050] Introduce a graph morphology difference metric function to calculate the difference value between the source domain graph and the target domain graph;

[0051] ;

[0052] wherein, δ denotes the bandwidth of the graph morphology difference metric function.

[0053] According to one aspect of the above technical solution, calculating the similarity loss between the source domain graph and the target domain graph according to the difference value, based on the cross-entropy loss function, obtaining the true distribution probability of the bearing category according to the source domain data and the target domain data, and through the fully connected layer in the deep residual neural network, obtaining the predicted distribution probability of the bearing category in the process of iteratively updating the nodes in the source domain data and the target domain data, so as to calculate the distribution difference loss between the true distribution probability of the bearing category and the predicted distribution probability of the bearing category, introducing a kernel function, and calculating the distance of the mean embedding in space between the source domain data and the target domain data through the maximum mean discrepancy function to obtain the mean discrepancy loss, specifically including:

[0054] Calculate the similarity loss between the source domain graph and the target domain graph according to the difference value;

[0055] ;

[0056] Customize the fault category items of the bearing;

[0057] Based on the definition of the fault category items, obtain the true distribution probability of the bearing category according to the source domain data and the target domain data;

[0058] Based on the definition of the fault category items, by using the fully connected layer in the deep residual neural network, the bearing category prediction distribution probability is obtained during the iterative update of the nodes in the source domain data and the target domain data;

[0059] Based on the cross-entropy loss function, according to the true distribution probability of the bearing category and the predicted distribution probability of the bearing category, the distribution difference loss between the true distribution probability of the bearing category and the predicted distribution probability of the bearing category is calculated;

[0060] ;

[0061] where, represents the probability that the bearing category in the true distribution is , represents the probability that the bearing category in the predicted distribution is , and F(P,Q) represents the cross-entropy loss function;

[0062] Introduce a kernel function, and calculate the mean difference loss according to the eigenvectors of the vibration signals in the source domain data and the eigenvectors of the vibration signals in the target domain data;

[0063] ;

[0064] where, x c represents the eigenvector of the c-th vibration signal in the source domain data, y β represents the eigenvector of the β-th vibration signal in the target domain data, φ represents the kernel function, g represents the number of vibration signals in the source domain data, and n represents the number of vibration signals in the target domain data.

[0065] According to one aspect of the above technical solution, the total loss function is calculated by combining the similarity loss, the distribution difference loss, and the mean difference loss, and the definition of the total loss function is as follows:

[0066] .

[0067] According to one aspect of the above technical solution, the gradient of the graph morphological difference metric function, the source domain graph, the target domain graph, the nodes, and the parameters of the graph kernel model are calculated in sequence based on the total loss function, specifically including:

[0068] Calculate the gradient of the graph morphological difference metric function based on the total loss function;

[0069] ;

[0070] where, ‌ represents the partial derivative symbol;

[0071] Based on the gradients of the graph morphological difference metric function, and through the chain rule, calculate the gradients of the source domain graph and the target domain graph respectively;

[0072] ;

[0073] ;

[0074] where sign represents the sign function, which is used to extract the direction of the node embedding, and the output of the sign function is ±1 or 0;

[0075] Based on the gradients of the source domain graph and the target domain graph, calculate the gradient of the node;

[0076] ; ;

[0077] Calculate the gradients of the weight parameters and bias parameters of the graph kernel model;

[0078] ; ;

[0079] ; ;

[0080] where represents the node v after the last update of neighbor aggregation, represents the node i after the last update of neighbor aggregation, and R represents the transpose.

[0081] The present invention also provides a bearing fault diagnosis system, including:

[0082] An acquisition module: used to acquire the signal data of the bearing in a known environment, and use the signal data in the known environment as source domain data, acquire the signal data of the bearing in other environments, and use the signal data in other environments as target domain data;

[0083] A denoising module; used to perform denoising processing on the source domain data and the target domain data;

[0084] An extraction module: used to extract features from the denoised source domain data and target domain data through a deep residual neural network combined with a self-attention mechanism to obtain initial feature vectors;

[0085] Specifically, the extraction module is used to introduce a deep residual neural network combined with a self-attention mechanism, where the deep residual neural network includes a multi-scale residual feature extraction module, a channel-temporal dual-path attention mechanism module, a normalization layer, a global average pooling layer, and a global maximum pooling layer;

[0086] Use the multi-scale residual feature extraction module to extract the time-domain and frequency-domain features in the denoised source domain data and target domain data;

[0087] Normalize the denoised source domain data and target domain data through the normalization layer;

[0088] Use the global average pooling layer and the global max pooling layer to reflect continuous fault features and capture transient shock features;

[0089] Calculate the global average value of the input features in the depth residual neural network through the global average pooling layer to obtain an initial feature vector;

[0090] Initialization module: used to regard the initial feature vector as a node, and perform weighted processing on the node in the time channel and frequency channel to initialize the label embedding of the node;

[0091] Specifically, the initialization module is used to: regard the initial feature vector as a node, and calculate the time-channel attention feature weight value of the node based on the channel-temporal dual-path attention mechanism module;

[0092] ;

[0093] ;

[0094] Among them; represents the node \(v\) of the source domain data in the depth residual neural network, represents the node \(i\) of the target domain data in the depth residual neural network, \(W1\) and \(W2\) are the parameters of the fully connected layer in the depth residual neural network, represents the activation function, GAP represents the global average pooling operation, and ReLU represents the non-linear activation function;

[0095] Calculate the frequency-channel attention feature weight value of the node based on the channel-temporal dual-path attention mechanism module;

[0096] ;

[0097] ;

[0098] Among them, \(W3\) is the parameter of the fully connected layer, and GRU is the gated recurrent unit in the depth residual neural network;

[0099] Fuse the time-channel attention feature weight value and the frequency-channel attention feature weight value to obtain a fused feature weight value;

[0100] ;

[0101] ;

[0102] Calculate the internal learning parameter γ in the depth residual neural network;

[0103] ;

[0104] ;

[0105] where GlobalPool(·) is the global average pooling operation, N ( v ) represents the node neighborhood feature; Wr represents the weight matrix, Wr ∈ R (d×2d) , and d represents the dimension of the output feature;

[0106] Initialize the label embedding of the nodes in the source domain data and the target domain data;

[0107] ;

[0108] ;

[0109] Graph calculation module: used to introduce a graph kernel model to iteratively update the nodes with initialized label embeddings in the source domain data and the target domain data respectively, so as to obtain a source domain graph and a target domain graph, and calculate the difference value between the source domain graph and the target domain graph through a graph morphological difference metric function;

[0110] The graph calculation module is specifically used for: taking the nodes with initialized label embeddings as the initial embedding values, and performing neighbor aggregation calculations layer by layer on the initial embedding values in the source domain data and the target domain data respectively through the graph kernel model;

[0111] ; ;

[0112] where k represents the number of neighbor points in the source domain data, N represents the total number of nodes in the source domain data, represents the embedding value corresponding to node u at the (k - 1)th layer in the source domain data, represents the neighbor aggregation feature of node v at the kth layer in the source domain data, j represents the number of neighbor points in the target domain data, C represents the total number of nodes in the target domain data, represents the embedding value corresponding to node f at the (j - 1)th layer in the target domain data, represents the neighbor aggregation feature of node i at the jth layer in the target domain data;

[0113] Based on the results of neighbor aggregation calculation, update the nodes initially embedded in the source domain data and the target domain data respectively;

[0114] ; ;

[0115] Among them, represents the activation function, W represents the weight, and b represents the bias, represents the updated node v in the k-th layer of the source domain data, represents the updated node i in the j-th layer of the target domain data;

[0116] By aggregating the updated nodes in the source domain data and the target domain data respectively, construct the graph embedding representation to obtain the source domain graph and the target domain graph respectively;

[0117] ; ;

[0118] Among them, represents the updated node v in the last layer K of the source domain data, V G represents the total number of updated nodes in the source domain data, represents the source domain graph, represents the updated node i in the last layer J of the target domain data, I H represents the total number of updated nodes in the target domain data, represents the target domain graph;

[0119] Introduce a graph morphology difference metric function to calculate the difference value between the source domain graph and the target domain graph;

[0120] ;

[0121] Among them, δ represents the bandwidth of the graph morphology difference metric function;

[0122] Loss calculation module: used to calculate the similarity loss between the source domain graph and the target domain graph according to the difference value, based on the cross-entropy loss function, obtain the true distribution probability of the bearing category according to the source domain data and the target domain data, and through the fully connected layer in the deep residual neural network, obtain the predicted distribution probability of the bearing category during the iterative update process of the nodes in the source domain data and the target domain data, so as to calculate the distribution difference loss between the true distribution probability of the bearing category and the predicted distribution probability of the bearing category, introduce a kernel function, calculate the distance of the mean embedding in space between the source domain data and the target domain data through the maximum mean discrepancy function, so as to obtain the mean difference loss, and combine the similarity loss, the distribution difference loss and the mean difference loss to calculate the total loss function;

[0123] The loss calculation module is specifically configured to: calculate the similarity loss between the source domain graph and the target domain graph according to the difference value;

[0124] ;

[0125] Customize the fault category items of the bearing;

[0126] Based on the definition of the fault category items, obtain the true distribution probability of the bearing category according to the source domain data and the target domain data;

[0127] Based on the definition of the fault category items, use the fully connected layer in the deep residual neural network to obtain the predicted distribution probability of the bearing category during the iterative update of the nodes in the source domain data and the target domain data;

[0128] Based on the cross-entropy loss function, calculate the distribution difference loss between the true distribution probability of the bearing category and the predicted distribution probability of the bearing category according to the true distribution probability of the bearing category and the predicted distribution probability of the bearing category;

[0129] ;

[0130] wherein, represents the probability that the bearing category in the true distribution is , represents the probability that the bearing category in the predicted distribution is , and F(P,Q) represents the cross-entropy loss function;

[0131] Introduce a kernel function, and calculate the mean difference loss according to the feature vectors of the vibration signals in the source domain data and the feature vectors of the vibration signals in the target domain data;

[0132] ;

[0133] wherein, x c represents the feature vector of the c-th vibration signal in the source domain data, y β represents the feature vector of the β-th vibration signal in the target domain data, φ represents the kernel function, g represents the number of vibration signals in the source domain data, and n represents the number of vibration signals in the target domain data;

[0134] The definition of the total loss function is as follows:

[0135] ;

[0136] Gradient calculation module: used to calculate gradients of the graph morphological difference metric function, the source domain graph, the target domain graph, the nodes, and the parameters of the graph kernel model in sequence based on the total loss function;

[0137] Specifically, the gradient calculation module is used to: calculate the gradient of the graph morphological difference metric function based on the total loss function;

[0138] ;

[0139] where, represents the symbol for partial derivative;

[0140] Based on the gradient of the graph morphological difference metric function and through the chain rule, calculate the gradients of the source domain graph and the target domain graph respectively;

[0141] ;

[0142] ;

[0143] where, sign represents the sign function, which is used to extract the direction of the node embedding, and the output of the sign function is ±1 or 0;

[0144] Based on the gradients of the source domain graph and the target domain graph, calculate the gradient of the nodes;

[0145] ; ;

[0146] Calculate the gradients of the weight parameters and bias parameters of the graph kernel model;

[0147] ; ;

[0148] ; ;

[0149] where, represents the node v after the last update of neighbor aggregation, represents the node i after the last update of neighbor aggregation, and R represents the transpose;

[0150] Optimization module: used to use the cross-entropy loss function to perform supervised learning on the classification results during the iterative update of the nodes in the source domain data and the target domain data, and jointly optimize the graph kernel model in combination with the graph morphological difference metric function and the maximum mean difference function.

[0151] The present invention also provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the bearing fault diagnosis method described above is implemented.

[0152] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the bearing fault diagnosis method described above is implemented.

[0153] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0154] By combining the distribution characteristics of the source domain and target domain graph structures, and integrating traditional domain difference measurement methods with the concept of kernel functions, the differences between graph structures are comprehensively captured. During the calculation process, the global distribution difference and local structure difference are dynamically balanced through an adaptive weight adjustment mechanism, and at the same time, the feature mapping in the high-dimensional Hilbert space is optimized to minimize the distribution difference between the source domain and the target domain, thereby improving the adaptability and diagnostic accuracy of the model in the target domain.

[0155] The present invention constructs a topological representation of the time-domain signal graph of the axle box bearing vibration signal, deeply excavates the inherent morphological features in the process of equipment state evolution, and establishes a hierarchical domain-invariant feature extraction mechanism based on neural networks. This method breaks through the limitations of traditional statistical distribution matching, realizes cross-domain knowledge transfer from the essence structure of data, and provides a new theoretical tool and technical path for equipment state migration diagnosis in complex industrial environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0156] Figure 1 is a flowchart of the bearing fault diagnosis method in the first embodiment of the present invention;

[0157] Figure 2 is a structural block diagram of the bearing fault diagnosis system in the second embodiment of the present invention;

[0158] Figure 3 is a structural block diagram of the electronic device in the third embodiment of the present invention;

[0159] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0160] To facilitate the understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the disclosure of the present invention more thorough and comprehensive.

[0161] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly on the other element or there can also be an intermediate element. When an element is considered to be "connected to" another element, it can be directly connected to the other element or there may be an intermediate element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0162] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0163] Please refer to Figure 1 , which shows a bearing fault diagnosis method in the first embodiment of the present invention, including the following steps:

[0164] S10, Obtain the signal data of the bearing under a known environment, and use the signal data under the known environment as the source domain data, obtain the signal data of the bearing under other environments, and use the signal data under other environments as the target domain data;

[0165] S20, Denoise the source domain data and the target domain data;

[0166] S30, Extract features from the denoised source domain data and target domain data through a deep residual neural network combined with a self-attention mechanism to obtain an initial feature vector;

[0167] S40, Regard the initial feature vector as a node, and perform weighted processing on the node in the time channel and the frequency channel to initialize the label embedding of the node;

[0168] S50, Introduce a graph kernel model to iteratively update the nodes after label embedding initialization in the source domain data and the target domain data respectively to obtain a source domain graph and a target domain graph, and calculate the difference value between the source domain graph and the target domain graph through a graph morphological difference metric function;

[0169] S60. Calculate the similarity loss between the source domain graph and the target domain graph according to the difference value. Based on the cross-entropy loss function, obtain the true distribution probability of the bearing category according to the source domain data and the target domain data, and through the fully connected layer in the deep residual neural network, obtain the predicted distribution probability of the bearing category during the iterative update of the nodes in the source domain data and the target domain data, so as to calculate the distribution difference loss between the true distribution probability of the bearing category and the predicted distribution probability of the bearing category. Introduce a kernel function, and calculate the distance of the mean embeddings of the source domain data and the target domain data in the space through the maximum mean discrepancy function to obtain the mean discrepancy loss, and combine the similarity loss, the distribution difference loss and the mean discrepancy loss to calculate the total loss function;

[0170] S70. Based on the total loss function, calculate the gradients of the graph morphology difference metric function, the source domain graph and the target domain graph, the nodes, and the parameters of the graph kernel model in sequence;

[0171] S80. Use the cross-entropy loss function to perform supervised learning on the classification results during the iterative update of the nodes in the source domain data and the target domain data, and combine the graph morphology difference metric function and the maximum mean discrepancy function to jointly optimize the graph kernel model.

[0172] It can be understood that the present invention comprehensively captures the differences between graph structures by combining traditional domain difference measurement methods and kernel function concepts according to the distribution characteristics of the source domain and target domain graph structures. During the calculation process, the global distribution difference and the local structure difference are dynamically balanced through an adaptive weight adjustment mechanism, and at the same time, the feature mapping in the high-dimensional Hilbert space is optimized to minimize the distribution difference between the source domain and the target domain, thereby improving the adaptation ability and diagnostic accuracy of the model in the target domain;

[0173] The present invention constructs a topological representation of the time-domain signal graph of the axle box bearing vibration signal, deeply excavates the inherent morphological features in the equipment state evolution process, and establishes a hierarchical domain-invariant feature extraction mechanism based on a neural network. This method breaks through the limitations of traditional statistical distribution matching, realizes cross-domain knowledge transfer from the essence structure of data, and provides a new theoretical tool and technical path for equipment state migration diagnosis in complex industrial environments.

[0174] Further, the specific steps of step S20 include:

[0175] Divide the source domain data into a source domain training set and a source domain test set, and divide the target domain data into a target domain training set and a target domain test set; it should be noted that both the source domain data and the target domain data are vibration signal data;

[0176] Divide the source domain training set, the source domain test set, the target domain training set, and the target domain test set into data samples according to a preset data length, and perform standardization processing on the data samples; this step is to enhance the robustness of the image recognition model;

[0177] Use the abnormal amplitude replacement method to correct the outliers in the standardized data samples.

[0178] It can be understood that the main purpose of this step is to reduce the interference of noise on the signal data, thereby improving the ability to diagnose bearings in the later stage.

[0179] Further, the step S30 specifically includes:

[0180] Introduce a deep residual neural network combined with a self-attention mechanism, where the deep residual neural network includes a multi-scale residual feature extraction module, a channel-temporal dual-path attention mechanism module, a normalization layer, a global average pooling layer, and a global maximum pooling layer;

[0181] Use the multi-scale residual feature extraction module to extract the time-domain and frequency-domain features in the denoised source domain data and target domain data;

[0182] Normalize the denoised source domain data and target domain data through the normalization layer; this step can make the data distribution more standard, improve the convergence speed of the model and the generalization ability of the model;

[0183] Use the global average pooling layer and the global maximum pooling layer to reflect continuous fault features and capture transient shock features; this is achieved by the dual-path operation of the global average pooling layer and the global maximum pooling layer;

[0184] Calculate the global average value of the input features in the deep residual neural network through the global average pooling layer for each channel to obtain an initial feature vector; this initial feature vector is a one-dimensional feature vector.

[0185] Further, the step S40 specifically includes:

[0186] Regard the initial feature vector as a node, and calculate the time-channel attention feature weight value of the node based on the channel-temporal dual-path attention mechanism module;

[0187] ;

[0188] ;

[0189] where; represents the node \(v\) of the source domain data in the deep residual neural network, Node \(i\) representing the target domain data in the deep residual neural network, \(W1\) and \(W2\) are the fully connected layer parameters in the deep residual neural network, \(\sigma\) represents the activation function, GAP represents the global average pooling operation, and ReLU represents the non-linear activation function;

[0190] Based on the channel-temporal dual-path attention mechanism module, calculate the frequency channel attention feature weight value of the node;

[0191] ;

[0192] ;

[0193] where \(W3\) is the fully connected layer parameter, and GRU is the gated recurrent unit in the deep residual neural network;

[0194] Fuse the time channel attention feature weight value and the frequency channel attention feature weight value to obtain the fused feature weight value;

[0195] ;

[0196] ;

[0197] Calculate the internal learning parameter \(\gamma\) in the deep residual neural network;

[0198] ;

[0199] ;

[0200] where GlobalPool(·) is the global average pooling operation, N ( v ) represents the node neighborhood feature; \(W_r\) represents the weight matrix, \(W_r\in R\) (d×2d) , and \(d\) represents the dimension of the output feature;

[0201] Perform label embedding initialization on the nodes in the source domain data and the target domain data;

[0202] ;

[0203] .

[0204] It can be understood that the initialization of node label embedding is a technique that maps nodes in a network to a low-dimensional vector space. It can preserve the similarity of nodes in the network, enabling the node relationships in the embedding space to approximately reflect the structure and properties of the original network. The basic idea of node embedding is to represent each node as a low-dimensional vector, and these vectors maintain a certain similarity in the network; graph representation is a data structure that constructs the relationship between nodes and edges. The graph structure can achieve fast retrieval of specific information in the graph, facilitating the extraction and analysis of graph morphological features. In traditional graph neural networks, node initialization usually directly uses the original features or generates initial embeddings through simple linear transformations. Attention-weighted initialization dynamically allocates feature weights and fuses cross-node information through the attention mechanism, significantly enhancing the representation ability of the initial features.

[0205] Furthermore, the specific steps of step S50 include:

[0206] Taking the nodes after label embedding initialization as the initial embedding values, and respectively performing neighbor aggregation calculations layer by layer on the initial embedding values in the source domain data and the target domain data through the graph kernel model;

[0207] ; ;

[0208] where k represents the number of neighbor nodes in the source domain data, N represents the total number of nodes in the source domain data, represents the embedding value corresponding to node u in the (k - 1)-th layer in the source domain data, represents the neighbor aggregation feature of node v in the k-th layer in the source domain data, j represents the number of neighbor nodes in the target domain data, C represents the total number of nodes in the target domain data, represents the embedding value corresponding to node f in the (j - 1)-th layer in the target domain data, represents the neighbor aggregation feature of node i in the j-th layer in the target domain data;

[0209] Based on the results of neighbor aggregation calculations, respectively perform node updates on the nodes with initial embeddings in the source domain data and the target domain data;

[0210] ; ;

[0211] where, represents the activation function, W represents the weight, b represents the bias, represents the updated node v in the k-th layer in the source domain data, represents the updated node i in the j-th layer in the target domain data;

[0212] By aggregating the updated nodes in the source domain data and the target domain data respectively, an embedded representation of the graph is constructed to obtain the source domain graph and the target domain graph respectively;

[0213] ; ;

[0214] Among them, represents the updated node v in the last layer K of the source domain data, and V G represents the total number of updated nodes in the source domain data, represents the source domain graph, represents the updated node i in the last layer J of the target domain data, and I H represents the total number of updated nodes in the target domain data, represents the target domain graph;

[0215] Introduce a graph shape difference metric function to calculate the difference value between the source domain graph and the target domain graph;

[0216] ;

[0217] Among them, δ represents the bandwidth of the graph shape difference metric function.

[0218] It can be understood that by calculating the difference value between the embeddings of the source domain graph and the target domain graph to measure the similarity between the two, the parameterized graph kernel function can better characterize the difference in graph shape features between the source domain and the target domain, ensuring that the domain adaptation process is more accurate;

[0219] This step is carried out in two ways. One way starts from the source domain data and gradually calculates the source domain graph, and the other way starts from the target domain data and gradually calculates the target domain graph.

[0220] Furthermore, in step S60, according to the difference value, calculate the similarity loss between the source domain graph and the target domain graph. Based on the cross-entropy loss function, obtain the true distribution probability of the bearing category according to the source domain data and the target domain data, and through the fully connected layer in the deep residual neural network, obtain the predicted distribution probability of the bearing category during the iterative update process of the nodes in the source domain data and the target domain data, so as to calculate the distribution difference loss between the true distribution probability of the bearing category and the predicted distribution probability of the bearing category. Introduce a kernel function, and calculate the distance between the mean embeddings of the source domain data and the target domain data in space through the maximum mean discrepancy function to obtain the mean discrepancy loss, specifically including:

[0221] Calculate the similarity loss between the source domain graph and the target domain graph according to the difference value;

[0222] ;

[0223] Customize the fault category items of the bearing;

[0224] Based on the definition of the fault category items, obtain the true distribution probability of the bearing category according to the source domain data and the target domain data;

[0225] Based on the definition of the fault category items, use the fully connected layer in the deep residual neural network to obtain the predicted distribution probability of the bearing category during the iterative update of the nodes in the source domain data and the target domain data;

[0226] Based on the cross-entropy loss function, calculate the distribution difference loss between the true distribution probability of the bearing category and the predicted distribution probability of the bearing category according to the true distribution probability of the bearing category and the predicted distribution probability of the bearing category;

[0227] ;

[0228] Wherein, represents the probability that the bearing category in the true distribution is , represents the probability that the bearing category in the predicted distribution is , and F(P,Q) represents the cross-entropy loss function;

[0229] Introduce a kernel function, and calculate the mean difference loss according to the feature vectors of the vibration signals in the source domain data and the feature vectors of the vibration signals in the target domain data;

[0230] ;

[0231] Wherein, x c represents the feature vector of the c-th vibration signal in the source domain data, y β represents the feature vector of the β-th vibration signal in the target domain data, φ represents the kernel function, g represents the number of vibration signals in the source domain data, and n represents the number of vibration signals in the target domain data.

[0232] It can be understood that after calculating the source domain graph and the target domain graph, and calculating the difference value, the inter-domain difference can be minimized through the kernel loss term, thereby improving the generalization performance of the model on the target domain. The graph shape difference metric function can better characterize the graph shape feature differences between the source domain and the target domain, ensuring that the domain adaptation process is more accurate; then customize the fault category items of the bearing. The true distribution probability of the bearing category can be obtained from the source domain data. Use a fully connected network to classify the graph embedding to obtain the fault category of the bearing. The fully connected network contains three layers, and each layer uses the ReLU activation function to enhance the non-linear expression ability. Finally, connect the Softmax activation function to output the classification result as a probability distribution. This step can obtain the predicted distribution probability of the bearing category during the iterative update process of the nodes in the source domain data and the target domain data, and then the cross-entropy loss function can be constructed. This function can measure the distribution difference loss between the predicted distribution probability of the bearing category by the model and the true distribution probability of the bearing category; finally, introduce the kernel function, and calculate the mean difference loss according to the feature vectors of the vibration signals in the source domain data and the feature vectors of the vibration signals in the target domain data. This step maps the data to a high-dimensional space through the kernel function and calculates the distance between the mean embeddings of the source domain data and the target domain data in the space.

[0233] Further, in step S60, the definition of the total loss function is as follows:

[0234] 。

[0235] It can be understood that by combining the similarity loss, the distribution difference loss, and the mean difference loss, the total loss function is obtained. Through the total loss function, the graph kernel function can be optimized to achieve the graph shape difference between the source domain and the target domain, and this difference is minimized.

[0236] Further, step S70 specifically includes:

[0237] Calculate the gradient of the graph shape difference metric function based on the total loss function;

[0238] ;

[0239] Among them, represents the symbol for partial derivative;

[0240] Based on the gradient of the graph shape difference metric function and through the chain rule, calculate the gradients of the source domain graph and the target domain graph respectively;

[0241] ;

[0242] ;

[0243] Among them, sign represents the sign function, which is used to extract the direction of the node embedding, and the output of the sign function is ±1 or 0;

[0244] Calculate the gradient of the node based on the gradients of the source domain graph and the target domain graph;

[0245] ; ;

[0246] Calculate the gradients of the weight parameters and bias parameters of the graph kernel model;

[0247] ; ;

[0248] ; ;

[0249] Among them, represents the node v after the last update of neighbor aggregation, represents the node i after the last update of neighbor aggregation, and R represents the transpose.

[0250] It can be understood that after minimizing the graph difference through the total loss function, the gradients of network parameters (such as graph kernel weights, biases, kernel function bandwidths, etc.) can be calculated, and the parameters can be updated to finally achieve cross-device diagnosis of bearing faults.

[0251] The present invention can use the Python programming language and mainly relies on the PyTorch deep learning framework for model construction and training. The implementation of the CNN model includes five convolutional layers, each followed by a batch normalization layer and a ReLU activation layer. Finally, a global average pooling layer is used to generate features. The graph embedding is mapped through a custom graph embedding layer, and the node embedding update of the parameterized WL graph kernel is realized. Domain adaptation is trained through a defined adaptive loss function, combined with the kernel similarity between the source domain and the target domain, using the backpropagation process.

[0252] In summary, for the bearing fault diagnosis method in the above embodiments of the present invention, 1. By combining a deep convolutional neural network and a parameterized graph kernel model, local and global features in the bearing vibration signal can be effectively extracted, significantly improving the accuracy of fault diagnosis;

[0253] 2. By introducing a domain adaptation graph kernel loss function and a cross-entropy loss function, the distribution difference between the source domain and the target domain is reduced, making the migration effect of the model better on different data sets, improving the generalization ability of the model. The parameterized graph kernel function, by introducing trainable weights, biases, and graph kernel function bandwidths, not only enhances the mathematical interpretability of the model but also can effectively quantify the graph morphology difference;

[0254] 3. High degree of automation, the diagnostic process does not rely on manual feature engineering, reduces the dependence on expert knowledge, and improves the diagnostic efficiency and accuracy.

[0255] Please refer to Figure 2 , the second embodiment of the present invention provides a bearing fault diagnosis system, including:

[0256] Acquisition module 11: used to acquire the signal data of the bearing in a known environment, and regard the signal data in the known environment as source domain data, acquire the signal data of the bearing in other environments, and regard the signal data in other environments as target domain data;

[0257] Denosing module 12; used to perform denoising processing on the source domain data and the target domain data;

[0258] Extraction module 13: used to perform feature extraction on the denoised source domain data and target domain data through a deep residual neural network combined with a self-attention mechanism to obtain an initial feature vector;

[0259] Initialization module 14: used to regard the initial feature vector as a node, and perform weighted processing on the node in the time channel and frequency channel to initialize the label embedding of the node;

[0260] Graph calculation module 15: used to introduce a graph kernel model to iteratively update the nodes after label embedding initialization in the source domain data and the target domain data respectively to obtain a source domain graph and a target domain graph, and calculate the difference value between the source domain graph and the target domain graph through a graph morphology difference metric function;

[0261] Loss calculation module 16: used to calculate the similarity loss between the source domain graph and the target domain graph according to the difference value, based on the cross-entropy loss function, obtain the true distribution probability of the bearing category according to the source domain data and the target domain data, and through the fully connected layer in the deep residual neural network, obtain the predicted distribution probability of the bearing category during the iterative update process of the nodes in the source domain data and the target domain data, so as to calculate the distribution difference loss between the true distribution probability of the bearing category and the predicted distribution probability of the bearing category, introduce a kernel function, calculate the distance of the mean embedding in space between the source domain data and the target domain data through the maximum mean discrepancy function to obtain the mean difference loss, and combine the similarity loss, the distribution difference loss and the mean difference loss to calculate the total loss function;

[0262] Gradient calculation module 17: used to perform gradient calculation on the graph morphology difference metric function, the source domain graph and the target domain graph, the node, and the parameters of the graph kernel model based on the total loss function in turn;

[0263] Optimization module 18: It is used to perform supervised learning on the classification results during the iterative update of the nodes in the source domain data and the target domain data by using the cross-entropy loss function, and jointly optimize the graph kernel model in combination with the graph morphological difference metric function and the maximum mean difference function.

[0264] The present invention also provides an electronic device. Please refer to Figure 3 , which shows the electronic device in the third embodiment of the present invention, including a memory 10, a processor 20, and a computer program 30 stored on the memory 10 and executable on the processor 20. When the processor 20 executes the computer program 30, the above-mentioned bearing fault diagnosis method is implemented.

[0265] Among them, the memory 10 includes at least one type of storage medium, and the storage medium includes flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. The memory 10 can be an internal storage unit of the electronic device in some embodiments, such as the hard disk of the electronic device. The memory 10 can also be an external storage device in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Further, the memory 10 can also include both the internal storage unit of the electronic device and the external storage device. The memory 10 can be used not only to store application software and various types of data installed in the electronic device, but also to temporarily store data that has been output or will be output.

[0266] Among them, the processor 20 can be an Electronic Control Unit (ECU, also known as a vehicle computer), a Central Processing Unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments, and is used to run the program code stored in the memory 10 or process data, such as executing an access restriction program.

[0267] It should be noted that Figure 3 the structure shown does not constitute a limitation on the electronic device. In other embodiments, the electronic device may include fewer or more components than shown, or combine certain components, or have a different component layout.

[0268] The embodiment of the present invention also provides a readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned bearing fault diagnosis method is implemented.

[0269] Those skilled in the art can understand that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or used in combination with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

[0270] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing as appropriate, and then stored in a computer memory.

[0271] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well-known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0272] In the description of this specification, the description referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0273] The above-described embodiments merely represent several implementation manners of the present invention. The description thereof is relatively specific and detailed, but it should not be construed as a limitation to the scope of the patent for the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all fall within the protection scope of the present invention. Therefore, the protection scope of the patent for the present invention shall be subject to the appended claims.

Claims

1. A bearing fault diagnosis method, characterized in that, The steps are as follows: Obtain the signal data of the bearing in a known environment, and use the signal data in the known environment as the source domain data. Obtain the signal data of the bearing in other environments, and use the signal data in other environments as the target domain data; Denoise the source domain data and the target domain data; Extract features from the denoised source domain data and target domain data through a deep residual neural network combined with a self-attention mechanism to obtain initial feature vectors; Regard the initial feature vectors as nodes, and perform weighted processing on the nodes in the time channel and frequency channel to initialize the label embedding of the nodes; Introduce a graph kernel model to iteratively update the nodes with initialized label embedding in the source domain data and the target domain data respectively to obtain a source domain graph and a target domain graph, and calculate the difference value between the source domain graph and the target domain graph through a graph morphology difference metric function; Calculate the similarity loss between the source domain graph and the target domain graph according to the difference value. Based on the cross-entropy loss function, obtain the true distribution probability of the bearing category according to the source domain data and the target domain data, and obtain the predicted distribution probability of the bearing category through the fully connected layer in the deep residual neural network during the iterative update of the nodes in the source domain data and the target domain data, so as to calculate the distribution difference loss between the true distribution probability of the bearing category and the predicted distribution probability of the bearing category. Introduce a kernel function, calculate the distance of the mean embedding in space between the source domain data and the target domain data through the maximum mean discrepancy function to obtain the mean discrepancy loss, and combine the similarity loss, the distribution difference loss and the mean discrepancy loss to calculate the total loss function; Based on the total loss function, calculate the gradients of the graph morphology difference metric function, the source domain graph and the target domain graph, the nodes, and the parameters of the graph kernel model in turn; Use the cross-entropy loss function to perform supervised learning on the classification results during the iterative update of the nodes in the source domain data and the target domain data, and combine the graph morphology difference metric function and the maximum mean discrepancy function to jointly optimize the graph kernel model.

2. The bearing fault diagnosis method according to claim 1, wherein The step of extracting features from the denoised source domain data and target domain data through a deep residual neural network combined with a self-attention mechanism to obtain initial feature vectors specifically includes: Introduce a deep residual neural network combined with a self-attention mechanism, where the deep residual neural network includes a multi-scale residual feature extraction module, a channel-temporal dual-path attention mechanism module, a normalization layer, a global average pooling layer, and a global maximum pooling layer; Use the multi-scale residual feature extraction module to extract the time domain and frequency domain features in the denoised source domain data and target domain data; Normalize the denoised source domain data and target domain data through the normalization layer; Use the global average pooling layer and the global maximum pooling layer to reflect continuous fault features and capture transient impact features; Performing global average calculation on the inner channels of the input features in the deep residual neural network through the global average pooling layer to obtain an initial feature vector.

3. The bearing fault diagnosis method according to claim 2, wherein Regarding the initial feature vector as a node, and performing weighted processing on the node in the time channel and frequency channel to initialize the label embedding of the node, specifically including: Regarding the initial feature vector as a node, and calculating the time-channel attention feature weight value of the node based on the channel-temporal dual-path attention mechanism module; ； ； Among them; denotes the node v of the source domain data in the deep residual neural network, denotes the node i of the target domain data in the deep residual neural network, and W1 and W2 are the full connection layer parameters in the deep residual neural network, denotes the activation function, GAP denotes the global average pooling operation, and ReLU denotes the non-linear activation function; Calculating the frequency-channel attention feature weight value of the node based on the channel-temporal dual-path attention mechanism module; ； ； where W3 is the parameter of the fully connected layer, and GRU is the gated recurrent unit in the deep residual neural network; Fusing the time-channel attention feature weight value and the frequency-channel attention feature weight value to obtain a fused feature weight value; ； ； Calculating the internal learning parameter γ in the deep residual neural network; ； ； Among them, GlobalPool(·) is the global average pooling operation, N ( v ) represents the node neighborhood feature; Wr represents the weight matrix, Wr ∈ R (d×2d) , d represents the dimension of the output feature; Initializing the label embedding of the nodes in the source domain data and the target domain data; ； 。 4. The bearing fault diagnosis method according to claim 3, characterized in that Introducing a graph kernel model to iteratively update the nodes with initialized label embeddings in the source domain data and the target domain data respectively to obtain a source domain graph and a target domain graph, and calculating the difference value between the source domain graph and the target domain graph through a graph morphology difference metric function, specifically including: Regarding the nodes with initialized label embeddings as initial embedding values, and performing neighbor aggregation calculation layer by layer on the initial embedding values in the source domain data and the target domain data respectively through the graph kernel model; ； ； where k represents the number of neighbor points in the source domain data, and N represents the total number of nodes in the source domain data. represents the embedding value corresponding to node u at the (k - 1)-th layer in the source domain data. represents the neighbor aggregation feature of node v at the k-th layer in the source domain data. j represents the number of neighbor points in the target domain data, and C represents the total number of nodes in the target domain data. represents the embedding value corresponding to node f at the (j - 1)-th layer in the target domain data. represents the neighbor aggregation feature of node i at the j-th layer in the target domain data. Based on the results of neighbor aggregation calculation, updating the nodes with initially embedded labels in the source domain data and the target domain data respectively; ； ； Among them, represents the activation function, W represents the weight, and b represents the bias, represents the updated node v in the k-th layer of the source domain data, represents the updated node i in the j-th layer of the target domain data; Constructing the embedding representation of the graph by aggregating the updated nodes in the source domain data and the target domain data respectively to obtain a source domain graph and a target domain graph respectively; ； ； Among them, represents the updated node v in the last layer K of the source domain data, V G represents the total number of updated nodes in the source domain data, represents the source domain graph, represents the updated node i in the last layer J of the target domain data, I H represents the total number of updated nodes in the target domain data, represents the target domain graph; Introducing a graph morphology difference metric function to calculate the difference value between the source domain graph and the target domain graph; ； where δ represents the bandwidth of the graph morphology difference metric function.

5. The bearing fault diagnosis method according to claim 4, wherein Calculating the similarity loss between the source domain graph and the target domain graph according to the difference value, based on the cross-entropy loss function, obtaining the true distribution probability of the bearing category according to the source domain data and the target domain data, and obtaining the predicted distribution probability of the bearing category through the fully connected layer in the deep residual neural network during the process of iteratively updating the nodes in the source domain data and the target domain data, so as to calculate the distribution difference loss between the true distribution probability of the bearing category and the predicted distribution probability of the bearing category, introducing a kernel function, and calculating the distance of the mean embedding in space between the source domain data and the target domain data through the maximum mean discrepancy function to obtain the mean discrepancy loss, specifically including: Calculating the similarity loss between the source domain graph and the target domain graph according to the difference value; ； Defining the fault category items of the bearing; Based on the definition of the fault category items, obtaining the true distribution probability of the bearing category according to the source domain data and the target domain data; Based on the definition of the fault category items, by using the fully connected layer in the deep residual neural network, the bearing category prediction distribution probability is obtained during the iterative update of the nodes in the source domain data and the target domain data; Based on the cross-entropy loss function, according to the true distribution probability of the bearing category and the predicted distribution probability of the bearing category, the distribution difference loss between the true distribution probability of the bearing category and the predicted distribution probability of the bearing category is calculated; ； Among them, represents the probability that the bearing category in the true distribution is , represents the probability that the bearing category in the predicted distribution is , and F(P, Q) represents the cross-entropy loss function; A kernel function is introduced, and the mean difference loss is calculated according to the feature vectors of the vibration signals in the source domain data and the feature vectors of the vibration signals in the target domain data; ； Among them, x c represents the feature vector of the c-th vibration signal in the source domain data, and y β represents the feature vector of the β-th vibration signal in the target domain data. φ represents the kernel function, g represents the number of vibration signals in the source domain data, and n represents the number of vibration signals in the target domain data.

6. The bearing fault diagnosis method according to claim 5, characterized in that Combining the similarity loss, the distribution difference loss and the mean difference loss to calculate the total loss function, and the definition of the total loss function is as follows: 。 7. The bearing fault diagnosis method according to claim 6, characterized in that Based on the total loss function, gradient calculations are sequentially performed on the graph morphology difference metric function, the source domain graph and the target domain graph, the nodes, and the parameters of the graph kernel model, specifically including: Calculating the gradient of the graph morphology difference metric function based on the total loss function; ； Among them, represents the symbol for partial derivative; Based on the gradient of the graph morphology difference metric function and through the chain rule, the gradients of the source domain graph and the target domain graph are calculated respectively; ； ； Among them, sign represents the sign function, which is used to extract the direction of the node embedding, and the output of the sign function is ±1 or 0; Calculating the gradient of the nodes based on the gradients of the source domain graph and the target domain graph; ； ； Performing gradient calculations on the weight parameters and bias parameters of the graph kernel model; ； ； ； ； Among them, represents node v after the last update of neighbor aggregation, represents node i after the last update of neighbor aggregation, and R represents the transpose.

8. A bearing fault diagnosis system, characterized in that, Including: An acquisition module: used to acquire the signal data of the bearing in the known environment, and take the signal data in the known environment as the source domain data, acquire the signal data of the bearing in other environments, and take the signal data in other environments as the target domain data; A denoising module; used to perform denoising processing on the source domain data and the target domain data; An extraction module: used to perform feature extraction on the denoised source domain data and target domain data through a deep residual neural network combined with a self-attention mechanism to obtain initial feature vectors; An initialization module: used to regard the initial feature vectors as nodes and perform weighted processing on the nodes in the time channel and the frequency channel to initialize the label embedding of the nodes; A graph calculation module: used to introduce a graph kernel model to perform iterative updates on the nodes after label embedding initialization in the source domain data and the target domain data respectively to obtain a source domain graph and a target domain graph, and calculate the difference value between the source domain graph and the target domain graph through a graph morphology difference metric function; Loss calculation module: used to calculate the similarity loss between the source domain graph and the target domain graph according to the difference value, obtain the true distribution probability of bearing categories based on the cross-entropy loss function according to the source domain data and the target domain data, and through the fully connected layer in the deep residual neural network, obtain the predicted distribution probability of bearing categories in the process of iteratively updating the nodes in the source domain data and the target domain data, so as to calculate the distribution difference loss between the true distribution probability of bearing categories and the predicted distribution probability of bearing categories, introduce a kernel function, calculate the distance of the mean embeddings of the source domain data and the target domain data in space through the maximum mean discrepancy function, so as to obtain the mean discrepancy loss, and calculate the total loss function by combining the similarity loss, the distribution difference loss and the mean discrepancy loss; Gradient calculation module: used to calculate the gradients of the graph morphology difference metric function, the source domain graph and the target domain graph, the nodes, and the parameters of the graph kernel model based on the total loss function in sequence; Optimization module: used to use the cross-entropy loss function to perform supervised learning on the classification results in the process of iteratively updating the nodes in the source domain data and the target domain data, and jointly optimize the graph kernel model in combination with the graph morphology difference metric function and the maximum mean discrepancy function.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the bearing fault diagnosis method according to any one of claims 1-7.

10. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the bearing fault diagnosis method according to any one of claims 1-7.

Images