Train bearing fault diagnosis method based on graph multi-scale complex entropy

Abstract

The invention belongs to the technical field of train fault detection, and discloses a train bearing fault diagnosis method based on graph multi-scale complex entropy, and the method comprises the steps: converting a collected original vibration signal of a to-be-detected piece from a time domain to a graph domain; coarsening the signals of the image domain to obtain a multi-scale coarse image; calculating a graph multi-scale complex entropy of the to-be-tested piece according to the multi-scale graph, and taking the graph multi-scale complex entropy as a feature vector of the to-be-tested piece; and inputting the features screened out from the feature vectors of the to-be-tested piece into a train bearing fault classifier to obtain a train bearing fault classification result, and completing train bearing fault diagnosis. The train bearing fault diagnosis method can effectively reduce the interference of background noise and impurity frequency at the same time, has strong fault feature extraction capability and high robustness, can provide higher accuracy for the fault diagnosis result of the bearing, realizes accurate diagnosis of different fault types and different fault degrees of the train bearing, and improves the fault diagnosis accuracy of the train bearing. The train bearing fault can be effectively monitored, and the safety and economical efficiency of train operation are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of train fault detection, relates to a train bearing fault detection technology, and in particular to a train bearing fault diagnosis method based on graph multi-scale complex entropy. Background Art

[0002] As a key component in rotating machinery, bearings play a vital role in transmitting speed and load, and their health is directly related to the safe operation of the entire train system. In rail transit equipment such as high-speed trains, bearings work for a long time in an environment of high speed, high load, and large temperature changes, and are prone to failures such as fatigue spalling, pitting, and wear. When a bearing fails, repeated collisions between the damaged point and other components will produce impact vibrations, which may cause train operation failures, economic losses, and even casualties in severe cases. Therefore, effective fault diagnosis and monitoring of train bearings is a key link in ensuring railway transportation safety.

[0003] Traditional train bearing fault diagnosis methods mainly rely on the analysis of vibration signals. Vibration signals carry key information about the bearing system during operation, including the working status, load conditions, friction and wear of the bearings, etc. Through in-depth analysis of these vibration signals, abnormal conditions of train bearings can be effectively identified, and predictive maintenance measures can be implemented to reduce unnecessary downtime and maintenance costs, thereby improving the reliability and safety of train operation.

[0004] However, in practical applications, the vibration signals of bearings are often accompanied by vibration signal interference from other mechanical components and strong noise background. These signals are often the comprehensive manifestation of the vibration response of multiple components, making it difficult to extract weak fault features, increasing the difficulty of diagnosis and the possibility of misjudgment. In addition, the complexity of the sensor transmission path causes signal energy attenuation or distortion, making it more difficult to distinguish between vibration signals containing bearing fault feature information and healthy vibration signals, further exacerbating the difficulty of bearing fault diagnosis. Summary of the invention

[0005] In view of this, the purpose of the present invention is to address the problems of low accuracy in vibration signal feature extraction, large background noise interference of vibration signals, and severe signal attenuation in the prior art, and to provide a train bearing fault diagnosis method based on graph multi-scale complex entropy. By performing graph signal processing on the fault signal, the health status indicator of the bearing is obtained, thereby achieving effective diagnosis of train bearing faults.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions to achieve it.

[0007] The train bearing fault diagnosis method based on graph multi-scale complex entropy provided by the present invention comprises the following steps: S1 converts the collected original vibration signal of the test piece from the time domain to the image domain; S2 performs coarsening processing on the signal in the image domain to obtain a multi-scale coarse image, which specifically includes the following sub-steps: S21 clusters the graph domain signal to obtain a graph segmentation matrix; S22 calculates the adjacency matrix and graph signal of the coarsened graph according to the graph partition matrix; Select multiple scale factors, repeat the above steps S21-S22, and coarsen the image domain signal into multiple coarse images of different scales, i.e., multi-scale coarse images; S3 calculates the multi-scale complex entropy of the DUT graph based on the multi-scale graph obtained in S2, and uses it as the feature vector of the DUT to define the graph complex entropy ,The specific steps of calculating the complex entropy of the graph are as follows: S31 reconstructs the phase space of the image signal under each scale factor and constructs m dimensional vector sequence; S32 Calculation m The Manhattan distance between the dimensional vector sequences is used as the similarity, and the statistical similarity is less than the threshold r The number of similarity probabilities is calculated to obtain the logarithmic mean value. P m ( r ), and similarly calculate the embedding dimension as m The logarithmic mean of the similarity probability under +1 P m+1 ( r ), P m ( r )and P m+1 ( r ) is E 1; S33 According to step S32 P m ( r )and P m+1 ( r ),calculate for E 2; S34 is obtained according to step S32 m The Manhattan distance between the dimensional vector sequences is estimated using the histogram method to estimate the empirical probability density function of the distance and calculated E 3. , P b represents the probability of falling into the b interval of the histogram, and B represents the number of intervals of the histogram; S35 calculates adjacentm The Pearson correlation coefficient between dimensional vector sequences is in the range of [-1, 1]. [-1, 1] is evenly divided into A subintervals, divide all correlation coefficients into subintervals, count the probability of falling into each subinterval, and calculate E 4. , P a Indicates that it falls into the subinterval a The probability of S36 For each scale factor, calculate , that is, the multi-scale complex entropy of the graph is obtained and used as the feature vector of the device under test; S4 inputs the features screened out from the feature vector of the test piece into the train bearing fault classifier to obtain the train bearing fault classification result and complete the train bearing fault diagnosis.

[0008] In the above step S1, the specific implementation method is to map the collected original time series signal into a two-dimensional matrix to obtain an undirected weighted graph. The mapping method can be a visible graph, a horizontal visible graph, a k-nearest neighbor graph, etc.; through this method, the information in the one-dimensional time series is stored in a graph matrix form containing a set of bidirectional relationships between objects and the strength of these relationships, so as to realize the analysis of time series signals in the form of a graph. Then, based on the graph spectrum theory, the graph matrix representation is obtained, including the adjacency matrix, the degree diagonal matrix and the Laplace matrix, and the eigenvalues ​​and eigenvectors of the normalized Laplace matrix are obtained to realize visual multidimensional analysis.

[0009] In the above step S2, a coarsening algorithm is proposed. , the original image signal is X , W is the adjacency matrix, U is the eigenvector matrix obtained by orthogonal decomposition of the graph Laplacian matrix, s is the scale factor, n is the number of nodes in the coarsened graph, and .

[0010] In the above step S21, each column of the eigenvector matrix is ​​standardized to [-1, 1], and the eigenvectors with absolute values ​​less than 0.001 are regarded as 0. The non-zero rate of each eigenvector is calculated to obtain the standard deviation of the non-zero rate, and the eigenvectors with non-zero rates higher than 3 times the standard deviation are selected. The K-means clustering algorithm is applied to the rows of the eigenvectors, and the parameters in the algorithm are k = n , N The rows are grouped n Class, according to the clustering results to get the graph segmentation matrix , P No.i Line j The column elements have the following values: ; In the above step S22, the adjacency matrix of the coarsened graph is calculated according to the graph partition matrix W c and image signal X c , the calculation formula is as follows: W c =P T WP ; X c =P T X .

[0011] In the above step S3, a graph complexity entropy is proposed The algorithm is as follows.

[0012] In the above step S31, based on the image signal obtained in step S2, m The specific steps for constructing the vector sequence of dimension are as follows: X c (When the scale factor is 1, the graph signal is X ) to reconstruct the phase space and form n A sequence of vectors , is the embedding dimension, that is, the dimension of the vector, where: ; ; In the above step S32, the distance is defined For vector x m ( i )and x m ( j ), that is: ; For each vector x m ( i ), we need to calculate the Manhattan distance between it and other vectors (including its own), so for each vector x m ( i )have n distance, given a threshold r , the statistical distance is less than the threshold rThe number of P m ( r )for: ; Calculating embedding dimensions m +1 P m+1 ( r ),but E 1 is: E 1= P m ( r )- P m+1 ( r ).

[0013] In the above step S33, according to the calculated value in S32 P m ( r )and P m+1 ( r ),but E 2 is: .

[0014] In the above step S34, all Manhattan distances obtained in step S32 (excluding i = j The distance at that time is calculated by using the histogram method to estimate the empirical probability density function of the distance. Suppose the histogram has B intervals, and the probability of all distances falling into each interval is ,definition E 3 is: .

[0015] In the above step S35, define For vector x m ( i )and x m ( j ), that is: ; in, express x m ( i )middle m The average of the values, express x m ( j )middlem The average of the values; that is: ; ; Calculate the Pearson correlation coefficient between all adjacent vectors and get n- 1 correlation coefficient, the range of the Pearson correlation coefficient is [-1, 1], and [-1, 1] is evenly divided into A subintervals, divide all correlation coefficients into subintervals, and count the probability of falling into each subinterval ,definition E 4 is: .

[0016] In the above step S36, for each graph under each scale factor, the values ​​calculated by S31-S35 are As the multi-scale complex entropy at the corresponding scale; the multi-scale complex entropy of all scale factors is constructed as the characteristic vector of the test piece.

[0017] In the above step S4, a model such as a random forest model, a support vector machine, a logistic regression or a decision tree is used as a train bearing fault classifier.

[0018] The above train bearing fault classifier is trained through the following steps: (1) Based on the training set, the random forest model is used to screen the features in the feature vector, and the screened features constitute the updated feature vector; (2) Reconstruct the training set based on the updated feature vector to obtain the reconstructed training set; (3) Use the reconstructed training set to train the train bearing fault classifier.

[0019] In the above step (1), the training set is obtained according to the following steps: different types of train bearing fault history sample data are used as training sample data, and the training data are processed according to the steps S1-S3 given above to obtain the feature vector of each sample; all samples and their feature vectors constitute the training set.

[0020] Using the random forest model to learn the training set, the mean impurity reduction (MDI), mean accuracy reduction (Mean Decrease Accuracy) or permutation importance of each feature can be calculated based on the learning results as the feature importance of the feature, and the feature importance is set as the feature screening indicator to screen out the features that meet the requirements (for example, sort the feature importance and screen out the top M (for example, the top 15 or top 20)) to form the updated feature vector. Then, based on the updated feature vector, the feature value of each sample in the training set is screened, and the screened feature values ​​are used to reconstruct the training set to obtain the reconstructed training set.

[0021] In the above step S4, the feature vector of the part to be tested is screened according to the features screened in the previous step (1).

[0022] Compared with the prior art, the train bearing fault diagnosis method based on graph multi-scale complex entropy provided by the present invention has the following beneficial effects: (1) Aiming at the problem of low accuracy of vibration signal feature extraction in the prior art, the present invention proposes a new feature extraction method, and combines the entropy theory to quantify the complexity and irregularity of the signal at multiple angles and multiple scales, and then proposes a multi-scale complex entropy graph, and uses the multi-scale complex entropy as a feature vector to identify train bearing faults; (2) The present invention also uses a tree classification algorithm as the basis of the classification model. First, features with higher feature importance are screened out from the multi-scale complex entropy of the graph, and then the train bearing fault is identified based on these features, which can improve the accuracy and effectiveness of train fault identification; (3) The present invention can effectively reduce the interference of background noise and impurity frequency at the same time, has a strong fault feature extraction capability, and has high robustness. It can provide a higher accuracy rate for the fault diagnosis results of the bearing, and realize accurate diagnosis of different fault types and fault degrees of train bearings. It can effectively monitor train bearing faults and improve the safety and economy of train operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a schematic diagram of the flow chart of the train bearing fault diagnosis method based on graph multi-scale complex entropy proposed in the present invention.

[0024] Figure 2 is the confusion matrix of the fault detection result in the embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following will clearly and completely describe the technical solutions of various embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Example

[0026] This embodiment provides a train bearing fault diagnosis method based on graph multi-scale complex entropy. Figure 1 As shown, the following steps are included: S1 converts the collected original vibration signal of the test piece from the time domain to the image domain; S2 performs coarsening processing on the signal in the image domain to obtain a multi-scale coarse image; S3 calculates the multi-scale complex entropy of the graph of the device under test based on the multi-scale graph obtained in S2, and uses it as the feature vector of the device under test; S4 inputs the features screened out from the feature vector of the test piece into the train bearing fault classifier to obtain the train bearing fault classification result and complete the train bearing fault diagnosis.

[0027] In the above step S1, during the process of collecting the original vibration signal, an acceleration sensor is installed on the surface of the train axle box, and the acceleration sensor located on the surface of the axle box obtains the train bearing fault signal, and the vibration signal is used as the original signal for fault diagnosis. The collected original time series signal is mapped into a two-dimensional matrix to obtain an undirected weighted graph, and the mapping method adopts a visual graph algorithm; through this method, the information in the one-dimensional time series is stored in a graph matrix form containing a set of bidirectional relationships between objects and the strength of these relationships, so that the time series signal can be analyzed in the form of a visual graph. Then, based on the graph spectrum theory, the graph matrix representation is obtained, including the adjacency matrix, the degree diagonal matrix and the Laplace matrix, and the eigenvalues ​​and eigenvectors of the normalized Laplace matrix are obtained to realize visual multidimensional analysis.

[0028] In this embodiment, the number of sampling points of the sample is set to 2048, and each element in the sample is used as a node in the graph. An undirected weighted graph is constructed according to the visibility relationship between the nodes, and the Euclidean distance between two nodes is used as the weight of the edge between the two nodes. The calculation formula is as follows: ; In the formula, x i and x j Node i and nodes j The value of , the defined weight is the Euclidean distance between two nodes, which truly reflects the difference between the vertices.

[0029] Based on the spectral graph theory, we get the matrix representation of the graph, including the adjacency matrix W , degree diagonal matrix D and the graph Laplacian matrix L ; In the degree diagonal matrix, for each node in the graph i , ;and L=DW , then L Perform standard orthogonal decomposition to obtain eigenvalues ​​and eigenvectors U .

[0030] By the following formula L Perform standard orthogonal decomposition and find eigenvalues ​​and eigenvectors: ; in, The elements on the diagonal are the eigenvalues ​​of, U The column is L The feature vector of .

[0031] In the above step S2, a coarsening algorithm is proposed to coarsen the signal in the image domain to obtain a multi-scale coarse image. s is the scale factor, n is the number of nodes in the coarsened graph, and In this embodiment, the scale factor is s= 1,2,…20.

[0032] For each scale factor, the coarsening process of the image domain signal includes the following steps: S21 clusters the graph domain signals to obtain the graph segmentation matrix.

[0033] In this step, each column of the eigenvector matrix is ​​first standardized to [-1, 1], and then the eigenvectors with absolute values ​​less than 0.001 are regarded as 0. The non-zero rate of each eigenvector (that is, the ratio of non-zero eigenvalues ​​in the eigenvector) is calculated, and the standard deviation of the non-zero rate of all eigenvectors is obtained. The eigenvectors with non-zero rates higher than 3 times the standard deviation are selected, and the K-means clustering algorithm is applied to the rows of the eigenvectors. The parameters in the algorithm are k = n , N The rows are grouped n Class, according to the clustering results to get the graph segmentation matrix , P No. i Line j The column elements have the following values: .

[0034] S22 calculates the adjacency matrix and graph signal of the coarsened graph based on the graph segmentation matrix.

[0035] In this step, the adjacency matrix of the coarsened graph is calculated based on the graph partition matrix. W c and image signal X c , the calculation formula is as follows: W c =P T WP ; X c =P T X .

[0036] For multiple scale factors, the above steps S21-S22 are repeated, so that the image domain signal is coarsened into multiple coarse images of different scales, that is, a multi-scale coarse image.

[0037] In the above step S3, a graph complexity entropy algorithm is proposed to extract the features of the test piece.

[0038] Defining Graph Complexity Entropy ,The specific steps of calculating the complex entropy of the graph are as follows: S31 reconstructs the phase space of the image signal under each scale factor and constructs m dimensional vector sequence.

[0039] In this step, based on the image signal obtained in step S2, m The specific steps for constructing the vector sequence of dimension are as follows: X c (When the scale factor is 1, the graph signal is X ) to reconstruct the phase space and form n A vector sequence is the embedding dimension, that is, the dimension of the vector. In this embodiment, the embedding dimension is 2, where: ; .

[0040] S32 Calculation m The Manhattan distance between the dimensional vector sequences is used as the similarity, and the statistical similarity is less than the threshold r The number of similarity probabilities is calculated to obtain the logarithmic mean value. P m ( r ), and similarly calculate the embedding dimension as m The logarithmic mean of the similarity probability under +1 P m+1 ( r ), Pm ( r )and P m+1 ( r ) is E 1。

[0041] In this step, define the distance For vector x m ( i )and x m ( j ), that is: ; For each vector x m ( i ), we need to calculate the Manhattan distance between it and other vectors (including its own), so for each vector x m ( i )have n distance, given a threshold r (Take 0.25 times the standard deviation of the time series), the statistical distance is less than the threshold r The number of P m ( r )for: ; Then calculate the embedding dimension m=3 Next P 3( r ),but E 1 is recorded as: E 1= P 2( r )- P 3( r ).

[0042] S33 According to step S32 P m ( r )and P m+1 ( r ),calculate for E 2.

[0043] According to the calculation in S32 P 2( r )and P 3( r ),but E 2 is recorded as: .

[0044] S34 Based on the Manhattan distance between the dimensional vector sequences obtained in step S32, the empirical probability density function of the distance is estimated using the histogram method, and the obtained E 3. , P b represents the probability of falling into the b interval of the histogram, and B represents the number of intervals of the histogram.

[0045] In this step, all Manhattan distances obtained in step S32 (excluding i = j The distance at that time is calculated by using the histogram method to estimate the empirical probability density function of the distance. Suppose the histogram has B intervals (the number of intervals in this embodiment is 500), and the probability of all distances falling into each interval (that is, the percentage of distances falling into each interval to all distances) is: ,definition E 3 is: .

[0046] S35 calculates the Pearson correlation coefficient between adjacent dimensional vector sequences. The correlation coefficient range is [-1, 1]. [-1, 1] is evenly divided into A subintervals, divide all correlation coefficients into subintervals, count the probability of falling into each subinterval, and calculate E 4. , P a Indicates that it falls into the subinterval a probability.

[0047] In this step, define For vector x m ( i )and x m ( j ), that is: ; in, express x m ( i )middle m The average of the values, express x m ( j )middle m The average of the values.

[0048] Calculate the Pearson correlation coefficient between all adjacent vectors and get n- 1 correlation coefficient, the range of the Pearson correlation coefficient is [-1, 1], and [-1, 1] is evenly divided into A subintervals (the number of intervals in this embodiment is 50), all correlation coefficients are divided into subintervals, and the probability of falling into each subinterval is counted ,definition E 4 is: .

[0049] S36 For each scale factor graph, calculate the multi-scale complex entropy of the graph , and use it as the feature vector of the device under test.

[0050] In this embodiment, for each graph under each scale factor, the values ​​calculated by S31-S35 are As the multi-scale complex entropy at the corresponding scale; the multi-scale complex entropy of all scale factors is constructed as the characteristic vector of the test piece.

[0051] For each scale factor s , obtain the rough graph through step S2, and then repeat the above steps S31 to S36 to calculate four entropy values, namely the complex entropy of the graph. s Taking 1, 2, ... 20, one sample can obtain 80 entropy values, and these 80 entropy values ​​are constructed as the feature vector of the test piece.

[0052] In the above step S4, during the process of collecting the original vibration signal, an acceleration sensor is installed on the surface of the train axle box, and the train bearing fault signal is obtained by the acceleration sensor located on the surface of the axle box, and the vibration signal is used as the original signal for fault diagnosis.

[0053] In this embodiment, a random forest model is used as a train bearing fault classifier.

[0054] The above train bearing fault classifier is trained through the following steps: (1) Based on the training set, the random forest model is used to screen the features in the feature vector, and the screened features constitute the updated feature vector.

[0055] The training set is obtained according to the following steps: different types of train bearing fault history sample data are used as training sample data, and the training data are processed according to the steps S1-S3 given above to obtain the feature vector of each sample; all samples and their feature vectors constitute the training set.

[0056] In this step, the random forest model is used to learn the training set. Based on the learning results, the mean impurity reduction (MDI) of each feature can be calculated as the feature importance of the feature.

[0057] The calculation process of mean impurity reduction (MDI) is as follows: Gini Impurity is used as the impurity indicator to calculate the reduction in impurity of the feature in the decision tree before and after the data is split. The average of the impurity reduction of the feature in all decision trees is the mean impurity reduction.

[0058] The MDI calculation formula is: ; Where T represents the number of decision trees. Indicates i The Gini impurity of the node before the split in the tree, Indicates i Gini impurity of the nodes after the split in the tree.

[0059] The feature importance is set as the index of feature screening, and the features that meet the requirements are screened out to form the updated feature vector. In this embodiment, as mentioned above, each sample has 80 entropy values ​​as feature values, and the feature importance of these 80 entropy values ​​is sorted from large to small, and the top 20 features are screened out to form the updated feature vector.

[0060] Then, based on the updated feature vector, the feature value of each sample in the training set is screened, and the training set is reconstructed using the screened feature values ​​to obtain a reconstructed training set.

[0061] (3) Use the reconstructed training set to train the train bearing fault classifier.

[0062] In this step, the random forest model trained in step (1) is further trained using the reconstructed training set to obtain the final train bearing fault classifier. The random forest model is trained using conventional methods disclosed in the art. For the training process, see Diesel Engine Turbocharger Fault Diagnosis Based on Random Forest Method, Jia Zheyu, Wen Huabing, Zhu Junchao, Zhao Zhenyu, Ship Science and Technology, Vol.45, No.6, 2023.

[0063] For the part to be tested, firstly, the feature vector of the part to be tested is screened according to the features screened in the previous step (1). Then, the screened features are input into the trained train bearing fault classifier to obtain the fault type of the part to be tested.

[0064] This embodiment uses the bearing vibration signal collected by the Railway Research and Education Center of the University of Birmingham for example verification.

[0065] Several samples of different fault types are collected. Seven different fault types are given here, as shown in Table 1. 30 vibration signal samples are collected for each health state, a total of 210 sample data, and the ratio of the training set to the test set is 1:1. The 210 sample data are randomly divided according to this ratio.

[0066] Table 1 Classification of train bearing health status

[0067] According to the steps S1-S3 given above, the multi-scale complex entropy of the graph of the training set and the test set samples is calculated.

[0068] According to the previous step (1), based on the training set, the features in the feature vector are screened using the random forest model, and then the training set is reconstructed using the screened features to train the random forest model as the train bearing fault classifier to obtain the final train bearing fault classifier.

[0069] The trained train bearing fault classifier is then tested using the test set. According to the features selected in the previous step (1), the feature vectors composed of the multi-scale complex entropy of the graph corresponding to each sample in the test set are selected, and then the selected features are input into the trained train bearing fault classifier to obtain the prediction results of the fault type of each sample. The confusion matrix of the final classification result is as follows: Figure 2 As shown, the results show that five sample fault types are classified incorrectly, and the classification accuracy is 95.2%. Therefore, this method can effectively and accurately diagnose different types and degrees of bearing faults, and can provide a higher accuracy rate for the bearing fault diagnosis results. It realizes the accurate diagnosis of train bearing fault types, can effectively monitor train bearing faults, and improve the safety and economy of train operation.

[0070] Those skilled in the art will appreciate that the embodiments herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A train bearing fault diagnosis method based on graph multi-scale complex entropy, comprising the following steps: S1 converts the collected original vibration signal of the test piece from the time domain to the image domain; S2 performs coarsening processing on the signal in the image domain to obtain a multi-scale coarse image, which specifically includes the following sub-steps: S21 clusters the graph domain signal to obtain a graph segmentation matrix; S22 calculates the adjacency matrix and graph signal of the coarsened graph according to the graph partition matrix; Select multiple scale factors, repeat the above steps S21-S22, and coarsen the image domain signal into multiple coarse images of different scales, i.e., multi-scale coarse images; S3 calculates the multi-scale complex entropy of the DUT graph based on the multi-scale graph obtained in S2, and uses it as the feature vector of the DUT to define the graph complex entropy ,The specific steps of calculating the complex entropy of the graph are as follows: S31 reconstructs the phase space of the image signal under each scale factor and constructs m dimensional vector sequence; S32 Calculation m The Manhattan distance between the dimensional vector sequences is used as the similarity, and the number of similarities less than the threshold r is counted to calculate the logarithmic mean of the similarity probability. P m ( r ), and similarly calculate the embedding dimension as m The logarithmic mean of the similarity probability under +1 P m+1 ( r ), P m ( r )and P m+1 ( r ) is E 1; S33 According to step S32 P m ( r )and P m+1 ( r ),calculate for E 2; S34 is obtained according to step S32 m The Manhattan distance between the dimensional vector sequences is estimated using the histogram method to estimate the empirical probability density function of the distance and calculated E 3. , P b represents the probability of falling into the b interval of the histogram, and B represents the number of intervals of the histogram; S35 calculates adjacent m The Pearson correlation coefficient between dimensional vector sequences is in the range of [-1, 1]. [-1, 1] is evenly divided into A subintervals, divide all correlation coefficients into subintervals, count the probability of falling into each subinterval, and calculate E 4. , P a Indicates that it falls into the subinterval a probability; S36 For each scale factor, calculate , that is, the multi-scale complex entropy of the graph is obtained and used as the feature vector of the device under test; S4 inputs the features screened out from the feature vector of the test piece into the train bearing fault classifier to obtain the train bearing fault classification result and complete the train bearing fault diagnosis.

2. The train bearing fault diagnosis method based on graph multi-scale complex entropy according to claim 1 is characterized in that: In step S1, the specific implementation method is to map the collected original time series signal into a two-dimensional matrix to obtain an undirected weighted graph; the mapping method is a visible graph, a horizontal visible graph or a k-nearest neighbor graph.

3. The train bearing fault diagnosis method based on graph multi-scale complex entropy according to claim 1 is characterized in that: In step S2, the original image , the original image signal is X , W is the adjacency matrix, U is the eigenvector matrix obtained by orthogonal decomposition of the graph Laplacian matrix, s is the scale factor, n is the number of nodes in the coarsened graph, and ; In step S21, the eigenvector matrix U Each column of is normalized to [-1, 1], and the eigenvectors with absolute values ​​less than 0.001 are regarded as 0. The non-zero rate of each eigenvector is calculated, and the standard deviation of the non-zero rate is obtained. The eigenvectors with non-zero rates higher than 3 times the standard deviation are selected, and the K-means clustering algorithm is applied to the rows of the eigenvectors. The parameters in the algorithm are k = n , N The rows are grouped n Class, according to the clustering results to get the graph segmentation matrix , P No. i Line j The column elements have the following values: ； In step S22, the adjacency matrix of the coarsened graph is calculated according to the graph partition matrix and image signal , the calculation formula is as follows: W c =P T WP ； X c =P T X 。 4. The train bearing fault diagnosis method based on graph multi-scale complex entropy according to claim 1 is characterized in that: In step S31, based on the image signal obtained in step S2, m The specific steps for constructing the vector sequence of dimension are as follows: X c Reconstruct the phase space to form n A sequence of vectors , is the embedding dimension, that is, the dimension of the vector, where: ； ； In step S32, the distance is defined For vector x m ( i )and x m ( j ), that is: ； For each vector x m ( i ), we need to calculate the Manhattan distance between it and other vectors, so for each vector x m ( i )have n distance, given a threshold r , the statistical distance is less than the threshold r The number of P m ( r )for: ； Calculating embedding dimensions m +1 P m+1 ( r ),but E 1 is: E 1= P m ( r )- P m+1 ( r )。 5. The train bearing fault diagnosis method based on graph multi-scale complex entropy according to claim 1 is characterized in that: In step S34, all Manhattan distances obtained in step S32 are not included. i = j The distance at that time; use the histogram method to estimate the empirical probability density function of the distance, assuming that the histogram has B intervals, and the probability of all distances falling into each interval is calculated as ,definition E 3 is: ； In step S35, define For vector x m ( i )and x m ( j ), that is: ； in, express x m ( i )middle m The average of the values, express x m ( j )middle m The average of the values; Calculate the Pearson correlation coefficient between all adjacent vectors and get n- 1 correlation coefficient, the range of the Pearson correlation coefficient is [-1, 1], and [-1, 1] is evenly divided into A subintervals, divide all correlation coefficients into subintervals, and count the probability of falling into each subinterval ,definition E 4 is: 。 6. The train bearing fault diagnosis method based on graph multi-scale complex entropy according to any one of claims 2 to 5, characterized in that: In step S36, for each scale factor, the values ​​calculated in steps S31-S35 are As the multi-scale complex entropy at the corresponding scale; the multi-scale complex entropy of all scale factors is constructed as the characteristic vector of the test piece.

7. The train bearing fault diagnosis method based on graph multi-scale complex entropy according to claim 1 is characterized in that: In step S4, a random forest model is used as a train bearing fault classifier.

8. The train bearing fault diagnosis method based on graph multi-scale complex entropy according to claim 1 or 7 is characterized in that: The train bearing fault classifier is trained through the following steps: (1) Based on the training set, the random forest model is used to screen the features in the feature vector, and the screened features constitute the updated feature vector; (2) Reconstruct the training set based on the updated feature vector to obtain the reconstructed training set; (3) Use the reconstructed training set to train the train bearing fault classifier.

9. The train bearing fault diagnosis method based on graph multi-scale complex entropy according to claim 8 is characterized in that: In step (1), the training set is obtained according to the following steps: different types of train bearing fault history sample data are used as training sample data, and the training data are processed according to the steps S1-S3 given above to obtain the feature vector of each sample; all samples and their feature vectors constitute the training set.

10. The train bearing fault diagnosis method based on graph multi-scale complex entropy according to claim 9 is characterized in that: In step (2), based on the training results, the average impurity reduction (MDI), the average accuracy reduction (MeanDecrease Accuracy) or the permutation importance (Permutation importance) of each feature is calculated as the feature importance of the feature, and the feature importance is set as the feature screening indicator to screen out the features that meet the requirements and form an updated feature vector; then, based on the updated feature vector, the feature values ​​of each sample in the training set are screened, and the training set is reconstructed using the screened feature values ​​to obtain a reconstructed training set.

Images