Intelligent production line rotating machinery health assessment method based on hypergraph convolution fuzzy network

Abstract

The application relates to an intelligent production line rotating machinery health evaluation method based on a hypergraph convolution fuzzy network, which comprises the following steps: collecting time domain signals generated by various faults of a rotating machinery system during operation to obtain corresponding time-frequency representations; dividing the time domain signals and the time-frequency representations according to fault types, adding labels, and constructing vibration data; using a hypergraph model to construct hypergraphs with the time domain signals and the time-frequency representations as graph nodes to obtain two hypergraphs under each fault type; using a weight fuzzy module to fuse the nodes of the two hypergraphs under each fault type; inputting the fused hypergraph features into a multilayer perception hybrid module; calculating the probabilities of all fault classes through an output full connection layer and a classification function to realize training, optimization and updating of the hypergraph convolution fuzzy network. The application increases the interpretability of the model and can obtain more robust rotating machinery system health evaluation results.

Description

Technical Field

[0001] The present invention relates to the technical field of health assessment of rotating machinery systems, and in particular to a health assessment method for rotating machinery in an intelligent production line based on a hypergraph convolutional fuzzy network. Background Art

[0002] Catastrophic failures of rotating machinery systems in smart production lines can result in significant economic losses. To prevent potential risks that could lead to such failures, health assessment is a core task in the maintenance of intelligent machinery. The Industrial Internet of Things (IIoT) can provide intelligent assessments with the big data generated by equipment operation. Currently, the IIoT focuses primarily on developing hardware and software for tracking the health status of monitored system components. Accurate health assessments can pave the way for this development. Existing methods for health assessment of rotating machinery systems fall into two main categories: fault identification based on mathematical statistics and fault detection based on machine learning. Fault identification based on mathematical statistics often uses statistical methods to extract identifiable feature vectors from the raw data and then employs classifiers to distinguish them. Common statistical features include feature entropy, kurtosis, mean, and spectrum. These methods require prior analysis based on prior knowledge, which can result in the loss of many rich features. Manually extracted feature indicators are closely related to the accuracy of health assessments, and any loss of these indicators can affect the final assessment results. Fault detection methods based on deep neural network models aim to develop deep nonlinear mapping structures that enable gradient calculation for complex problems. Specifically, they establish an iterative training process based on large data streams to extract knowledge. Deep network structures are complex, and generally speaking, the interpretability of such deep models is extremely weak.

[0003] In the current fault detection methods based on deep neural network models, feature extraction in deep learning problems lacks rich fault semantics, that is, there is a lack of relevant descriptions of the interpretability of the model, which makes it impossible for researchers to further understand the working mechanism of the deep model and improve the rationality of the model structure. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention provides an intelligent production line rotating machinery health assessment method based on a hypergraph convolutional fuzzy network, aiming to increase the interpretability of the model and thereby obtain more robust rotating machinery system health assessment results.

[0005] The technical solution adopted in the present invention is as follows:

[0006] A method for health assessment of rotating machinery in an intelligent production line based on a hypergraph convolutional fuzzy network. The hypergraph convolutional fuzzy network includes a hypergraph model, a weighted fuzzy module, a multi-layer perception hybrid module, an output fully connected layer, and a classification function.

[0007] The method comprises:

[0008] S1. Collect time domain signals generated by various faults during the operation of the rotating machinery system, divide the time domain signals according to the fault type and add labels, obtain the time-frequency representation of each time domain signal, and mark the corresponding labels to form vibration data;

[0009] S2. Input the vibration data into the hypergraph model, construct a hypergraph using the time domain signal and its time-frequency representation as graph nodes, and obtain two hypergraphs for each fault type;

[0010] S3. Using the weight fuzzy module to perform data fusion on the nodes of the two hypergraphs under each fault type, to achieve fusion of the two hypergraph features and obtain a fused hypergraph;

[0011] S4. Input the fused hypergraph features into the multi-layer perception hybrid module to suppress knowledge divergence in the graph structure;

[0012] S5. Calculate the probabilities of all fault classes through the output fully connected layer and the classification function, and calculate the loss function of the fault assessment to achieve the training, optimization, and update of the hypergraph convolutional fuzzy network;

[0013] S6. Use the trained hypergraph convolutional fuzzy network to perform health assessment on the rotating machinery system.

[0014] Further technical solutions are:

[0015] In step S3, the weight fuzzy module is used to perform data fusion on the nodes of the two hypergraphs under each fault type, including:

[0016] Initialize the connection between the nodes of the two hypergraphs;

[0017] Calculate the node fuzzy entropy between the two hypergraphs in sequence:

[0018]

[0019] Among them, a is the adjustment factor, v i , The i-th node in the hypergraph constructed for the time domain signal and the j-th node in the hypergraph constructed for the time-frequency representation respectively;

[0020] The node fuzzy entropy is used as the fuzzy membership to calculate the fusion weight:

[0021]

[0022] Among them, m and n are the total number of nodes in the hypergraph constructed by time domain signal and the hypergraph constructed by time-frequency representation, respectively;

[0023] Based on the fusion weight I, the node relationship in the two hypergraphs is determined according to the following fuzzy rules:

[0024]

[0025] Among them, 1 indicates a strong relationship, 0 indicates a slight relationship, and ξ is a predefined threshold;

[0026] Two or more nodes with strong relationships are connected through hyperedges to replace the node initialization connection, thereby transferring the most critical semantics of the two hypergraphs to the fused hypergraph.

[0027] The structure of the multi-layer perception hybrid module includes a first graph convolution layer, a self-attention mechanism, a second graph convolution layer, a first multi-layer perceptron and a second multi-layer perceptron, a third graph convolution layer and an embedding vector;

[0028] The first multilayer perceptron and the second multilayer perceptron are composed of multiple fully connected layers of the same size;

[0029] The features output by the first convolutional layer Build the dependency relationship of the self-attention mechanism:

[0030]

[0031] Among them, ψ1 is the output feature of the self-attention mechanism, conv1(), conv2(), conv3() represent the first graph convolution operation, the second graph convolution operation, and the third graph convolution operation in the self-attention mechanism respectively;

[0032] The features output by the second graph convolutional layer After linear transformation and mapping in the first multilayer perceptron:

[0033]

[0034] where ψ2 is the output feature of the first multilayer perceptron, w1 and w2 are the weights of the first and second fully connected layers in the first multilayer perceptron, σ(·) is the unit nonlinearity Gaussian error linear unit; LayerNorm() is the layer regularization operation;

[0035] The second multilayer perceptron Mapping with ψ2:

[0036]

[0037] Where ψ3 is the output feature of the second multilayer perceptron, w3 and w4 are the weights of the first and second fully connected layers in the second multilayer perceptron respectively;

[0038] Align the ψ1, ψ2, and ψ3 dimensions and complete the embedding vector by feedforward combination The embedding vector is also used as the input of the third graph convolutional layer to achieve the accumulation of the first layer of knowledge in the last layer.

[0039] In step S2, the information transmission mechanism of the graph convolutional neural network used to construct the hypergraph is as follows:

[0040]

[0041] In the above formula:

[0042] X l and X l-1 represents the output and input of the graph convolutional neural network, σ(·) represents the ReLU activation function, and θ l-1 are network parameters;

[0043] is the smallest non-negative eigenvector, which can be calculated by the identity matrix I, where:

[0044] D v Is a diagonal matrix, representing the node degree matrix, and the elements on the diagonal are the degree d(v) of each node v in the hypergraph;

[0045] D e is a diagonal matrix representing the degree matrix of hyperedges, where the elements on the diagonal are the degree δ(e) of each hyperedge e in the hypergraph;

[0046] H is dimensional point-edge incidence matrix, They are respectively the node set, hyperedge set, and weight set assigned to hyperedges of the hypergraph.

[0047] The degree of a node is calculated as follows:

[0048]

[0049] Among them, v j ,e j are the jth node and hyperedge in the hypergraph, w(e j ) represents the hyperedge e j The average value of the heat core distance between any two vertices on it;

[0050] The degree of a hyperedge is calculated as follows:

[0051]

[0052] In step S2, the optimization of the hypergraph is transformed into finding the eigenvector corresponding to the minimum eigenvalue of the matrix bundle (L, A) to achieve graph segmentation;

[0053] L is the Laplacian matrix of the hypergraph,

[0054] A is the adjacency matrix of the hypergraph, The value a of each element in the adjacency matrix i,j =w(e n ), w(e n ) represents the nth hyperedge e n Node v at any two vertices i ,v j The average value of the hot core distance is:

[0055]

[0056] x i ,x j and v i ,v j The corresponding time-frequency representation or time domain signal, u is the Gaussian kernel size.

[0057] Before step S4, the method further includes:

[0058] Repeat S1 to S3, use more vibration data to iteratively train the hypergraph model, expand the number of graph nodes and layers of the fused hypergraph, and construct the hypergraph model into a deep neural network with high fault feature extraction capability.

[0059] The vibration data obtained in step S1 is expressed as follows:

[0060]

[0061] in, is a set of time domain signals The corresponding time-frequency representation, c represents the set of all fault classes, the lth fault class is the jth time domain signal and its time-frequency representation vector under the lth fault type, and N represents the number of data.

[0062] The time domain signals generated by various faults during the operation of the rotating mechanical system are collected, including:

[0063] An acceleration sensor is installed on the cylinder head of the fuel and valve system of the rotating mechanical system, and a reference standard reflective label of the dead point signal measured by the photoelectric speed sensor is affixed to the main shaft end of the rotating mechanical system. Based on the dead point signal, a time domain signal with the same time length is collected.

[0064] A cross entropy function is used as a loss function for calculating fault assessment to implement the training, optimization, and updating of the hypergraph convolutional fuzzy network.

[0065] The beneficial effects of the present invention are as follows:

[0066] This paper introduces the idea of ​​fuzzy reasoning into the traditional graph convolutional neural network method, and proposes a hypergraph convolutional fuzzy architecture with a multi-layer perception hybrid module and a weighted fuzzy module. It can well solve the lack of rich fault semantics in feature extraction in traditional deep learning problems, and successfully realizes the fusion of different fault features.

[0067] This paper designs a fuzzy rule-driven graph construction method to induce a dynamically changing cascading behavior of constructing data from two different input feature subsets and perform data fusion in the graph, thereby transferring the most critical semantics of the two graphs into the final constructed hypergraph.

[0068] The present invention uses a novel weighted coding to generate heterogeneous embedding, namely, multi-layer perceptual hybrid embedding, to suppress knowledge divergence in the graph structure, enabling the model to further learn fault characteristics, better enhance the semantic transfer between graphs, and effectively capture the key features between data, thereby obtaining more robust rotating machinery system health assessment results.

[0069] Other features and advantages of the present invention will be set forth in the following description or may be learned by practicing the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 Flowchart of a method according to an embodiment of the present invention.

[0071] Figure 2 Schematic diagram of the hypergraph convolutional fuzzy network structure of an embodiment of the present invention.

[0072] Figure 3 This is a schematic diagram of the fuzzy rule-driven graph construction principle of an embodiment of the present invention.

[0073] Figure 4 This is a vibration signal collected during the experimental verification process of the embodiment of the present invention.

[0074] Figure 5 This is the prediction result obtained by experimental verification of the embodiment of the present invention. DETAILED DESCRIPTION

[0075] The specific embodiments of the present invention are described below with reference to the accompanying drawings.

[0076] See also Figure 1 and Figure 2 , this embodiment provides a method for health assessment of rotating machinery in an intelligent production line based on a hypergraph convolutional fuzzy network, wherein the hypergraph convolutional fuzzy network includes a hypergraph model, a weight fuzzy module, a multi-layer perception hybrid module, an output fully connected layer, and a classification function;

[0077] See also Figure 1 , the method comprising:

[0078] S1. Collect time domain signals, i.e., vibration signals, generated by various faults during the operation of the rotating machinery system, divide the time domain signals according to the fault type and add labels, obtain the time-frequency representation of each time domain signal, and mark the corresponding labels to form vibration data;

[0079] As a specific implementation, the acquisition of the vibration signal includes:

[0080] Install the acceleration sensor on the cylinder head of the fuel and gas distribution system of the rotating mechanical system, and stick the reference standard reflective label of the dead point signal measured by the photoelectric speed sensor on the main shaft end of the rotating mechanical system. Perform key time positioning based on the dead point signal to obtain vibration signals with the same time length.

[0081]

[0082] The vibration signal Divide into m types of fuel and gas distribution system fault types, and set corresponding fault labels Y = {y 1 ,y 2 ,…,y m}, where x m The vibration signal representing the mth fault type, y m A label vector representing the mth fault type.

[0083] The constructed vibration data can be expressed as follows:

[0084]

[0085] in, is a set of time domain signals The corresponding time-frequency representation, c represents the set of all fault classes, the lth fault class is the jth time domain signal and its time-frequency representation vector under the lth fault type, and N represents the number of data.

[0086] S2. Input the vibration data into the hypergraph model, construct a hypergraph using the time domain signal and its time-frequency representation as graph nodes, and obtain two hypergraphs for each fault type.

[0087] The hypergraph constructed by time domain signal under the type l fault can be expressed as A set of nodes representing a hypergraph, represents the set of directed hyperedges, Represents the weight matrix assigned to the hyperedge. The larger the value in the weight matrix, the greater the influence of the corresponding edge. At the same time, for any hyperedge is a node set A subset of . The hypergraph constructed by the time-frequency representation of the corresponding time-domain signal is

[0088] As a specific embodiment, the construction of the hypergraph includes:

[0089] (1) Calculate the degree of the node:

[0090]

[0091] where v j ,e j are the jth node and hyperedge in the hypergraph, respectively, and w(e j ) represents the average value of the hot kernel distance between any two vertices of the hyperedge e j .

[0092] (2) Calculate the degree of the hyperedge:

[0093]

[0094] H is a point-edge association matrix of dimension , as known from the above, and ε are the node set and hyperedge set of the hypergraph, respectively. In the point-edge association matrix, if a node is on a hyperedge, then H(v,e) = 1, otherwise H(v,e) = 0.

[0095] (3) Set the diagonal matrix D v , which represents the degree matrix of the node, and the elements on the diagonal are the degrees d(v) of each node v in the hypergraph; set the diagonal matrix D e , which represents the degree matrix of the hyperedge, and the elements on the diagonal are the degrees δ(e) of each hyperedge e in the hypergraph.

[0096] Calculate the adjacency matrix of the hypergraph representing the connection relationship between each node As known from the above, is the weight matrix assigned to the hyperedge; the value of each element in the adjacency matrix a i,j = w(e n ), w(e n ) represents the average value of the hot kernel distance between the nodes v i ,v j at any two vertices of the nth hyperedge e n .

[0097]

[0098] where x i ,x j are the time-frequency representation or time-domain signal corresponding to v i ,v j , respectively, and u is the size of the Gaussian kernel.

[0099] (4) The optimization of the hypergraph is transformed into solving the eigenvector corresponding to the minimum eigenvalue of the matrix bundle (L, A) to achieve graph segmentation.

[0100] Among them, the Laplacian matrix of the hypergraph The minimum non-negative eigenvector Δ can be calculated using the identity matrix I:

[0101]

[0102] (5) Information transfer of the graph convolutional neural network based on the computational hypergraph model:

[0103]

[0104] Among them, X l and X l-1 represents the output and input of the graph convolutional neural network, σ(·) represents the Recti-fied Linear Units (ReLU) activation function, and θ l-1 are network parameters.

[0105] S3. Utilize the weight fuzzy module to perform data fusion on the nodes of the two hypergraphs under each fault type, realize the fusion of the two hypergraph features, and obtain a fused hypergraph for representing the information cascade.

[0106] The specific steps include:

[0107] Initialize the connection between the nodes of the two hypergraphs;

[0108] Calculate the node fuzzy entropy between the two hypergraphs in sequence:

[0109]

[0110] Among them, a is the adjustment factor, v i , The i-th node in the hypergraph constructed for the time domain signal and the j-th node in the hypergraph constructed for the time-frequency representation respectively;

[0111] The node fuzzy entropy is used as the fuzzy membership to calculate the fusion weight:

[0112]

[0113] Among them, m and n are the total number of nodes in the hypergraph constructed by time domain signal and the hypergraph constructed by time-frequency representation, respectively;

[0114] Based on the fusion weight I, the node relationship in the two hypergraphs is determined according to the following fuzzy rules:

[0115]

[0116] Among them, 1 indicates a strong relationship, 0 indicates a slight relationship, and ξ is a predefined threshold;

[0117] If the relationship is strong, the nodes are connected by a hyperedge, otherwise no connection is required. By connecting two or more nodes with strong relationships by a hyperedge, the node initial connection is replaced, thereby transferring the most critical semantics of the two hypergraphs into the fused hypergraph.

[0118] This embodiment uses a fuzzy weight module to focus on the node entropy of the two hypergraphs, that is, their semantic degrees, so as to transfer the most critical semantics in the two hypergraphs to the final constructed hypergraph (the fused hypergraph).

[0119] like Figure 3 As shown, each hyperedge can establish a connection between two or more nodes, thus being able to construct a many-to-many relationship representation, such as e1 and e2 in the figure. For the convenience of description, Figure 3 An incidence matrix is ​​used in which each row (A→J) represents a node and each column (e1→e3) represents a hyperedge. If a node is part of a particular hyperedge, the corresponding element of the incidence matrix position is set to 1, otherwise it is set to 0.

[0120] S4. Input the fused hypergraph features into the multi-layer perception hybrid module to suppress knowledge divergence in the hypergraph structure.

[0121] like Figure 1 As shown, the structure of the multi-layer perception hybrid module includes a first graph convolution layer (i.e., graph convolution layer 1), a self-attention mechanism, a second graph convolution layer (i.e., graph convolution layer 2), a first multi-layer perceptron (i.e., multi-layer perceptron 1) and a second multi-layer perceptron (i.e., multi-layer perceptron 2), a third graph convolution layer (i.e., graph convolution layer 3) and an embedding vector;

[0122] The first multilayer perceptron and the second multilayer perceptron are composed of multiple fully connected layers of the same size. As a preferred embodiment, each multilayer perceptron includes a fully connected layer 1, a layer regularization, and a fully connected layer 2.

[0123] The features output by the first convolutional layer Build the dependency relationship of the self-attention mechanism:

[0124]

[0125] Among them, ψ1 is the output feature of the self-attention mechanism, conv1(), conv2(), conv3() represent the first graph convolution operation, the second graph convolution operation, and the third graph convolution operation in the self-attention mechanism respectively;

[0126] The features output by the second graph convolutional layer After linear transformation and mapping in the first multilayer perceptron:

[0127]

[0128] where ψ2 is the output feature of the first multilayer perceptron, w1 and w2 are the weights of the first and second fully connected layers in the first multilayer perceptron, σ(·) is the unit nonlinearity Gaussian error linear unit (GeLU); LayerNorm() is the layer regularization operation;

[0129] The second multilayer perceptron Mapping with ψ2:

[0130]

[0131] Where ψ3 is the output feature of the second multilayer perceptron, w3 and w4 are the weights of the first and second fully connected layers in the second multilayer perceptron respectively;

[0132] Align the dimensions of ψ1, ψ2, and ψ3, and then complete the embedding vector through feedforward combination The embedding vector is also used as the input of the third graph convolutional layer to achieve the accumulation of the first layer of knowledge in the last layer.

[0133] S5, through the output fully connected layer (i.e. Figure 1 The fully connected layer 4) and the softmax classification function calculate the probabilities of all fault classes and the loss function of fault assessment to realize the training, optimization and update of the hypergraph convolutional fuzzy network.

[0134] The hypergraph convolutional neural fuzzy network takes the time-frequency representation of vibration data and the time-domain signal as input, constructs the hypergraph fault features obtained by the vibration signal and its time-frequency representation into respective graph nodes, and uses the weighted fuzzy module to complete the data fusion of the graph nodes, that is, the fusion of the two hypergraph fault features. It realizes the extraction of key fault information from the two inputs and reconstructs it into a fused hypergraph feature. The calculation method is as follows:

[0135]

[0136] Where X represents the hypergraph fault feature extracted from the vibration signal in the graph structure, represents the hypergraph fault features extracted from the graph structure by time-frequency representation, Represents hypergraph feature fusion.

[0137] The update of the hypergraph convolutional fuzzy network aims to learn a classifier from a labeled vibration data set. The label probabilities of all classes are calculated using a softmax function. The calculation process is as follows:

[0138]

[0139] in, is the predicted value of the label, w C and b C represents the learnable parameters of the classifier, and F represents the final feature representation

[0140] As a specific implementation, the cross entropy function is used as the loss function for calculating fault assessment:

[0141]

[0142] in, Represents the loss function value, y i is the label value of the i-th sample, is the predicted value of the i-th sample, and N is the number of samples.

[0143] S6. Use the trained hypergraph convolutional fuzzy network to perform health assessment on the rotating machinery system.

[0144] As an improvement to the above embodiment, before step S4, the method further includes:

[0145] Repeat S1 to S3, use more vibration data to iteratively train the hypergraph model, expand the number of graph nodes and layers of the fused hypergraph to improve the graph structure's ability to extract fault features, and construct the hypergraph model into a deep neural network.

[0146] As a specific implementation, during iterative training of the hypergraph model, the two input data are fed into their respective graph structures in the same batch size, thereby extracting hypergraph features from both. Each iterative training uses vibration data from different time batches.

[0147] The effectiveness of the method of this embodiment is verified by specific experiments below.

[0148] The method of this embodiment is used to identify multiple faults in the fuel and gas distribution systems of a rotating machinery system in a smart production line. The method includes the following steps:

[0149] Step 1: Use acceleration sensors and photoelectric speed sensors to collect vibration signals of different fault types of the experimental system, with a sampling frequency of 25kHz, including intake valve clearance 0.1mm larger, intake valve clearance 0.2mm larger, exhaust valve clearance 0.1mm larger, exhaust valve clearance 0.2mm larger, intake filter blockage, exhaust pipe blockage, lubrication advance angle reduced by 3°, lubrication advance angle increased by 3°, abnormal cylinder combustion, cylinder wear, serious cylinder failure and normal, a total of 12 different types of vibration signal data, each of which has 500 samples, and each sample is labeled. On this basis, the time-frequency representation of each sample vibration data is extracted, and the corresponding labels are marked to construct a data set. Among them, the collected vibration data is as follows Figure 4 As shown in the figure, the lower part is the time domain signal, and the upper part is the corresponding time-frequency representation. The system fault types and sample numbers are shown in Table 1:

[0150] Table 1 System failure types and sample numbers

[0151] Fault type Sample size Intake valve clearance is 0.1 mm larger 500 Intake valve clearance is 0.2 mm larger 500 Exhaust valve clearance is 0.1 mm larger 500 Exhaust valve clearance is 0.2 mm larger 500 Air intake filter clogged 500 Exhaust pipe clogged 500 Lubrication advance angle reduced by 3° 500 Lubrication advance angle increased by 3° 500 Abnormal cylinder combustion 500 Cylinder working wear 500 Serious cylinder failure 500 normal 500

[0152] Step 2: The dataset is randomly divided into a training set and a test set in a ratio of 8:2. On this basis, the vibration signal and time-frequency representation in the training set are input into the hypergraph model in parallel to construct their respective graph nodes. After the graph is constructed, the corresponding hypergraph features with fault information are output.

[0153] Step 3: Through the weight fuzzy module, the vibration signal in the hypergraph feature, i.e., the graph structure of the vibration signal, and the hypergraph feature of the time-frequency representation, i.e., the graph structure of the time-frequency signal, are fused to obtain the fused hypergraph feature.

[0154] In the fourth step, repeat the second and third steps and input all batches of vibration signals and time-frequency representations into the hypergraph model, thereby increasing the number of graph nodes and layers and improving the hypergraph model's training and learning of fault features.

[0155] Step 5: Combine the hypergraph model, weight fuzzy module, multi-layer perception hybrid module, fully connected layer and Softmax classification layer to build a hypergraph convolutional fuzzy network, and input the fused hypergraph features into the multi-layer perception hybrid module.

[0156] Step 6: The features of the graph convolution output of the first layer in the multi-layer perception hybrid module are used to construct dependencies using the self-attention mechanism, and the knowledge divergence in the graph structure is suppressed under the hybrid embedding of the multi-layer perceptron and the embedding vector.

[0157] Step 7: Use the output of the multi-layer perception hybrid module as the input of the full connection. Finally, calculate the probabilities of all fault classes in the Softmax classification layer, and use the cross entropy function as the loss function for fault assessment to achieve training, optimization, and update of the overall model.

[0158] Step 8: Save the hypergraph convolutional fuzzy network model with the optimal parameters to perform health assessment on the test set of the rotating machinery system.

[0159] The hypergraph convolutional fuzzy network model is used to identify multiple faults in the fuel and gas distribution systems. The prediction results are as follows: Figure 5 As shown in the figure, the predicted value successfully achieves complete classification of faults, verifying the effectiveness of the method proposed in this embodiment in the health assessment of rotating machinery systems in intelligent production lines.

[0160] Those skilled in the art will understand that the foregoing descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art will be able to modify the technical solutions described in the foregoing embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A method for health assessment of rotating machinery in intelligent production lines based on hypergraph convolutional fuzzy networks, characterized in that: The hypergraph convolutional fuzzy network includes a hypergraph model, a weight fuzzy module, a multi-layer perception hybrid module, an output fully connected layer and a classification function; The method comprises: S1. Collect time domain signals generated by various faults during the operation of the rotating machinery system, divide the time domain signals according to the fault type and add labels, obtain the time-frequency representation of each time domain signal, and mark the corresponding labels to form vibration data; S2. Input the vibration data into the hypergraph model, construct a hypergraph using the time domain signal and its time-frequency representation as graph nodes, and obtain two hypergraphs for each fault type; S3. Using the weight fuzzy module to perform data fusion on the nodes of the two hypergraphs under each fault type, to achieve fusion of the two hypergraph features and obtain a fused hypergraph; S4. Input the fused hypergraph features into the multi-layer perception hybrid module to suppress knowledge divergence in the graph structure; S5. Calculate the probabilities of all fault classes through the output fully connected layer and the classification function, and calculate the loss function of the fault assessment to achieve the training, optimization, and update of the hypergraph convolutional fuzzy network; S6. Performing a health assessment on a rotating machinery system using the trained hypergraph convolutional fuzzy network; The weighted fuzzy module is used to perform data fusion on the nodes of the two hypergraphs under each fault type, including: Initialize the connection between the nodes of the two hypergraphs; Calculate the node fuzzy entropy between the two hypergraphs in sequence: Among them, a is the adjustment factor, v i , The i-th node in the hypergraph constructed for the time domain signal and the j-th node in the hypergraph constructed for the time-frequency representation respectively; The node fuzzy entropy is used as the fuzzy membership to calculate the fusion weight: Among them, m and n are the total number of nodes in the hypergraph constructed by time domain signal and the hypergraph constructed by time-frequency representation, respectively; Based on the fusion weight I, the node relationship in the two hypergraphs is determined according to the following fuzzy rules: Among them, 1 indicates a strong relationship, 0 indicates a slight relationship, and ξ is a predefined threshold; Two or more nodes with strong relationships are connected through hyperedges to replace the node initialization connection, thereby transferring the most critical semantics of the two hypergraphs to the fused hypergraph.

2. The method according to claim 1, characterized in that The structure of the multi-layer perception hybrid module includes a first graph convolution layer, a self-attention mechanism, a second graph convolution layer, a first multi-layer perceptron and a second multi-layer perceptron, a third graph convolution layer and an embedding vector; The first multilayer perceptron and the second multilayer perceptron are composed of multiple fully connected layers of the same size; The features output by the first convolutional layer Build the dependency relationship of the self-attention mechanism: Among them, ψ1 is the output feature of the self-attention mechanism, conv1(), conv2(), conv3() represent the first graph convolution operation, the second graph convolution operation, and the third graph convolution operation in the self-attention mechanism respectively; The features output by the second graph convolutional layer After linear transformation and mapping in the first multilayer perceptron: where ψ2 is the output feature of the first multilayer perceptron, w1 and w2 are the weights of the first and second fully connected layers in the first multilayer perceptron, σ(·) is the unit nonlinearity Gaussian error linear unit; LayerNorm() is the layer regularization operation; The second multilayer perceptron Mapping with ψ2: Where ψ3 is the output feature of the second multilayer perceptron, w3 and w4 are the weights of the first and second fully connected layers in the second multilayer perceptron respectively; Align the ψ1, ψ2, and ψ3 dimensions and complete the embedding vector by feedforward combination The embedding vector is also used as the input of the third graph convolutional layer to achieve the accumulation of the first layer of knowledge in the last layer.

3. The method according to claim 1, characterized in that In step S2, the information transmission mechanism of the graph convolutional neural network used to construct the hypergraph is as follows: In the above formula: X l and X l-1 represents the output and input of the graph convolutional neural network, σ(·) represents the ReLU activation function, and θ l-1 are network parameters; is the smallest non-negative eigenvector, which can be calculated by the identity matrix I, where: D v Is a diagonal matrix, representing the node degree matrix, and the elements on the diagonal are the degree d(v) of each node v in the hypergraph; D e is a diagonal matrix representing the degree matrix of hyperedges, where the elements on the diagonal are the degree δ(e) of each hyperedge e in the hypergraph; H is dimensional point-edge incidence matrix, ε、 They are respectively the node set, hyperedge set, and weight set assigned to hyperedges of the hypergraph.

4. The method according to claim 3, characterized in that The degree of a node is calculated as follows: Among them, v j ,e j are the jth node and hyperedge in the hypergraph, w(e j ) represents the hyperedge e j The average value of the heat core distance between any two vertices on it; The degree of a hyperedge is calculated as follows:

5. The method according to claim 3, characterized in that In step S2, the optimization of the hypergraph is transformed into finding the eigenvector corresponding to the minimum eigenvalue of the matrix bundle (L, A) to achieve graph segmentation; L is the Laplacian matrix of the hypergraph, A is the adjacency matrix of the hypergraph, The value a of each element in the adjacency matrix i,j =w(e n ), w(e n ) represents the nth hyperedge e n Node v at any two vertices i ,v j The average value of the hot core distance is: x i ,x j and v i ,v j The corresponding time-frequency representation or time domain signal, u is the Gaussian kernel size.

6. The method according to claim 1, characterized in that Before step S4, the method further includes: Repeat S1 to S3, use more vibration data to iteratively train the hypergraph model, expand the number of graph nodes and layers of the fused hypergraph, and construct the hypergraph model into a deep neural network with high fault feature extraction capability.

7. The method according to claim 1, characterized in that The vibration data obtained in step S1 is expressed as follows: in, is a set of time domain signals The corresponding time-frequency representation, c represents the set of all fault classes, the lth fault class is the jth time domain signal and its time-frequency representation vector under the lth fault type, and N represents the number of data.

8. The method according to claim 1, characterized in that The time domain signals generated by various faults during the operation of the rotating mechanical system are collected, including: An acceleration sensor is installed on the cylinder head of the fuel and valve system of the rotating mechanical system, and a reference standard reflective label of the dead point signal measured by the photoelectric speed sensor is affixed to the main shaft end of the rotating mechanical system. Based on the dead point signal, a time domain signal with the same time length is collected.

9. The method according to claim 1, characterized in that A cross entropy function is used as a loss function for calculating fault assessment to implement the training, optimization, and updating of the hypergraph convolutional fuzzy network.

Images