Unmanned aerial vehicle cluster cross-domain fault diagnosis method based on multi-source domain depth migration network

Abstract

The invention discloses an unmanned aerial vehicle cluster cross-domain fault diagnosis method based on a multi-source domain depth migration network, and the method comprises the steps: firstly, providing a space-time Bayesian graph wavelet convolution Transformer in combination with an unmanned aerial vehicle cluster communication topology in the aspect of unmanned aerial vehicle cluster fault feature extraction; space-time coupled fault features are extracted from existing flight state monitoring data under different operation conditions; taking the existing different operation conditions as different source domains, and taking the new operation condition as a target domain; then, constructing a multi-source domain cross convolution Transform, and learning a mutual relation between fault features of different source domains in a segment-to-segment mode to obtain a multi-source domain fusion feature; and finally, providing a weight-shared three-branch cross-domain alignment convolution Transformer to perform feature alignment on the multi-source domain fusion features and target domain fault features in a segment-to-segment manner, extracting domain invariant features, and improving fault diagnosis performance under a new operation condition by using the source domain fault features. According to the method, the problem of fault diagnosis of the unmanned aerial vehicle cluster in the target domain based on limited flight data is solved, and the accuracy and generalization of cross-domain fault diagnosis are improved.

Description

Technical Field

[0001] The present invention belongs to the field of UAV cluster fault diagnosis, and specifically relates to a UAV cluster cross-domain fault diagnosis method based on a multi-source domain deep migration network. Background Art

[0002] Compared with a single drone, the drone swarm system, which is a collaborative system of multiple drones, has obvious advantages in terms of task completion time, fault tolerance, working range, etc., which makes it widely used in real-life scenarios such as small-scale precision agriculture, collaborative transportation, and commercial performances. However, due to the interconnectivity between drones in the drone swarm system, when one or more drones fail, these failures will not only affect their own normal flight, but also reduce the performance of neighboring communication drones, thereby adversely affecting the entire drone swarm system and even causing the risk of the entire system crashing. Therefore, it is of great practical significance to study drone swarm fault diagnosis technology, timely discover the failure of the drone swarm system, facilitate timely maintenance decisions, and ensure the reliable operation of swarm flight.

[0003] With the continuous advancement of industrial-related data transmission and computing technologies, a large number of flight status monitoring signals of drone cluster systems are collected, which has led to the widespread development of data-driven fault diagnosis methods. Data-driven fault diagnosis models can learn potential fault features from a large number of flight status monitoring signals, and establish a nonlinear mapping relationship between features and fault classification labels to achieve classification prediction of faults. Data-driven fault diagnosis methods generally include two types of shallow machine learning methods and deep learning methods. Shallow machine learning methods include support vector machines, relevance vector machines, and hidden Markov models. However, these methods can only learn shallow features, have limited fault diagnosis performance, and require complex manual feature engineering, which is very time-consuming and labor-intensive. With the rapid development of deep learning technologies with powerful deep representation learning capabilities, such as recurrent neural networks (RNNs), convolutional neural networks (CNNs) and their various variants such as long short-term memory (LSTM) and gated recurrent units (GRUs), deep learning models have become more effective fault diagnosis technologies. Deep learning models can perform end-to-end fault diagnosis directly from raw flight status monitoring data without any manual feature engineering. These deep learning methods have achieved good diagnostic performance in fault diagnosis. However, due to structural limitations such as the gradient vanishing and explosion problems of recurrent neural networks, and the limited receptive field of convolutional neural networks, these CNN, RNN and their variant models are not yet fully capable of modeling the global dependencies of long-period flight status monitoring data.

[0004] In actual engineering, drone swarms need to perform different tasks. According to different task requirements, drone swarms need to fly under different operating conditions (i.e., different domains), such as different communication topologies, formation shapes, and flight speeds. Under new operating conditions, the fault data of drone swarms is very limited, which makes it impossible to fully train the data-driven model, resulting in insufficient learning of fault features, large model uncertainty, poor generalization performance, and low fault diagnosis accuracy. However, a large number of flights under new operating conditions to obtain sufficient fault data require high flight costs. Considering that drone swarms accumulate a large amount of historical fault data under completed tasks, and when drone swarms have the same fault under different tasks, there are similar fault features. Therefore, it is of engineering significance to study how to use existing fault data to solve the fault diagnosis problem based on a small amount of limited data under new operating conditions. However, due to the differences in communication topology, flight formation shape, and speed of drone swarms under different operating conditions, there are differences in data distribution of fault features, which leads to the failure of fault diagnosis algorithms. Transfer learning can handle the problem of data distribution differences in different domains. Therefore, it is of engineering significance to study a transfer learning method to solve the fault diagnosis problem of drone swarms based on limited data under new operating conditions. Summary of the invention

[0005] Purpose of the invention: The present invention provides a cross-domain fault diagnosis method for UAV clusters based on a multi-source domain deep migration network, which overcomes the problems of difficulty in extracting fault features of UAV clusters, difficulty in fusing multi-source domain features, and difficulty in migrating cross-domain features. It solves the fault diagnosis problem of UAV clusters based on limited flight data under new operating conditions, and improves the accuracy and generalization of cross-domain fault diagnosis.

[0006] Technical solution: The cross-domain fault diagnosis method of a drone cluster based on a multi-source domain deep migration network described in the present invention comprises the following steps:

[0007] (1) Collect flight status monitoring data corresponding to different faults occurring under different operating conditions of the UAV cluster, perform data preprocessing, divide the data into source domain and target domain, and construct training set and test set;

[0008] (2) constructing and initializing a multi-source domain deep migration network model, the model includes a domain feature extractor based on a spatiotemporal Bayesian graph wavelet convolutional transformer, a multi-source domain cross convolutional transformer, a weight-sharing three-branch cross-domain aligned convolutional transformer, and source domain and target domain classifiers;

[0009] (3) Input N source domain samples and one target domain sample in the training set into the domain feature extractor to extract fault features in different domains;

[0010] (4) Input the N source domain fault features into the multi-source domain cross convolutional Transformer, learn the relationship between the fault features of different source domains in a segment-by-segment manner, and obtain the multi-source domain fusion features;

[0011] (5) Inputting the N source domain fault features into the corresponding N source domain classifiers to obtain the fault classification results of the N source domains;

[0012] (6) The multi-source domain fusion features obtained in step (4) and the target domain fault features obtained in step (3) are simultaneously input into the weight-sharing three-branch cross-domain alignment convolutional Transformer, and feature alignment is performed in a segment-by-segment manner to obtain high-level source domain and target domain fault features and cross-domain related features;

[0013] (7) Input the fault features obtained in step (6) into the target domain classifier to obtain the fault classification results corresponding to the three branches;

[0014] (8) Repeat steps (3) to (7) until the maximum number of predictions N is reached. p , get N for each training sample p fault classification results in different domains, and calculate the average value of the fault classification results in different domains as the final classification result;

[0015] (9) Calculate the loss function L that integrates multiple factors total , including the hierarchical multi-metric domain difference loss L dis , classification loss and distillation loss L dtl , the back-propagation algorithm is used to update the network parameters to minimize the overall loss value;

[0016] (10) Repeat steps (2) to (9) to traverse all training samples until the maximum number of training rounds M is reached. epoch , we obtain the optimal multi-source domain deep migration network model;

[0017] (11) Activate the network in the target domain data flow direction of the optimal multi-source domain deep migration network model, input the test set samples into the optimal multi-source domain deep migration network model, and calculate the corresponding fault classification results;

[0018] (12) Repeat step (11) until the maximum number of predictions N is reached p , get N for each test sample p fault classification results and calculate the average value of the fault classification results as the final classification result. p The prediction variance V corresponding to the fault classification result p , quantifying the uncertainty of classification predictions.

[0019] Furthermore, the different operating conditions of the drone cluster in step (1) include the flight of the drone cluster in different flight formation shapes and different flight speeds under different communication topologies; each operating condition is regarded as a domain; flight status monitoring data under N existing operating conditions are collected as N source domain data sets, and limited flight status monitoring data under a new operating condition is collected as a target domain data set.

[0020] Furthermore, the different faults described in step (1) include sensor and actuator failures of a single drone, sensor and actuator failures of two or more drones at the same time, communication interruption failures between two drones, and propagation failures of non-communicating drones.

[0021] Furthermore, the flight status monitoring data in step (1) includes the absolute value of the relative distance change between the current UAV and its neighboring communicating UAV in the lateral and longitudinal directions, the speed deviation and the average value of the overall speed, the lateral and longitudinal speeds, climb angle, yaw angle, bank angle and throttle opening of the current UAV.

[0022] Furthermore, the implementation process of step (2) is as follows:

[0023] Given N source domain input samples and a target domain input sample First, the samples of the source domain and the target domain are respectively extracted through the domain feature extractor to extract the fault features in each domain. Afterwards, the N source domain fault features are obtained Two operations are performed: first, the fault features of each domain are input into the respective source domain classifiers to obtain the classification prediction results, i.e. j=1,...,N; secondly, divide the N source domain fault features into N pairs of feature combinations in the ratio of 1:N-1, namely These combinations are then input into the multi-source domain cross convolution Transformer F mcct (·) to obtain multi-source domain fusion features Next, the multi-source domain fusion features and target domain fault characteristics Input into the weight-sharing three-branch cross-domain alignment convolutional Transformer to perform feature alignment in a segment-by-segment manner to obtain high-level source domain fault features and target domain fault characteristics And cross-domain related features Finally, the obtained features are input into the target domain classifier Obtain the target domain fault classification results.

[0024] Furthermore, the source domain and target domain classifiers in step (2) have the same network structure, including two Bayesian linear layers, ReLU activation function and Softmax function layer; the source domain fault feature is established by the classifier j=1,...,N, high-level source domain fault characteristics and target domain fault characteristics And cross-domain related features The nonlinear mapping relationship between the fault classification label and the

[0025] Furthermore, the domain feature extractor in step (3) includes a common feature extractor F co (·) and domain-specific feature extractors The common feature extractor is composed of a weight-sharing spatiotemporal Bayesian graph wavelet convolutional Transformer; the specific domain feature extractors are composed of independent spatiotemporal Bayesian graph wavelet convolutional Transformers;

[0026] The spatiotemporal Bayesian graph wavelet convolution Transformer includes multi-head spatiotemporal convolution attention, Bayesian linear layer, residual connection and layer normalization; the multi-head spatiotemporal convolution attention includes H spatiotemporal convolution attentions arranged in parallel, each spatiotemporal convolution attention extracts spatiotemporal coupling fault features from different scales; the input sample passes through H spatiotemporal convolution attentions to obtain H spatiotemporal coupling fault features and splices them, and then passes through two layers of Bayesian linear layers with residual connection and layer normalization in sequence to obtain the final spatiotemporal coupling fault features;

[0027] The spatiotemporal convolutional attention includes multi-layer multi-scale adaptive graph wavelet convolution, local and global scaling dot product attention, dilated causal convolution network layer and dimension adjustment layer; first, multi-layer multi-scale adaptive graph wavelet convolution processes input samples Get the corresponding query vector Q h , the key vector K h Sum value vector V h , and input into the local scaled dot product attention to obtain the spatial feature sequence Then the dimension adjustment layer adjusts the dimension of the spatial feature sequence to obtain where d model =Ne d ×N f ; Then, two parallel dilated causal convolutional network layers learn the spatial feature sequence The local context dependency of the corresponding query vector Q h ′, key vector K h ′; Similarly, the dimension adjustment layer adjusts the input samples Dimensions and take it as the value vector Vh ′; The query vector Q h ′, key vector K h ′ and value vector V h ′Input into the global scaled dot product attention to adaptively integrate the local temporal features of the input samples into the modeling of global dependencies to simultaneously model local and global spatiotemporal coupling features and extract spatiotemporal coupling fault features;

[0028] The multi-layer multi-scale adaptive graph wavelet convolution is composed of K layers of multi-scale adaptive graph wavelet convolution in series; each multi-scale adaptive graph wavelet convolution includes |S| scale adaptive graph wavelet convolutions to extract spatial features in input samples from different scales; each adaptive graph wavelet convolution is implemented by graph wavelet transform; the graph wavelet transform uses a set of suitable wavelet bases provided by wavelet transform to project the graph signal from the vertex domain to the spectral domain and extract fault features in the spectral domain; the wavelet base is defined as Where Ne is the number of drones, each wavelet basis ψ s,i Corresponding to the signal of the i-th UAV at scale s; the graph wavelet with scale parameter s on the i-th UAV can be written in the following matrix multiplication form:

[0029] Ψ s =UG s U T

[0030] in, is the scaling matrix, λ l is the normalized graph Laplacian matrix The lth eigenvalue of , U is The eigenvector matrix of ; the normalized graph Laplacian matrix It is obtained from the adjacency matrix A corresponding to the UAV cluster communication topology, which is defined as Among them I n is the identity matrix of the corresponding dimension, is the angle matrix The graph wavelet basis is approximated by Chebyshev polynomials as follows:

[0031]

[0032] c i,s =2e -s J i (-s)

[0033] in, is the i-th order Chebyshev polynomial, Q is the number of Chebyshev polynomials, J i(-s) is the first kind of Bessel function; for input samples with graph structure Wavelet transform of the graph is The corresponding inverse transformation is The adaptive graph wavelet convolution with scale s is defined as:

[0034]

[0035] Among them, s is a graph wavelet basis with scale s, W s (i,j) are learnable parameters, is the output signal, σ(·) is the nonlinear activation function; the adaptive graph wavelet convolutions with different scales are connected in parallel and combined with residual connections as an adaptive component to form a multi-scale adaptive graph wavelet convolution, and then multiple multi-scale adaptive graph wavelet convolutions are connected in series to form a multi-layer multi-scale adaptive graph wavelet convolution; in the domain d∈{d s1 ,...,d sN ,d t The corresponding output of the s-th scale of the K-th layer adaptive graph wavelet convolution in} is:

[0036]

[0037] in, is the initial input of the network, and are the graph wavelet basis and its inverse at scale s, Θ d,s is the diagonal matrix of the graph convolution kernel, and is a trainable parameter matrix, σ(·) is a nonlinear activation function; in the domain d∈{d s1 ,...,d sN ,d t}, the final output of the multi-layer multi-scale adaptive graph wavelet convolution is the average value of the output features of |S| parallel adaptive graph wavelet convolutions in the K-th layer of multi-scale adaptive graph wavelet convolution, which is defined as follows:

[0038]

[0039] Furthermore, the multi-source domain cross convolution Transformer in step (4) includes N multi-source cross convolution attention layers, residual connections, layer normalization, and linear layers; which are used to learn the relationship between different source domain features and extract multi-source domain fusion features;

[0040] The multi-source cross convolution attention includes an expanded causal convolutional network layer, a repeated layer, and a scaled dot product attention; each N source domain fault feature is sequentially regarded as a value vector, and the remaining N-1 source domain fault features are regarded as key vectors and value vectors; specifically, the N source domain fault features are sequentially divided into N pairs of feature combinations according to a ratio of 1:N-1, namely Input N pairs of feature combinations into N multi-source cross convolutional attentions respectively, It corresponds to The splicing features of the fault features of the remaining source domains except ; After the expansion causal convolutional network layer, the value vector V is obtained cn , After the expansion causal convolutional network layer, the query vector Q is obtained cn , key vector K cn , the repeated layer passes through the value vector V cn The copy operation is performed to make its dimension the same as the corresponding query vector Q cn , key vector K cn Consistent; the query vector Q cn , key vector K cn Sum value vector V cn Input the scaled dot product attention, calculate the attention, learn the relationship between each source domain feature and the other source domain features in a segment-by-segment manner, and obtain N multi-source domain interaction features {Attn c1 ,...,Attn cN}; Further, the N multi-source domain interaction features are concatenated, and then further fused through two layers of linear layers with residual connection and layer normalization to obtain the final multi-source domain fusion feature

[0041] Furthermore, the weight-shared three-branch cross-domain aligned convolutional Transformer in step (6) includes a source branch, a source-target branch, and a target branch; the multi-source domain fusion features and the target domain fault features are aligned in a segment-by-segment manner to extract domain-invariant features, thereby utilizing the source domain fault features to improve the fault diagnosis performance in the target domain; the source branch, the source-target branch, and the target branch are all composed of a weight-shared convolutional Transformer; the multi-source domain fusion features are aligned with the target domain fault features ... Input to the source branch, and processed by the dilated causal convolutional network layer to obtain the query vector Q s , key vector K s Sum value vector V s , and then through multi-head self-attention, two layers of residual connection, layer normalization and linear layer to obtain high-level source domain fault features Similarly, the target domain fault characteristics Input into the target branch, and processed by the dilated causal convolutional network layer to obtain the query vector Qt , key vector K t Sum value vector V t , and then through multi-head self-attention, two layers of residual connection, layer normalization and linear layer to obtain high-level target domain fault features The query vector Q in the source branch s and the key vector K of the target branch t Sum value vector V t At the same time, it is input into the multi-head cross-domain attention of the source-target branch to learn the relationship between the source domain and target domain features, and obtain cross-domain related features; then the multi-source domain fusion features are and high-level features of the target domain Splicing to get the cross-domain combined feature F cd , and input it into the residual connection and layer normalization in the source-target branch to fuse with the cross-domain alignment features, and then pass through the linear layer, residual connection and layer normalization to obtain the cross-domain related features Next, the cross-domain related features obtained by the three branches are and They are respectively input into the weight-shared target domain classifier and the corresponding fault classification value is output.

[0042] Furthermore, the hierarchical multi-metric domain difference loss L in step (9) is dis Used to optimize the distribution difference of fault features at different levels of each pair of target domains and source domains; it includes the low-level domain distribution loss and the high-level domain distribution loss obtained by weighted addition of the maximum average difference loss and the CORAL distance, which are used to measure the distribution difference between the fault features of the source domain samples after passing through the common feature extractor and the specific domain feature extractor and the corresponding target domain fault features;

[0043] The classification loss It is used to optimize the difference between all classifier output values ​​and fault classification label values; cross entropy loss is selected as the classification loss, and the total classification loss is in represents the classification loss value corresponding to the N source domain classifier, The classification loss value of the source branch and the classification loss value of the target branch corresponding to the target domain classifier;

[0044] The distillation loss L dtl It is used to improve the fault diagnosis performance in the target domain by using the fault features in the source domain. Specifically, the cross-domain related features of the source-target branch output in the weight-shared three-branch cross-domain alignment convolutional Transformer are The fault classification probability distribution output in the target domain classifier is used as a soft label to supervise the training of the target branch through the following distillation loss to improve the classification performance of the diagnosis model on the target domain samples.

[0045]

[0046] Among them, q k and p k are the probabilities of the classification category k corresponding to the output of the source-target branch and the target branch respectively;

[0047] The comprehensive multi-factor loss function L total for,

[0048]

[0049] Among them, λ and β are weight parameters.

[0050] Beneficial effects: Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. In terms of fault feature extraction, the spatiotemporal Bayesian graph wavelet convolution Transformer proposed in the present invention can integrate physically meaningful wavelet transform into graph convolutional neural network to design multi-scale adaptive graph wavelet convolution to extract multi-scale spatial fault features; then the local temporal features of the input samples are adaptively integrated into the modeling of global dependency relationships by global scaling dot product self-attention to simultaneously model local and global temporal features, thereby realizing the extraction of spatiotemporal coupling features, which has high generalization and interpretability;

[0052] 2. In terms of multi-source domain feature fusion, the present invention proposes a multi-source domain cross convolution Transformer that divides N fault features into N pairs of feature combinations in a ratio of 1:N-1 and inputs them into N multi-source cross convolution attentions. The relationship between different source domain features is learned in a segment-by-segment manner, which can fully fuse multi-source domain features and extract more valuable multi-source domain fusion features.

[0053] 3. In terms of cross-domain feature alignment, a weight-sharing three-branch cross-domain alignment convolutional Transformer is proposed. Through three weight-sharing source branches, source-target branches, and target branches, the multi-source domain fusion features and the target domain fault features are aligned in a segment-by-segment manner, and domain-invariant features are extracted, so as to improve the fault diagnosis performance in the target domain by using the source domain features.

[0054] 4. In terms of loss function optimization, the present invention proposes a comprehensive multi-factor loss function including hierarchical multi-metric domain difference loss, classification loss and distillation loss, which can optimize feature distribution differences and classification losses in different domains from multiple angles and improve fault diagnosis performance in the target domain;

[0055] In summary, the present invention overcomes the problems of difficulty in extracting fault features of drone clusters, difficulty in fusing multi-source domain features, and difficulty in migrating cross-domain features. It solves the problem of fault diagnosis of drone clusters based on limited flight data under new operating conditions, and improves the accuracy and generalization of cross-domain fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a flow chart of the present invention;

[0057] Figure 2 This is a schematic diagram of the multi-source domain deep migration network structure;

[0058] Figure 3 Schematic diagram of the spatiotemporal Bayesian graph wavelet convolution Transformer structure;

[0059] Figure 4 Schematic diagram of the multi-source domain cross convolution Transformer structure;

[0060] Figure 5 Four different operating conditions of drone swarms: (a) is a square flight formation and communication topology; (b) is a V-shaped flight formation and communication topology; (c) is a pentagonal flight formation and communication topology; (d) is a cross-shaped flight formation and communication topology;

[0061] Figure 6 The fault diagnosis results of the UAV cluster based on the method of the present invention under four cross-domain diagnosis tasks. DETAILED DESCRIPTION

[0062] The present invention is further described in detail below with reference to the accompanying drawings.

[0063] like Figure 1 As shown, the present invention provides a cross-domain fault diagnosis method for drone clusters based on a multi-source domain deep migration network, and the specific implementation process is as follows:

[0064] Step 1: Collect the flight status monitoring data corresponding to different faults occurring under different operating conditions (different domains) of the drone cluster, perform data preprocessing, divide the data into source domain and target domain, and construct training set and test set.

[0065] Different operating conditions of drone clusters include the flight of drone clusters in different flight formation shapes and at different flight speeds under different communication topologies; each operating condition is regarded as a domain; the flight status monitoring data under N operating conditions are collected as N source domain data sets, and the flight status monitoring data under one operating condition is taken as a target domain data set.

[0066] Different failures that can occur in drone swarms include sensor and actuator failures of a single drone, simultaneous sensor and actuator failures of two or more drones, communication interruption failures between two drones, and propagation failures of non-communicating drones.

[0067] The flight status monitoring data includes the absolute value of the relative distance change between the current UAV and its neighboring communication UAV in the lateral and longitudinal directions, the speed deviation and the average value of the overall speed, the lateral and longitudinal speed of the current UAV, the climb angle, yaw angle, bank angle and throttle opening.

[0068] The flight status monitoring data is normalized to the interval [0,1] by the minimum-maximum normalization formula. The sliding window method is adopted, with a step size of 1, and a sliding window of length 10 is slid from the beginning to the end of the flight status monitoring data to achieve segmented status monitoring data, and samples are generated in units of sliding windows. Each sample corresponds to a fault classification label value, and finally a sample set that can be used for network model training and testing is generated.

[0069] N labeled source domain samples and some labeled target domain samples are used as training sets, and unused labeled target domain samples are used as test sets.

[0070] Step 2: Figure 2 As shown in the figure, a multi-source domain deep migration network model is constructed and initialized, including a domain feature extractor based on the spatiotemporal Bayesian graph wavelet convolutional Transformer, a multi-source domain cross-convolutional Transformer, a weight-sharing three-branch cross-domain aligned convolutional Transformer, and source domain and target domain classifiers.

[0071] The data flow of the multi-source domain deep migration network model is: given N source domain input samples and a target domain input sample First, the samples of the source domain and the target domain are respectively extracted through the domain feature extractor to extract the fault features in each domain. Afterwards, the N source domain fault features are obtained Two operations are performed: first, the fault features of each domain are input into the respective source domain classifiers to obtain the classification prediction results, i.e. j=1,...,N; secondly, divide the N source domain fault features into N pairs of feature combinations in the ratio of 1:N-1, namely These combinations are then input into the multi-source domain cross convolution Transformer F mcct (·) to obtain multi-source domain fusion features Next, the multi-source domain fusion features and target domain fault characteristics Input into the weight-sharing three-branch cross-domain alignment convolutional Transformer to perform feature alignment in a segment-by-segment manner to obtain high-level source domain fault features and target domain fault characteristics And cross-domain related features Finally, the obtained features are input into the target domain classifier Obtain the target domain fault classification results.

[0072] Step 3: Input N source domain samples and one target domain sample in the training set into the domain feature extractor to extract fault features in different domains.

[0073] The domain feature extractor based on spatiotemporal Bayesian graph wavelet convolutional Transformer includes a common feature extractor F co (·) and domain-specific feature extractors The common feature extractor consists of a weight-shared spatiotemporal Bayesian graph wavelet convolutional Transformer, and the specific domain feature extractors consist of independent spatiotemporal Bayesian graph wavelet convolutional Transformers.

[0074] like Figure 3 As shown in the figure, the spatiotemporal Bayesian graph wavelet convolution transformer includes multi-head spatiotemporal convolution attention, Bayesian linear layer, residual connection and layer normalization. The multi-head spatiotemporal convolution attention includes H spatiotemporal convolution attentions arranged in parallel; the working principle of the spatiotemporal Bayesian graph wavelet convolution transformer is: the input sample passes through H spatiotemporal convolution attentions to obtain H spatiotemporal coupling fault features and splice them, and then passes through two layers of Bayesian linear layers with residual connection and layer normalization in sequence to obtain the final spatiotemporal coupling fault features.

[0075] like Figure 3 As shown in the figure, multi-head spatiotemporal convolutional attention is a key core module for extracting spatiotemporal coupling features of drone cluster systems. It consists of H spatiotemporal convolutional attentions arranged in parallel, each of which extracts spatiotemporal coupling fault features from different scales. Each spatiotemporal convolutional attention includes multi-layer multi-scale adaptive graph wavelet convolution, local and global scaling dot product attention, dilated causal convolution network layer and dimension adjustment layer. Its working principle is: First, multi-layer multi-scale adaptive graph wavelet convolution extracts input samples The spatial features of different scales at each time point in the query vector Q are obtained by aggregating them. h , key vector K h Sum value vector V h , and then input into the local scaled dot product attention to establish the local dependency of spatial features to obtain the spatial feature sequence Then adjust the dimension of the spatial feature sequence through the dimension adjustment layer to where d model =Ne d ×N f ; Then, two parallel dilated causal convolutional network layers process the spatial feature sequence Extract local temporal features and output them as query vector Q h ′ and key vector K h ′ Input to the global scaled dot product self-attention; input sample at the same time After the dimension adjustment layer, the dimension is adjusted to and as the value vector V h ′Input the global scaled dot product self-attention; finally, the global scaled dot product self-attention adaptively integrates the local temporal features of the input samples into the modeling of the global dependency to simultaneously model the local and global spatiotemporal coupling features, thereby realizing the extraction of spatiotemporal coupling features, which can effectively improve the generalization of feature learning.

[0076] The multi-layer multi-scale adaptive graph wavelet convolution includes K layers of multi-scale adaptive graph wavelet convolution in series; each multi-scale adaptive graph wavelet convolution includes |S| scale adaptive graph wavelet convolutions to extract spatial features from input samples at different scales; each adaptive graph wavelet convolution is implemented by graph wavelet transform. Graph wavelet transform uses a set of suitable wavelet bases provided by wavelet transform to project the graph signal from the vertex domain to the spectral domain and extract fault features in the spectral domain; the wavelet base is defined as Where Ne is the number of drones, each wavelet basis ψ s,i Corresponding to the signal of the i-th UAV at scale s; the graph wavelet with scale parameter s on the i-th UAV can be written in the following matrix multiplication form:

[0077] Ψ s =UG s U T

[0078] in, is the scaling matrix, λ l is the normalized graph Laplacian matrix The lth eigenvalue of , U is The eigenvector matrix of . Normalized graph Laplacian matrix It is obtained from the adjacency matrix A corresponding to the UAV cluster communication topology, which is defined as Among them I n is the identity matrix of the corresponding dimension, is the angle matrix The graph wavelet basis is approximated by Chebyshev polynomials as follows:

[0079]

[0080] c i,s =2e -s J i (-s)

[0081] in, is the i-th order Chebyshev polynomial, Q is the number of Chebyshev polynomials, J i (-s) is the first kind of Bessel function; for input samples with graph structure Wavelet transform of the graph is The corresponding inverse transformation is The adaptive graph wavelet convolution with scale s is defined as:

[0082]

[0083] Among them, s is a graph wavelet basis with scale s, W s (i,j) are learnable parameters, is the output signal, σ(·) is the nonlinear activation function. The adaptive graph wavelet convolutions with different scales are connected in parallel and combined with residual connections as an adaptive component to form a multi-scale adaptive graph wavelet convolution. Then, multiple multi-scale adaptive graph wavelet convolutions are connected in series to form a multi-layer multi-scale adaptive graph wavelet convolution. In the domain d∈{d s1 ,...,d sN ,d t The corresponding output of the s-th scale of the K-th layer adaptive graph wavelet convolution in} is:

[0084]

[0085] in, is the initial input of the network, and are the graph wavelet basis and its inverse at scale s, Θ d,s is the diagonal matrix of the graph convolution kernel, and is a trainable parameter matrix, σ(·) is a nonlinear activation function. s1 ,...,d sN ,d t}, the final output of the multi-layer multi-scale adaptive graph wavelet convolution is the average value of the output features of |S| parallel adaptive graph wavelet convolutions in the K-th layer of multi-scale adaptive graph wavelet convolution, which is defined as follows:

[0086]

[0087] In summary, the entire processing process of the spatiotemporal Bayesian graph wavelet convolution Transformer can be described by the formula:

[0088] F d =LayerNorm(MCdropout(ReLU(F out1 W h2 +b h2 ))+F out1 )

[0089]

[0090] Fea st =Concat(Attn 1 ,...,Attn H )

[0091]

[0092] in, Represents the output of multi-layer multi-scale adaptive graph wavelet convolution in the h-th spatiotemporal convolutional attention, and the weight of the dilated causal convolution layer and and bias b mf , Attn h represents the h-th spatiotemporal convolutional attention output, W h1 and W h2 is the weight of the Bayesian linear layer, b h1 and b h2 is the bias of the Bayesian linear layer, MCdropout(·) is the random dropout function, and F d is the output feature.

[0093] Step 4: Input N source domain fault features into Figure 4 The multi-source domain cross-convolution Transformer shown learns the relationship between fault features of different source domains in a segment-by-segment manner to obtain multi-source domain fusion features.

[0094] The multi-source cross-convolution Transformer includes N multi-source cross-convolution attentions, residual connections, layer normalization, and linear layers; it is used to learn the relationship between different source domain features and extract multi-source domain fusion features. The multi-source cross-convolution attention includes an expanded causal convolutional network layer, a repeated layer, and a scaled dot product attention; each of the N source domain fault features is used as a value vector, and the remaining N-1 source domain fault features are used as key vectors and value vectors. Specifically, the N source domain fault features are divided into N pairs of feature combinations in a ratio of 1:N-1, namely Input N pairs of feature combinations into N multi-source cross convolutional attentions respectively, It corresponds to The splicing features of the fault features of the remaining source domains except ; After the expansion causal convolutional network layer, the value vector V is obtained cn , After the expansion causal convolutional network layer, the query vector Q is obtained cn , key vector K cn , the repeated layer passes through the value vector V cn The copy operation is performed to make its dimension the same as the corresponding query vector Q cn , key vector K cn Consistent; the query vector Q cn , key vector K cn Sum value vector V cn Input the scaled dot product attention, calculate the attention, learn the relationship between each source domain feature and the other source domain features in a segment-by-segment manner, and obtain N multi-source domain interaction features {Attn c1 ,...,Attn cN Furthermore, the N multi-source domain interaction features are concatenated and then further fused through two layers of linear layers with residual connections and layer normalization to obtain the final multi-source domain fusion feature. The entire processing process of multi-source domain cross convolution Transformer can be described as:

[0095]

[0096] Fea ca =Concat(Attn c1 ,...,Attn cN )

[0097]

[0098] Among them, Attn cn represents the n-th multi-source domain interaction feature, ★ Cross-correlation calculation, dilating the weights of the causal convolutional network layer and and bias b cn , is the fault feature of the nth source domain, For the corresponding The concatenated features of the fault features of the remaining source domains except , Repeat(·) is the repeated layer, It is the final multi-source domain fusion feature.

[0099] Step 5: Input the N source domain fault features into the corresponding N source domain classifiers to obtain the fault classification results of the N source domains.

[0100] The source domain and target domain classifiers have the same network structure, including two Bayesian linear layers, ReLU activation function and Softmax function layer; the source domain fault features are established through the classifier j=1,...,N, high-level source domain fault characteristics and target domain fault characteristics And cross-domain related features The nonlinear mapping relationship between the source domain classifier and the fault classification label can be described as follows:

[0101]

[0102] in, is the corresponding source domain classifier, is the corresponding fault classification result, d s ∈{d s1 ,...,d sN}.

[0103] Step 6: Input the multi-source domain fusion features obtained in step 4 and the target domain fault features obtained in step 3 into the weight-sharing three-branch cross-domain alignment convolutional Transformer at the same time, perform feature alignment in a segment-by-segment manner, and obtain high-level source domain and target domain fault features as well as cross-domain related features.

[0104] like Figure 2 As shown in the figure, the weight-sharing three-branch cross-domain aligned convolutional Transformer includes a source branch, a source-target branch, and a target branch. The multi-source domain fusion features are aligned with the target domain fault features in a segment-by-segment manner to extract domain-invariant features, so as to improve the fault diagnosis performance in the target domain by using the source domain fault features; the source branch, the source-target branch, and the target branch are all composed of a weight-sharing convolutional Transformer. Multi-source domain fusion features Input to the source branch, and processed by the dilated causal convolutional network layer to obtain the query vector Q s , key vector K s Sum value vector V s , and then through multi-head self-attention, two layers of residual connection, layer normalization and linear layer to obtain high-level source domain fault features Similarly, the target domain fault characteristics Input into the target branch, and processed by the dilated causal convolutional network layer to obtain the query vector Q t , key vector K t Sum value vector V t , and then through multi-head self-attention, two layers of residual connection, layer normalization and linear layer to obtain high-level target domain fault features The query vector Q in the source branch sand the key vector K of the target branch t Sum value vector V t At the same time, it is input into the multi-head cross-domain attention of the source-target branch to learn the relationship between the source domain and target domain features, and obtain cross-domain related features; the multi-head cross-domain attention is realized by multi-head self-attention; then the multi-source domain fusion features are and high-level features of the target domain Splicing to get the cross-domain combined feature F cd , and input it into the residual connection and layer normalization in the source-target branch to fuse with the cross-domain alignment features, and then pass through the linear layer, residual connection and layer normalization to obtain the cross-domain related features Next, the cross-domain related features obtained by the three branches are and They are input into the weight-shared target domain classifier respectively, and the corresponding fault classification value is output. The processing process of different branches of the weight-shared three-branch cross-domain aligned convolutional Transformer can be described as follows. The source branch is:

[0105]

[0106] The target branch is:

[0107]

[0108] The source-target branches are:

[0109]

[0110] Among them, W s q , W s k , and W s v is the weight of the dilated causal convolutional network layer, b s is the bias of the dilated causal convolutional network layer, H m is the number of attentions in multi-head self-attention, W s1 and W s2 is the weight of the linear layer, b s1 and b s2 is the bias of the linear layer.

[0111] Step 7: Input the fault features obtained in step 6 into the target domain classifier to obtain the fault classification results corresponding to the three branches.

[0112] like Figure 2 As shown in Figure 1, the target domain classifier has the same structure as the source domain classifier mentioned above. A target domain classifier can be described as follows:

[0113]

[0114] in, is the corresponding target domain classifier, is the corresponding fault classification result, d t ∈{d ts ,d tst ,d tt}.

[0115] Step 8: Repeat steps 3 to 7 until the maximum number of predictions N is reached. p , get N for each training sample p The fault classification results in different domains are calculated and the average value of the fault classification results in different domains is taken as the final classification result.

[0116] Step 9: Calculate the loss function L based on multiple factors total Including hierarchical multi-metric domain difference loss L dis , classification loss and distillation loss L dtl , the back-propagation algorithm is used to update the network parameters to minimize the overall loss value.

[0117] Hierarchical multi-metric domain difference loss L dis It is used to optimize the distribution difference of different levels of features of each pair of target domain and source domain, and help the network learn domain invariant features. It consists of two parts: low-level domain distribution loss and high-level domain distribution loss based on maximum average difference loss and CORAL distance; they are used to reduce the distribution distance between the features of source domain samples after passing through the common feature extractor and the specific domain feature extractor and the corresponding target domain features; the maximum average difference and CORAL distance metric functions are defined as follows. Given two domains and The maximum mean difference can be obtained by the following unbiased estimate:

[0118]

[0119] Where H is the reproducing kernel Hilbert space with characteristic kernel K, and φ(·) represents the mapping function that can map the original x to the reproducing kernel Hilbert space. The CORAL distance is the distance between the covariances of the data in the two domains and is defined as follows:

[0120]

[0121] in, represents the Frobenius norm of the square matrix, C s and C s Represent the covariance matrix of the source domain and target domain features respectively. Based on the above definition of the maximum mean difference and CORAL distance, taking the jth source domain as an example, the low- and high-level domain distribution loss and high-level domain distribution loss and hierarchical multi-metric domain difference loss L dis It can be defined as follows:

[0122]

[0123] Among them, ε is the weight parameter, F MMD (·) represents the maximum average difference loss, F CORAL (·) is the CORAL distance, F co (·) is the common feature extractor, is a domain-specific feature extractor. Therefore, by minimizing minL dis The parameters of the common feature extractor and the domain-specific feature extractor are optimized to narrow the data distribution difference between the source domain features and the target domain features.

[0124] Classification Loss d∈{d s1 ,...,d sN ,d tt ,d ts}Using the cross entropy loss defined as follows, for a given data set

[0125]

[0126] in, is the true label value, is the output value of the classifier. The total classification loss is in represents the classification loss value corresponding to the N source domain classifier, are the source branch classification loss value and the target branch classification loss value corresponding to the target domain classifier; the classification loss is used to optimize the difference between all classifier output values ​​and the fault classification label value.

[0127] Distillation loss It is used to use source domain features to assist fault diagnosis in the target domain to improve the target domain fault diagnosis performance under small samples; specifically, the cross-domain related features of the source-target branch output in the three-branch cross-domain alignment convolutional Transformer are The fault classification probability distribution output in the target domain classifier is used as a soft label to supervise the training of the target branch through the distillation loss defined as follows to improve the classification performance of the model on the target domain samples:

[0128]

[0129] Among them, q k and p kare the probabilities of classification category k corresponding to the output from the source-target branch and the target branch, respectively.

[0130] Therefore, the loss function L that integrates multiple factors total It can be defined as:

[0131]

[0132] Among them, λ and β are weight parameters.

[0133] Step 10: Repeat steps 2 to 9 to traverse all training samples until the maximum number of training rounds M is reached. epoch , and obtain the optimal multi-source domain deep migration network model.

[0134] Step 11: activate the network in the target domain data flow direction of the optimal multi-source domain deep migration network model, input the test set samples into the optimal multi-source domain deep migration network model, and calculate the corresponding fault classification results.

[0135] Step 12: Repeat step 11 until the maximum number of predictions N is reached. p , get N for each test sample p fault classification results and calculate the average value of the fault classification results as the final classification result. p The prediction variance V corresponding to the fault classification result p , quantifying the uncertainty of classification predictions.

[0136] Prediction Variance p Defined as,

[0137]

[0138] Among them, σ 2 YesN p The variance of the classification prediction results, Softplus(·) is the softplus activation function used to ensure the effectiveness of the variance. Higher prediction variance V p Indicates N p The random prediction results are more discrete and uncertain. On the contrary, the lower the prediction variance V p Indicates N p The random prediction results are more stable, with less uncertainty, and the prediction results are more reliable.

[0139] Based on the semi-physical simulation platform of the drone cluster system, the corresponding flight status monitoring data when different faults occur in the drone cluster under different operating conditions are collected. The semi-physical simulation platform of the drone cluster system is mainly composed of a Raspberry Pi cluster, a server, a switch, a router and a data analysis terminal (PC). Each pair of Raspberry Pis in the Raspberry Pi cluster simulates a drone, one of which is used as a flight controller system, and the other is used to simulate the dynamics, actuators, sensors and environment of the drone. The Raspberry Pi sends the information to the router through the switch, and finally sends it to the database server and the neighbor computing server for information exchange. The database server is connected to the router and is used to store the status information of the drone cluster. The neighbor computing server accesses the database, solves the neighbor drone information of each drone according to the communication topology structure and sends the data to the flight controller. The flight controller mainly solves and generates control instructions based on the received sensor data and the neighbor drone information, and sends them to the dynamic environment part to control the drone flight cluster. The data analysis terminal (computer) obtains the status information of the drone cluster stored in the database and displays the flight status of the cluster drone through the situation display terminal software.

[0140] N = 4 operating conditions were selected, namely square, V-shaped, pentagonal and cross flight formations and the corresponding communication topologies and flight speeds, as shown in Table 1 and Figure 5 As shown, (a) is a square flight formation and communication topology; (b) is a V-shaped flight formation and communication topology; (c) is a pentagonal flight formation and communication topology; (d) is a cross-shaped flight formation and communication topology. Under each operating condition, 12 types of drone cluster faults are set as shown in Table 2, and the flight status monitoring data shown in Table 3 are collected. In order to verify the fault diagnosis performance of the present invention based on limited data under the new operating conditions of the drone cluster, four cross-domain diagnosis tasks of the drone cluster are set, as shown in Table 4.

[0141] Table 1 Description of four flight conditions of UAV swarm

[0142] Working conditions Communication topology Formation shape Flight speed (m / s) Flight time(s) FT1 square square 35 600 FT2 V-Shape V-Shape 25 600 FT3 pentagon pentagon 30 600 FT4 Cross Cross 20 600

[0143] Table 2 UAV cluster failure types

[0144] Fault Tags Fault type Health - 0 Health status Fault-1 Multiplicative failure of single sensor Fault-2 Single sensor additive fault Fault-3 Single-machine sensor time-varying fault Fault-4 Single engine multiplication failure Fault-5 Additive failure of single control surface Fault-6 Combined failure of two machines Fault-7 Combined failure of two machines Fault-8 Single machine propagation failure outside the communication topology Fault-9 Two-machine propagation failure outside the communication topology Fault-10 Three-machine propagation failure outside the communication topology Fault-11 Communication failure between the two machines Fault-12 Communication failure between the two machines

[0145] Table 3 UAV swarm flight status monitoring data

[0146]

[0147] Table 4 Four cross-domain diagnosis tasks of drone clusters

[0148]

[0149] In order to evaluate the fault diagnosis performance of different models, three evaluation indicators, accuracy, recall rate, and F1 score, are selected and defined as follows: Among them, TP, TN, FP and FN are the number of correctly classified positive samples, the number of correctly classified negative samples, the number of incorrectly classified negative samples, and the number of incorrectly classified positive samples, respectively.

[0150] like Figure 6 As shown in Table 5, the fault diagnosis results of the drone cluster based on the method of the present invention under four cross-domain diagnosis tasks. According to the experimental results, the method of the present invention can obtain accurate fault diagnosis results under different cross-domain fault diagnosis tasks, which shows the effectiveness of the method. In addition, in all tasks, the diagnostic uncertainty is kept at a low level without large fluctuations, which verifies the reliability of the fault diagnosis results.

[0151] Table 5 Diagnosis results of drone clusters under four cross-domain fault diagnosis tasks

[0152] Task Accuracy Recall F1 Score Uncertainty A 94.58％ 94.53％ 94.51％ 0.023 B 93.96％ 93.96％ 93.99％ 0.015 C 94.41％ 94.40％ 94.38％ 0.020 D 94.58％ 94.58％ 94.54％ 0.018

[0153] In order to further verify the superiority of the method of the present invention, the existing integrated transfer learning method, multi-source domain adaptive network, and multi-source weighted deep transfer network are selected as comparative analysis methods. Table 6 shows the corresponding fault diagnosis results of different methods under task A.

[0154] Table 6 Diagnosis results of different methods under task A

[0155] method Accuracy Recall F1 score Method of the present invention 94.58％ 94.53％ 94.51％ Ensemble transfer learning methods 89.63％ 89.07％ 89.25％ Multi-source domain adaptation network 87.36％ 87.42％ 87.38％ Multi-source partial migration network 86.78％ 86.57％ 86.74％

[0156] According to the comparative diagnostic results of the method of the present invention and the integrated transfer learning method, the multi-source domain adaptive network and the multi-source partial transfer network in Table 6, it can be seen that the UAV cluster cross-domain fault diagnosis method based on the multi-source domain deep transfer network proposed in the present invention obtains the best values ​​in the three diagnostic performance evaluation indicators, which indicates that this method can provide more accurate and reliable UAV cluster cross-domain fault diagnosis results.

[0157] The method of the present invention can extract more accurate, rich and complete spatiotemporal coupling fault characteristics and uncertainty information from the UAV cluster flight status monitoring data, and has high generalization and robustness.

[0158] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A cross-domain fault diagnosis method for drone clusters based on a multi-source domain deep migration network, characterized in that: The following steps are involved: (1) Collect flight status monitoring data corresponding to different faults occurring under different operating conditions of the UAV cluster, perform data preprocessing, divide the data into source domain and target domain, and construct training set and test set; (2) constructing and initializing a multi-source domain deep migration network model, the model includes a domain feature extractor based on a spatiotemporal Bayesian graph wavelet convolutional transformer, a multi-source domain cross convolutional transformer, a weight-sharing three-branch cross-domain aligned convolutional transformer, and source domain and target domain classifiers; (3) Input N source domain samples and one target domain sample in the training set into the domain feature extractor to extract fault features in different domains; (4) Input the N source domain fault features into the multi-source domain cross convolutional Transformer, learn the relationship between the fault features of different source domains in a segment-by-segment manner, and obtain the multi-source domain fusion features; (5) Inputting the N source domain fault features into the corresponding N source domain classifiers to obtain the fault classification results of the N source domains; (6) The multi-source domain fusion features obtained in step (4) and the target domain fault features obtained in step (3) are simultaneously input into the weight-sharing three-branch cross-domain alignment convolutional Transformer, and feature alignment is performed in a segment-by-segment manner to obtain high-level source domain and target domain fault features and cross-domain related features; (7) Input the fault features obtained in step (6) into the target domain classifier to obtain the fault classification results corresponding to the three branches; (8) Repeat steps (3) to (7) until the maximum number of predictions N is reached. p , get N for each training sample p fault classification results in different domains, and calculate the average value of the fault classification results in different domains as the final classification result; (9) Calculate the loss function L that integrates multiple factors total , including the hierarchical multi-metric domain difference loss L dis , classification loss and distillation loss L, and use the back-propagation algorithm to update the network parameters to minimize the overall loss value; dtl (10) Repeat steps (2) to (9) to traverse all training samples until the maximum number of training rounds M is reached. epoch , we obtain the optimal multi-source domain deep migration network model; (11) Activate the network in the target domain data flow direction of the optimal multi-source domain deep migration network model, input the test set samples into the optimal multi-source domain deep migration network model, and calculate the corresponding fault classification results; (12) Repeat step (11) until the maximum number of predictions N is reached p , get N for each test sample p fault classification results and calculate the average value of the fault classification results as the final classification result. p The prediction variance V corresponding to the fault classification result p , quantifying the uncertainty of classification predictions.

2. The cross-domain fault diagnosis method for drone clusters based on a multi-source domain deep migration network according to claim 1 is characterized in that: The different operating conditions of the drone cluster in step (1) include the flight of the drone cluster in different flight formation shapes and different flight speeds under different communication topologies; each operating condition is regarded as a domain; flight status monitoring data under N existing operating conditions are collected as N source domain data sets, and limited flight status monitoring data under a new operating condition is collected as a target domain data set.

3. The cross-domain fault diagnosis method for drone clusters based on a multi-source domain deep migration network according to claim 1 is characterized in that: The different faults described in step (1) include sensor and actuator failures of a single drone, sensor and actuator failures of two or more drones at the same time, communication interruption failures between two drones, and communication failures of non-communicating drones.

4. The cross-domain fault diagnosis method for drone clusters based on a multi-source domain deep migration network according to claim 1 is characterized in that: The flight status monitoring data in step (1) includes the absolute value of the relative distance change between the current UAV and its neighboring communicating UAV in the lateral and longitudinal directions, the speed deviation and the average value of the overall speed, the lateral and longitudinal speeds, the climb angle, the yaw angle, the bank angle and the throttle opening of the current UAV.

5. The cross-domain fault diagnosis method for drone clusters based on a multi-source domain deep migration network according to claim 1 is characterized in that: The implementation process of step (2) is as follows: Given N source domain input samples and a target domain input sample First, the samples of the source domain and the target domain are respectively extracted through the domain feature extractor to extract the fault features in each domain. Afterwards, the N source domain fault features are obtained Two operations are performed: first, the fault features of each domain are input into the respective source domain classifiers to obtain the classification prediction results, i.e. Second, N source domain fault features are divided into N pairs of feature combinations in the ratio of 1:N-1. These feature combinations are then input into the multi-source domain cross convolution Transformer F mcct (·) to obtain multi-source domain fusion features Next, the multi-source domain fusion features and target domain fault characteristics Input into the weight-sharing three-branch cross-domain alignment convolutional Transformer to perform feature alignment in a segment-by-segment manner to obtain high-level source domain fault features and target domain fault characteristics And cross-domain related features Finally, the obtained features are input into the target domain classifier Obtain the target domain fault classification results.

6. The cross-domain fault diagnosis method for drone clusters based on a multi-source domain deep migration network according to claim 1 is characterized in that: The source domain and target domain classifiers in step (2) have the same network structure, including two Bayesian linear layers, ReLU activation function and Softmax function layer; the source domain fault features are established through the classifier High-level source domain fault characteristics and target domain fault characteristics And cross-domain related features The nonlinear mapping relationship between the fault classification label and the 7. The method for cross-domain fault diagnosis of drone clusters based on a multi-source domain deep migration network according to claim 1 is characterized in that: The domain feature extractor in step (3) includes a common feature extractor F co (·) and domain-specific feature extractors The common feature extractor is composed of a weight-sharing spatiotemporal Bayesian graph wavelet convolutional Transformer; the specific domain feature extractors are composed of independent spatiotemporal Bayesian graph wavelet convolutional Transformers; The spatiotemporal Bayesian graph wavelet convolution Transformer includes multi-head spatiotemporal convolution attention, Bayesian linear layer, residual connection and layer normalization; the multi-head spatiotemporal convolution attention includes H spatiotemporal convolution attentions arranged in parallel, each spatiotemporal convolution attention extracts spatiotemporal coupling fault features from different scales; the input sample passes through H spatiotemporal convolution attentions to obtain H spatiotemporal coupling fault features and splices them, and then passes through two layers of Bayesian linear layers with residual connection and layer normalization in sequence to obtain the final spatiotemporal coupling fault features; The spatiotemporal convolutional attention includes multi-layer multi-scale adaptive graph wavelet convolution, local and global scaling dot product attention, dilated causal convolution network layer and dimension adjustment layer; first, multi-layer multi-scale adaptive graph wavelet convolution processes input samples Get the corresponding query vector Q h , the key vector K h Sum value vector V h , and input into the local scaled dot product attention to obtain the spatial feature sequence Then the dimension adjustment layer adjusts the dimension of the spatial feature sequence to obtain where d model =Ne d ×N f ; Then, two parallel dilated causal convolutional network layers learn the spatial feature sequence The local context dependency of the corresponding query vector Q h ′, key vector K h ′; Similarly, the dimension adjustment layer adjusts the input samples Dimensions and take it as the value vector V h ′; The query vector Q h ′, key vector K h ′ and value vector V h ′Input into the global scaled dot product attention to adaptively integrate the local temporal features of the input samples into the modeling of global dependencies to simultaneously model local and global spatiotemporal coupling features and extract spatiotemporal coupling fault features; The multi-layer multi-scale adaptive graph wavelet convolution is composed of K layers of multi-scale adaptive graph wavelet convolution connected in series; each multi-scale adaptive graph wavelet convolution includes |S| scale adaptive graph wavelet convolutions to extract spatial features in input samples from different scales; each adaptive graph wavelet convolution is implemented by graph wavelet transform; the graph wavelet transform uses a set of suitable wavelet bases provided by wavelet transform to project the graph signal from the vertex domain to the spectral domain, and extract fault features in the spectral domain; The wavelet basis is defined as Where Ne is the number of drones, each wavelet basis ψ s,i Corresponding to the signal of the i-th UAV at scale s; the graph wavelet with scale parameter s on the i-th UAV can be written in the following matrix multiplication form: Ψ s =UG s AT T in, is the scaling matrix, λ l is the normalized graph Laplacian matrix The lth eigenvalue of , U is The eigenvector matrix of ; the normalized graph Laplacian matrix It is obtained from the adjacency matrix A corresponding to the UAV cluster communication topology, which is defined as Among them I n is the identity matrix of the corresponding dimension, is the angle matrix The graph wavelet basis is approximated by Chebyshev polynomials as follows: c i,s =2e -s J i (-s) in, is the i-th order Chebyshev polynomial, Q is the number of Chebyshev polynomials, J i (-s) is the first kind of Bessel function; for input samples with graph structure Wavelet transform of the graph is The corresponding inverse transformation is The adaptive graph wavelet convolution with scale s is defined as: Among them, s is a graph wavelet basis with scale s, W s (i,j) are learnable parameters, is the output signal, σ(·) is the nonlinear activation function; the adaptive graph wavelet convolutions with different scales are connected in parallel and combined with residual connections as an adaptive component to form a multi-scale adaptive graph wavelet convolution, and then multiple multi-scale adaptive graph wavelet convolutions are connected in series to form a multi-layer multi-scale adaptive graph wavelet convolution; in the domain d∈{d s1 ,...,d sN ,d t The corresponding output of the s-th scale of the K-th layer adaptive graph wavelet convolution in} is: in, is the initial input of the network, and are the graph wavelet basis and its inverse at scale s, Θ d,s is the diagonal matrix of the graph convolution kernel, and is a trainable parameter matrix, σ(·) is a nonlinear activation function; in the domain d∈{d s1 ,...,d sN ,d t }, the final output of the multi-layer multi-scale adaptive graph wavelet convolution is the average value of the output features of |S| parallel adaptive graph wavelet convolutions in the K-th layer of multi-scale adaptive graph wavelet convolution, which is defined as follows:

8. The cross-domain fault diagnosis method for drone clusters based on a multi-source domain deep migration network according to claim 1 is characterized in that: The multi-source domain cross convolution Transformer in step (4) includes N multi-source cross convolution attention layers, residual connections, layer normalization, and linear layers; it is used to learn the relationship between different source domain features and extract multi-source domain fusion features; The multi-source cross convolution attention includes an expanded causal convolutional network layer, a repeated layer, and a scaled dot product attention; N source domain fault features are sequentially regarded as value vectors, and the remaining N-1 source domain fault features are regarded as key vectors and value vectors; specifically, the N source domain fault features are sequentially divided into N pairs of feature combinations according to a ratio of 1:N-1, namely Input N pairs of feature combinations into N multi-source cross convolutional attentions respectively, It corresponds to The splicing features of the fault features of the remaining source domains except ; After the dilated causal convolutional network layer, the value vector V is obtained cn , After the expansion causal convolutional network layer, the query vector Q is obtained cn , key vector K cn , the repeated layer passes through the value vector V cn The copy operation is performed to make its dimension the same as the corresponding query vector Q cn , key vector K cn Consistency; The query vector Q cn , key vector K cn Sum value vector V cn Input the scaled dot product attention, calculate the attention, learn the relationship between each source domain feature and the other source domain features in a segment-by-segment manner, and obtain N multi-source domain interaction features {Attn c1 ,...,Attn cN }; Further, the N multi-source domain interaction features are concatenated, and then further fused through two layers of linear layers with residual connection and layer normalization to obtain the final multi-source domain fusion feature 9. The cross-domain fault diagnosis method for drone clusters based on a multi-source domain deep migration network according to claim 1 is characterized in that: Step (6) The weight-shared three-branch cross-domain aligned convolutional Transformer includes a source branch, a source-target branch, and a target branch; the multi-source domain fusion features and the target domain fault features are aligned in a segment-by-segment manner to extract domain-invariant features, thereby utilizing the source domain fault features to improve the fault diagnosis performance in the target domain; the source branch, the source-target branch, and the target branch are all composed of a weight-shared convolutional Transformer; the multi-source domain fusion features are aligned with the target domain fault features ... Input to the source branch, and processed by the dilated causal convolutional network layer to obtain the query vector Q s , key vector K s Sum value vector V s , and then through multi-head self-attention, two layers of residual connection, layer normalization and linear layer to obtain high-level source domain fault features Similarly, the target domain fault characteristics Input into the target branch, and processed by the dilated causal convolutional network layer to obtain the query vector Q t , key vector K t Sum value vector V t , and then through multi-head self-attention, two layers of residual connection, layer normalization and linear layer to obtain high-level target domain fault features The query vector Q in the source branch s and the key vector K of the target branch t Sum value vector V t At the same time, it is input into the multi-head cross-domain attention of the source-target branch to learn the relationship between the source domain and target domain features, and obtain cross-domain related features; Then, the multi-source domain features are fused and high-level features of the target domain Splicing to get the cross-domain combined feature F cd , and input it into the residual connection and layer normalization in the source-target branch to fuse with the cross-domain alignment features, and then pass through the linear layer, residual connection and layer normalization to obtain the cross-domain related features Next, the cross-domain related features obtained by the three branches are and They are respectively input into the weight-shared target domain classifier and the corresponding fault classification value is output.

10. The cross-domain fault diagnosis method for drone clusters based on a multi-source domain deep migration network according to claim 1 is characterized in that: Step (9) The hierarchical multi-metric domain difference loss L is used to optimize each pair of dis The distribution difference of fault features at different levels in the target domain and the source domain; it includes the low-level domain distribution loss and the high-level domain distribution loss obtained by weighted addition of the maximum average difference loss and the CORAL distance, which are used to measure the distribution difference between the fault features of the source domain samples after passing through the common feature extractor and the specific domain feature extractor and the corresponding target domain fault features; The classification loss It is used to optimize the difference between all classifier output values ​​and fault classification label values; cross entropy loss is selected as the classification loss, and the total classification loss is in Represents the classification loss value corresponding to N source domain classifiers, The classification loss value of the source branch and the classification loss value of the target branch corresponding to the target domain classifier; The distillation loss L is used to improve the fault diagnosis performance in the target domain by utilizing the fault features in the source domain; dtl Specifically, the weight-shared three-branch cross-domain alignment convolution Transformer outputs the cross-domain related features of the source-target branch The fault classification probability distribution output in the target domain classifier is used as a soft label to supervise the training of the target branch through the following distillation loss to improve the classification performance of the diagnosis model on the target domain samples. Among them, q k and p k are the probabilities of the classification category k corresponding to the output of the source-target branch and the target branch respectively; The loss function L of the comprehensive multi-factors is: total Among them, λ and β are weight parameters.

Images