326 results about "Message passing" patented technology

The invention discloses a machine vision-based resistor disc defect online detection system and a grading method, relates to the technical field of industrial machine vision detection, and solves the defect problems in the aspects of multi-scale defect dynamic perception, cross-level feature interaction and process adaptive optimization in the prior art. According to the scheme, metal reflection interference is inhibited through Retinex illumination correction and a combined denoising model; adopting a deformable convolution kernel and cavity space pyramid pooling to realize gradient entropy driving dynamic sensing of the multi-scale defect; constructing a bidirectional cross-layer attention network to realize early fusion of high-resolution details and high-level semantics; modeling local-global feature physical association based on a graph attention network and a self-supervised message passing mechanism; integrating reinforcement learning and a memristor random calculation unit to form a closed-loop parameter optimization system; according to the method, the multi-scale defect detection precision, the cross-modal feature fusion efficiency and the system adaptive capacity under complex working conditions are remarkably improved.

The invention belongs to the field of intelligent traffic systems, and relates to a multi-mode space-time traffic flow modeling method supporting large-scale road network real-time prediction, and the method comprises the steps: firstly designing a space-time prediction framework facing a dynamic traffic network, and then carrying out the training to obtain a final prediction model; the space-time prediction framework comprises a data embedding layer, a space-time coding module and a deep modeling and output module based on an expert hybrid mechanism; the data embedding layer comprises two parallel channels of time embedding and spectral domain space embedding and a time-space data fusion module; the space-time coding module comprises a block-level sparse time attention module, a space attention-message passing module and a weighted fusion layer which are parallel; the deep modeling and output module based on the expert hybrid mechanism comprises an MoE dynamic expert modeling module, a full-connection mapping module, a jump connection layer and an output layer; according to the design, the response speed, the prediction precision and the cross-regional adaptive capacity of the model in a high-heterogeneity scene are improved.

The invention discloses a network protocol reverse analysis method based on deep learning and a graph neural network, and the method comprises the steps: basic field detection: carrying out the byte-level feature extraction of a binary data stream through sliding window embedding, bidirectional LSTM coding and knowledge enhanced CRF decoding, and outputting a structured field labeling sequence; performing protocol format clustering based on the sequence: calculating a multi-dimensional similarity through an improved Needleman-Wunsch algorithm, realizing automatic classification of unknown protocols in combination with a dynamic density clustering algorithm optimized by LSH, and outputting a protocol cluster; and performing composite structure analysis based on the protocol cluster: constructing a protocol syntax tree based on a graph neural network, performing multiple rounds of message passing through a graph attention network GAT attention mechanism, identifying a nested structure and performing recursive analysis, and generating a multi-level protocol syntax tree. According to the method, the automation degree and accuracy of complex protocol analysis can be remarkably improved.

The invention discloses a relation extraction method and system based on a graph neural network, and belongs to the technical field of natural language processing. A target text is obtained, word segmentation, part-of-speech tagging and named entity recognition are carried out, and an entity set is extracted; constructing a text graph structure containing multiple edge types based on the entity set; performing feature coding on nodes in the graph to generate an initial feature vector fusing semantic, part-of-speech and position information; inputting the graph into the graph neural network model, and obtaining high-order node representation through multi-layer message passing and aggregation; modeling the entity pair in combination with the structure path and the context information, and inputting a multi-channel classification network to predict the relationship type of the multi-channel classification network; and finally, outputting an entity relationship triple according to a prediction result. The method has stronger semantic modeling ability and structure expression ability in a relation extraction task, and is suitable for scenes such as knowledge graph construction and information extraction systems.

The invention discloses a construction environment sudden change risk field prediction method fusing geological and meteorological data, and belongs to the technical field of construction engineering risk prediction, and the method comprises the steps: obtaining geological data, meteorological data and construction progress data of a construction area, carrying out the structural processing of the multi-modal heterogeneous data, and carrying out the construction engineering risk prediction. And generating graph data containing the multi-modal features. According to the method, a space-time adaptive graph neural network prediction model based on an attention mechanism is adopted, weights of different features can be dynamically calculated and allocated according to real-time environment and construction data, and a space-time propagation relationship of risks among geographic nodes is captured by using a message passing mechanism of a graph. According to the method, data of different modals can be deeply fused, accurate prediction of the construction environment sudden change risk is realized, corresponding early warning is triggered according to the risk level, the accuracy and timeliness of risk prediction are improved, and scientific support is provided for intelligent decision making of constructional engineering.

The embodiment of the invention provides a low-altitude digital management method and device based on an urban space-time grid, and the method and device achieve the precise division of an airspace through a plurality of layers of grid units by creatively building a unified urban space-time reference grid system. And designing a space-time grid data assignment model, processing multi-source heterogeneous data by using a deep neural network and a space-time convolutional layer, and realizing intelligent mapping of grid features through an attention mechanism. And constructing an airspace situation awareness model based on a graph neural network, and capturing a space-time dependency relationship among grid units in combination with a message passing mechanism to realize accurate prediction and hierarchical management and control of an airspace state. According to the method, the defects of the traditional technology in the aspects of airspace division, data processing, situation awareness and the like are effectively overcome, and the accuracy and practicability of urban low-altitude digital management are remarkably improved.

The embodiment of the invention relates to the technical field of data storage, and discloses a cross-K8S cluster configuration change storage method and device based on a block chain, and the method comprises the steps: obtaining a transaction request message passing the compliance verification of an intelligent contract engine, and the transaction request message comprises a configuration change request; obtaining log information of the configuration change request in the transaction request message executed by the target K8S cluster, wherein the log information further comprises configuration change cluster difference information; the log information is stored in a fragmented mode based on the IPFS network, and content hash codes of the log information are obtained; and taking the transaction request message and the content hash code as complete transaction information, and storing the transaction request message and the content hash code in a transaction pool. Decentralized storage is achieved based on an IPFS network fragmentation storage mode, the problem that a centralized database or a log system is maliciously modified or historical records are deleted easily in centralized storage is solved, and auditing integrity and data storage safety are guaranteed.

The invention relates to an unmanned aerial vehicle path planning method and system based on GNN and high-order security constraints. The method comprises the steps that a navigation scene where an unmanned aerial vehicle is located is represented as a heterogeneous directed graph, message passing and feature updating are conducted through the GNN, a risk-aware attention mechanism is introduced, and an interpretable decision result is output; a differentiable HoCBF-QP optimization layer is introduced, an original control command output by a strategy network is used as input, quadratic programming with high-order control barrier function constraints is solved online, minimum-amplitude safety correction is carried out on the control command, and an actuator control command meeting safety constraints is output; starting a HoCBF safety shield during operation so as to strictly ensure that all safety constraints are met before execution; a calculation task of the whole control cycle is modeled into a directed acyclic graph form, and parallel execution is carried out on heterogeneous multiple cores by utilizing a real-time scheduling strategy. The problem that unmanned aerial vehicle navigation control is not effective and unified in three aspects of structure, safety and scheduling is solved.

The invention discloses an electric power engineering cost simulation analysis method and system based on digital twinning, and relates to the technical field of simulation analysis, and the method comprises the steps: obtaining multiple engineering task multi-modal data and task dependency relationships for electric power engineering; processing the multi-modal data of the plurality of engineering tasks and the task dependency relationship based on a digital twinning technology so as to construct an electric power engineering virtual model; inputting the electric power engineering virtual model into a graph neural network so as to extract task node features corresponding to each engineering task virtual module through a message passing mechanism; and defining an input state of a reinforcement learning model according to each extracted task node feature so as to generate a task resource allocation action for each engineering task through reinforcement learning. Therefore, the digital twinborn technology, the graph theory feature extraction technology and the reinforcement learning technology are comprehensively applied, real-time monitoring and simulation analysis are carried out on cost and resource consumption of different links in the electric power engineering, and resource allocation is intelligently optimized.

The invention provides a drug response prediction method based on a gene relation network and a drug substructure. The drug response prediction method comprises the following steps: performing feature extraction on three multi-modal features of gene expression, copy number variation and mutation of a cell line by adopting an enhanced graph attention network; acquiring a substructure of the medicine by using a gating message passing neural network, and extracting features by enhancing a graph attention network; constructing two complementary graph structures, namely an atom-bond graph and a bond-angle graph, and performing feature extraction and message passing on the two graphs by using a graph convolutional neural network so as to extract 3D features of the medicine; the characteristics of the drug and the cell line are integrated through concat operation, and drug response prediction is carried out through mlp. According to the drug response prediction method provided by the invention, cell line features are extracted by establishing and using three kinds of omics data with different emphasis, and drug feature extraction is performed by using a drug molecular map, a drug 3D structure, a drug molecular fingerprint and a drug substructure relationship.

The invention discloses a transformer fault diagnosis and root cause positioning method based on a space-time diagram neural network, and relates to the field of transformer fault diagnosis, and the method comprises the steps: S1, carrying out the preprocessing of the structure information and DGA time series data of a transformer, and obtaining a topological network structure diagram and a DGA time series; s2, obtaining a spatial vector based on a message passing mechanism of a graph convolutional network; s3, a time sequence vector is obtained in combination with a Transform encoder and multi-head self-attention; s4, obtaining space-time fusion features; s5, constructing a multi-task prediction head based on the space-time fusion feature, the fault type historical data and the fault root cause; and S6, carrying out fault detection and outputting a corresponding fault type and root cause positioning result. According to the application, the accuracy of fault type identification can be remarkably improved, and accurate positioning of the fault root cause is realized.

One or more systems, devices, computer program products and / or computer-implemented methods of use provided herein relate to predicting an optimized result for a graph neural network (GNN). A system can comprise a memory configured to store computer executable components; and a processor configured to execute the computer executable components stored in the memory, wherein the computer executable components comprise: a fusion component that that models non-linear modality correlations within and across entities through Hirschfeld-Gebelein-Re'nyi maximal correlation (MaxCorr) embeddings that generates a multi-graph that preserves identities of modalities and entities; and a multi-graph neural network (MGNN) component for task-informed reasoning in multi-graphs, that learns parameters defining entity-modality graph connectivity and message passing in an end-to-end fashion.

The disclosure relates to methods and systems of partitioning-based scalable weighted aggregation composition for embeddings learned from knowledge graphs for training neural networks to perform downstream machine-learning tasks. For example, a system may access a knowledge graph comprising a plurality of nodes and partition the knowledge graph into a plurality of partitions based on edge densities between nodes of the knowledge graph. The system may perform partition-wise encoding using compositional message passing between nodes that enables learning from neighboring nodes. The system may generate an embedding for each node and each relation type in each partition based on the partition-wise encoding using compositional message passing. The system may concatenate the generated embeddings from the plurality of partitions. The system may train a global neural network for a downstream prediction task based on the concatenated embeddings using one or more weight matrices.

The invention relates to the technical field of infrared light machine system thermal deformation prediction, in particular to a multi-material structure thermally induced stress deformation prediction method based on a graph neural network. The method comprises the steps of data set establishment, graph structure establishment, graph neural network model establishment and training and model and parameter optimization. Finite element nodes correspond to graph nodes, finite element edges correspond to graph edges, an encoder-message passing-decoder architecture model is established, and node states are updated through a three-layer physical symmetry message passing mechanism. Physical constraint loss including minimum displacement smoothness constraint and stress continuity constraint is innovatively added into a loss function. Compared with traditional finite element calculation, the method has the advantages that the speed is increased by more than 100 times, high hardware adaptability is achieved, the black box limitation of a data-driven neural network model is broken through, thermally induced stress deformation analysis caused by different material coefficients can be processed, the adaptability to geometric changes is high, and good engineering application value is achieved.

The invention discloses an integrated circuit layout design teaching assistance method, and relates to the field of integrated circuits, and the method comprises the steps: receiving a circuit diagram creation instruction, and generating a circuit diagram drawing interface according to the circuit diagram creation instruction; identifying types and connection relations of circuit devices in the circuit diagram drawing interface; according to the identified circuit device type and the connection relation, generating a corresponding circuit schematic diagram data structure by adopting a graph neural network GNN algorithm based on message passing; converting the schematic circuit diagram data structure into a layout data structure; the layout data structure comprises geometric parameters and layout position data of layout elements; and according to the layout data structure, generating a circuit layout by using a reinforcement learning algorithm based on a graph neural network. In view of low integrated circuit layout design efficiency in the prior art, the design efficiency is improved by constructing a multi-objective optimized Markov decision process, utilizing deep learning and reinforcement learning algorithms and the like.

The invention relates to the technical field of building engineering construction management, and discloses a house steel building construction prediction progress and deployment system based on deep learning. According to the system, periodic stable paragraphs and non-periodic fluctuation paragraphs are recognized by analyzing the time sequence form of construction flow data, and construction links are stripped accordingly. And performing cross mapping on the hoisting event sequence in the link and the environmental monitoring reading of the associated link to generate a link toughness spectrogram representing the construction robustness. And determining a resource demand based on the spectrum graph, and forming a multi-dimensional resource demand vector, so as to drive a graph convolutional network to construct a dynamic construction deduction graph. And carrying out message passing and neighborhood aggregation on the graph, analyzing a key bottleneck path, and iteratively generating a progress prediction and resource allocation scheme. According to the method, the construction dynamic toughness can be evaluated, and accurate positioning and dynamic optimization of resource bottlenecks are realized.

The invention discloses a mining equipment state prediction method based on digital twinning and multi-modal fusion, and relates to the technical field of intelligent operation and maintenance, and the method comprises the steps: inputting a multi-modal alignment working condition window sequence into a digital twinning mechanism link, carrying out the same-window ideal response deduction, outputting an ideal multi-modal response, and constructing a twinning dynamic residual error; meanwhile, segmenting a residual structure event to generate a residual event token set; positioning a multi-modal data fragment based on the residual event token set, executing cross-modal attention alignment, and performing topology propagation aggregation in combination with a mechanical topology observation mapping table to generate a topology constraint fusion table set; and inputting the topological constraint fusion table set into the graph neural network, carrying out message passing in combination with a mechanical topological graph, outputting a component state and a complete machine state, and packaging the component state and the complete machine state into a mining equipment state prediction set. According to the method, structured analysis of the twinborn dynamic residual error is realized, and the method is used for accurately positioning an abnormal event and improving the sensitivity and timeliness of state prediction.

The invention discloses a three-dimensional geographic environment real-time intelligent deduction method based on a multi-modal large model, and relates to the technical field of computer vision. The method is used for solving the technical problem of unified modeling and real-time deduction of a dynamic object and a static environment in a three-dimensional geographical environment. The method comprises the following steps: firstly, resolving a camera pose through a motion recovery structure algorithm to generate a sparse point cloud, and separating a dynamic foreground object in a monitoring video to extract motion features; thirdly, initializing a three-dimensional Gaussian distribution set based on the sparse point cloud, extracting semantic features through a visual encoder, and mapping the semantic features to corresponding Gaussian distribution; thirdly, a topological graph structure of Gaussian distribution is constructed, motion features are used as initial excitation, and coordinate offset and appearance variation of each distribution are iteratively updated through message passing calculation; finally, Gaussian distribution attributes are dynamically updated, a continuous deduction image sequence is synthesized through micro-rasterization rendering, and high-reality real-time simulation of dynamic evolution of the three-dimensional geographical environment is achieved.

The invention relates to the technical field of market research data analysis, in particular to a market research data analysis method and system.The method comprises the steps that multi-source data is collected and cleaned through a multi-channel self-adaptive data integration technology, and a comprehensive data set is generated; constructing a graph network structure based on a graph neural network, updating node features through a message passing mechanism, generating market prediction data, and optimizing an investigation path based on a reinforcement learning algorithm; comparing the market feedback data with the prediction data through a feedback self-learning mechanism, and adjusting an investigation strategy to obtain an optimized investigation path; the performance of the market under extreme conditions is simulated through Monte Carlo simulation, the abnormal behaviors of the market are identified through dynamic outlier detection, a coping strategy is generated, and an investigation path is adjusted. The method effectively improves the accuracy of investigation data and the allocation efficiency of investigation resources, and adapts to a complex market environment.

A system and method for generating thermostable variants of a protein is disclosed. The system receives a three-dimensional structure of a target protein and identifies mutable regions, including solvent-exposed residues and loop regions. Conserved and active site residues are excluded from mutation through a fixed-position mask. A message-passing neural network (MPNN) generates mutant sequences at unmasked positions, executed under multiple temperature parameters. Design scores based on Shannon entropy and log probability are computed, and high-confidence variants are selected. Predicted structures for selected variants are evaluated using structural and sequence-based features to compute stability scores. A ranked list of thermostable variants is generated. Top candidates undergo molecular dynamics simulations to compute dynamic metrics such as RMSD, radius of gyration, SASA, and ddG, and are re-ranked accordingly. The system enables accurate, constraint-driven protein design with high structural and functional fidelity, suitable for industrial and therapeutic applications.

The invention provides a power grid fault propagation path identification method and system based on a graph neural network, and relates to the technical field of power grid fault identification, and the method comprises the steps: obtaining a power grid topological structure and fault observation data, constructing a direction perception graph neural network, achieving the one-way message passing based on potential energy difference, and obtaining a node potential energy value and embedded representation; edges are screened to generate a directed graph, and candidate paths are generated in the potential energy gradient direction; and verifying the satisfaction degree of the graph neural network evaluation path to the physical constraint through an invariant, and finally selecting the path with the minimum topological entropy. According to the invention, the accuracy and reliability of fault propagation path identification are improved.

The invention discloses an oil reservoir water content prediction method based on a graph neural network and Transform, and belongs to the technical field of oil and gas field development, and the method comprises the following steps: S1, converting an oil reservoir injection-production model at each time point into a graph structure, and constructing a connection network from a water injection well to a production well; and S2, constructing a GNT proxy model in combination with the graph neural network and Transform, training the GNT proxy model, and predicting the water content of each production well. According to the constructed GNT proxy model, an oil reservoir injection-production structure is converted into graph data through a GCN module, the information transmission strength between nodes is dynamically adjusted through an edge feature enhanced message transmission mechanism, and inter-well communication features are accurately captured; and the multi-head self-attention mechanism of the Transform encoder module is combined to process the time sequence dependence, so that the long-distance time dynamic relationship is effectively modeled, and an efficient and reliable agent model is provided for the dynamic analysis of the oil reservoir.

The invention discloses an industrial product quality risk association method based on a graph neural network, and relates to the field of industrial manufacturing, and the method comprises the steps: collecting multi-source industrial data, carrying out the preprocessing of the multi-source industrial data, obtaining a standardized data table, and constructing an industrial element heterogeneous graph based on the standardized data table; inputting the industrial element heterogeneous graph into a graph attention network basic model for node representation learning, and setting a supervised learning task taking a product batch node in the industrial element heterogeneous graph as a prediction target and taking a historical quality detection result as a supervised label; and introducing a meta-path self-learning layer in a message passing process of the graph attention network basic model. According to the method, deterministic prediction is upgraded into probabilistic inference, internal confidence degree measurement is provided for each risk prediction value, and automatic, accurate and interpretable correlation analysis and traceability positioning of the industrial quality risk are realized.

The invention belongs to the technical field of metabonomics large model construction, and relates to a metabonomics large model construction method based on multi-dimensional database fusion, a global unique ID is allocated for metabolites, and multi-source heterogeneous data is uniformly coded into a multi-modal feature matrix, so that information loss caused by data isomerism in a traditional method is avoided; a dynamic graph generator is adopted, a multi-modal feature matrix and a prior path adjacency matrix are combined, dynamic edge weights are generated through similarity and learnable parameters to construct a metabolic relationship graph, a multi-modal graph attention mechanism is utilized to distribute adaptive weights for different modal features, graph convolution message passing is executed, and the metabolic relationship graph is obtained. Generating a node embedding matrix fused with multi-modal information and an optimized dynamic metabolism relation graph; besides, through a meta-learning method, each database is split into meta-tasks, a dynamic graph is used for training a task exclusive classifier in an inner cycle, a dynamic graph generator and meta-model parameters are jointly optimized through query set loss in an outer cycle, and the cross-database generalization ability of the model is remarkably improved.

The invention discloses a wind power plant data synchronous correction method based on view topology and a mask graph neural network, and belongs to the field of wind power plant data processing and artificial intelligence space-time sequence prediction.The method comprises the steps that firstly, historical space-time observation data of unit nodes are obtained, and a space-time feature tensor is constructed; generating a mask matrix; extracting space and time sequence waveform correlation characteristics of a plurality of unit nodes, and constructing a global multi-view adjacency matrix; inputting the spatio-temporal feature tensor and the global multi-view adjacency matrix into a spatio-temporal diagram neural network, performing blocking and shielding operation in a message passing stage of the network, extracting full-field features, and outputting an initial deduction sequence; and obtaining a boundary physical residual error, reversely compensating the boundary physical residual error to the initial deduction sequence, and generating and outputting final correction data. According to the method, test data high-fidelity error correction under high concurrency missing rate and strong noise interference is realized, and the space-time diagram calculation memory overhead of a bottom-layer deep learning framework is remarkably reduced by designing a static topology cache mechanism.

The invention belongs to the technical field of multi-view clustering, and discloses an end-to-end multi-view clustering method and system for dynamic cross-view node interaction. The method comprises the following steps: constructing graph structure information according to node features in a multi-view data set; and inputting the graph structure information and the node features into the multi-view graph neural network model to obtain a clustering result. The multi-view graph neural network model comprises an encoder, a multi-layer perceptron, a decoder and a hyperspherical clustering module with regularization constraint; the encoder comprises a plurality of encoder layers, and each encoder layer comprises an in-view representation learning module and a cross-view representation learning module. According to the method, an unsupervised end-to-end node-level cross-view message transmission mode is provided, direct information interaction between cross-view nodes is established, complementary information propagation and consistent semantic information learning are effectively promoted, clustering is carried out by adopting a hypersphere with regularization constraint, and the clustering efficiency is improved. Balanced distribution and good separability between different categories can be ensured.

The invention relates to a UGT enzyme subtype substrate specificity prediction method and model. On the basis of a directional message passing neural network, graph structure characterization of a small molecule compound and features of specific protein binding sites of UGT enzyme are deeply fused, a bimodal prediction normal form of'molecule + protein binding sites' is designed, a deep learning model is constructed, conversion from compound center prediction to molecule-enzyme binding site comprehensive prediction is achieved, and the prediction accuracy is improved. And accurate classification prediction can be carried out on UGT enzyme substrates and non-substrates.

The invention discloses an intelligent supply and demand matching dynamic optimization method and device and a medium, and relates to the technical field. The method comprises the steps that demand side data and supply side data are collected, the demand side data comprise user interaction logs and real-time positions, and the supply side data comprise inventory data, logistics track data and production line load data; fusing demand side data and supply side data through a knowledge graph, inputting the demand side data and the supply side data into a graph neural network, encoding the demand side data and the supply side data into high-dimensional vectors, and generating a supply-demand matching preliminary screening candidate set; expanding a preliminary screening candidate set through a message passing mechanism of the graph neural network; and searching a non-dominated solution set by adopting a multi-objective optimization algorithm, and distributing index weights in real time so as to sort the expanded preliminary screening candidate sets. According to the method, real-time fusion of multi-source data is realized, a multi-target optimal solution is generated by using deep learning and operational research algorithms, and the strategy is continuously adjusted and optimized in combination with reinforcement learning.

The invention discloses a material performance prediction method and system based on deep learning, and relates to the technical field of material performance prediction, and the method comprises the steps: building a cross-level message passing network based on a double-layer graph data structure, extracting atomic-level and functional group-level feature embedding, and generating a molecular global latent variable through bidirectional cross-scale attention interaction; fusing molecular global latent variables with atomic-scale and functional group-scale features, applying physical constraints in combination with a chemical bonding rule base and a group contribution theory, and constructing a physical property prediction model; obtaining candidate molecules through autoregression generation and molecular force field verification based on a molecular global latent variable and a physical property prediction model; and screening candidate molecules by using the physical property prediction model, and outputting a final molecule set through molecular dynamics simulation and synthesis feasibility evaluation verification. Through multi-stage verification of molecular dynamics simulation and synthesis feasibility evaluation, a high-tension ring or a non-synthesized structure is effectively eliminated, so that the generated molecule has high performance and manufacturability.

The invention discloses a wind turbine generator fault diagnosis method based on knowledge embedded heterogeneous graph learning, which comprises the following steps: firstly, acquiring time sequence historical operation data of a multivariable sensor from a wind power plant, constructing a domain knowledge-based heterogeneous graph by using the processed multivariable time sequence data, and then carrying out heterogeneous graph representation learning; respectively inputting the domain knowledge heterogeneous graph into a heterogeneous interactive attention module to calculate an attention vector between nodes and a heterogeneous message passing module to calculate information between the nodes, and then weighting and aggregating the calculated attention vector and the information between the nodes from a source node to a target node; and obtaining the feature representation of each heterogeneous graph through end-to-end training. And finally, inputting the global graph representation into a full connection layer, outputting probabilities of different fault types by using a SoftMax classifier, and finally obtaining a fault type label of each state. According to the method, the heterogeneous graph embedded by the domain knowledge is constructed, so that fault features can be learned more effectively, and the fault diagnosis accuracy is improved.