Diagnostic scheme generation method and system based on multi-scale hypergraph and constraint optimization

Abstract

The invention discloses a diagnosis scheme generation method and system based on multi-scale hypergraph and constraint optimization, and belongs to the technical field of industrial equipment health management. The method comprises the following steps: receiving fault diagnosis query information expressed by a natural language, and carrying out semantic analysis on the fault diagnosis query information to form semantic guide information; task nodes are extracted from the technical information set, and a multi-scale nested subgraph is constructed according to the relation between the task nodes; based on the multi-scale nested subgraph, dynamically reconstructing a task path, representing task nodes as hypergraph nodes, and constructing a multi-scale hypergraph; performing structure sensing regularization processing on the multi-scale hypergraph to obtain the multi-scale hypergraph; performing cross-scale attention processing on the multi-scale hypergraph to obtain a multi-scale fusion hypergraph; and generating fault diagnosis paths based on the multi-scale fusion hypergraph, sorting the fault diagnosis paths, and outputting a diagnosis scheme. According to the invention, the intelligent level and decision-making efficiency of complex system fault diagnosis can be significantly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial equipment health management, and in particular relates to a diagnosis scheme generation method and system based on multi-scale hypergraph and constraint optimization. BACKGROUND

[0002] In the field of industrial equipment health management, the generation of fault diagnosis schemes mainly relies on the comprehensive use of equipment knowledge, historical data and expert experience. At present, a relatively basic and linear processing method is mainly adopted. Generally, information from different sources, such as equipment manuals, maintenance records, expert rules, etc., is established into independent, isolated knowledge bases or databases. When facing a new fault diagnosis task, keyword matching or rule-based query is mainly used to search for historical cases or preset fault handling processes directly related to the current fault symptoms in the independent knowledge base, and finally a diagnosis scheme is formed. Effective integration of fault diagnosis knowledge is the core challenge to improve the accuracy and efficiency of diagnosis. SUMMARY

[0003] To overcome the shortcomings of the prior art, the purpose of the present application is to provide a diagnosis scheme generation method and system based on multi-scale hypergraph and constraint optimization, which can significantly improve the intelligent level and decision efficiency of complex system fault diagnosis.

[0004] To achieve the purpose of the application, in a first aspect, the embodiments of the present application provide a diagnosis scheme generation method based on multi-scale hypergraph and constraint optimization, which comprises: receiving fault diagnosis query information expressed in natural language, and performing semantic analysis on the fault diagnosis query information to form semantic guide information; extracting task nodes from a technical information set based on the semantic guide information, and constructing multi-scale nested subgraphs according to the relationships between the task nodes; based on the multi-scale nested subgraphs, dynamically reconstructing a task path, representing the task nodes as hypergraph nodes, and constructing a multi-scale hypergraph; performing structure-aware regularization processing on the multi-scale hypergraph to obtain a regularized multi-scale hypergraph; performing cross-scale attention processing on the regularized multi-scale hypergraph to obtain a multi-scale fusion hypergraph; generating a fault diagnosis path based on the multi-scale fusion hypergraph, sorting the fault diagnosis path, and outputting the fault diagnosis path ranked at the top as a diagnosis scheme.

[0005] In a possible implementation manner, the semantic analysis of the fault diagnosis query information to form semantic guide information comprises: Performing intent recognition and semantic decomposition on the fault diagnosis query information to obtain semantic representation information; Performing context semantic modeling and entity extraction processing on the semantic representation information to obtain structured information; Performing reasoning on the structured information to obtain potential constraint information, and the structured information and the potential constraint information form the semantic guidance information.

[0006] In a possible implementation, the method further includes: based on the semantic guidance information, extracting task nodes from a technical information set and constructing multi-scale nested subgraphs according to relationships between the task nodes, including: Based on the semantic guidance information and the technical information set, performing structured matching processing to extract task nodes, the task nodes including core diagnostic targets, related components, and key working condition features; Based on parent-child relationships and dependency relationships between the task nodes, constructing multi-scale nested subgraphs, wherein each multi-scale nested subgraph corresponds to a task structure at a different abstraction level.

[0007] In a possible implementation, based on the multi-scale nested subgraphs, dynamically reconstructing a task path, representing the task nodes as hypergraph nodes, and constructing a multi-scale hypergraph, including: According to relationships between the semantic guidance information and the task nodes in the multi-scale nested subgraphs, dynamically adjusting a node sequence of the task path to obtain a reconstructed task path; Mapping the task nodes in the multi-scale nested subgraphs to hypergraph nodes, and an attribute vector of the hypergraph node including semantic features, structural features, and constraint features; Based on the reconstructed task path and the hypergraph nodes, constructing a multi-scale hypergraph.

[0008] In a possible implementation, the method further includes: based on the multi-scale hypergraph, performing structure-aware regularization processing to obtain a regularized multi-scale hypergraph, including: Calculating semantic similarity scores of nodes and hyperedges of the multi-scale hypergraph, for each node, selecting hyperedges with high semantic similarity scores to construct a sparse connection matrix of the multi-scale hypergraph; Based on the sparse connection matrix, retaining hyperedges with connection strengths greater than or equal to a preset threshold to form a pruned sparse connection matrix; Performing logical verification and conflict reconciliation on the pruned sparse connection matrix by constructing constraint conditions to obtain the regularized multi-scale hypergraph.

[0009] In a possible implementation, the logical verification and conflict reconciliation of the pruned sparse connection matrix by constructing constraint conditions to obtain the regularized multi-scale hypergraph includes: constructing a structure constraint hypergraph, the structure constraint hypergraph including constraint conditions for nodes and hyperedges of the multi-scale hypergraph; modeling the structure constraint hypergraph by using a hypergraph neural network to learn node selection weights and hyperedge constraint tension, to obtain a structure consistency optimization model; performing logical verification and conflict reconciliation on the pruned sparse connection matrix based on the structure consistency optimization model, to obtain a regularized multi-scale hypergraph.

[0010] In a possible implementation, the cross-scale attention processing of the regularized multi-scale hypergraph to obtain a multi-scale fusion hypergraph includes: for each structure scale, calculating semantic attention weights between each node of the multi-scale hypergraph and hyperedges connected to the node, performing node information aggregation and updating node representation based on the semantic attention weights, to obtain a fused hypergraph node; performing linear mapping on hyperedges of the multi-scale hypergraph at each structure scale, and fusing hyperedges of different structure scales, to obtain a fused hypergraph hyperedge; splicing the fused hypergraph node and the fused hypergraph hyperedge, to obtain the multi-scale fusion hypergraph.

[0011] In a possible implementation, the method includes the following steps. based on the multi-scale fusion hypergraph, matching the fault diagnosis query information with hypergraph nodes of the multi-scale fusion hypergraph to generate candidate diagnosis paths; performing structure consistency and semantic rationality evaluation on each of the candidate diagnosis paths, and sorting the candidate diagnosis paths according to the evaluation results; outputting the candidate diagnosis paths with high ranking as diagnosis schemes.

[0012] To achieve the object of the application, in a second aspect, the embodiments of the present application provide a system, including a storage medium and one or more processors; the storage medium stores a computer program, and the computer program is called by the one or more processors to implement any one of the above-mentioned diagnosis scheme generation methods based on multi-scale hypergraph and constraint optimization.

[0013] The technical scheme provided by the embodiments of the present application brings at least the following beneficial effects: The present application aims at the core problem of low accuracy and efficiency of diagnostic scheme generation caused by the isolation of knowledge bases and the dependence on keyword matching in the prior art, and realizes a technical breakthrough through multi-scale hypergraph modeling and constraint consistency optimization. The technical effects include: through semantic analysis and cross-scale attention mechanism, deep fusion of heterogeneous knowledge, breaking the "knowledge island", realizing the qualitative change from linear retrieval to complex network reasoning; dynamically reconstructing and optimizing the diagnostic path to ensure the logical rigor and global optimality of the scheme generation. The scheme significantly improves the accuracy and efficiency of fault diagnosis, can quickly generate an optimal diagnostic path that is practical and feasible, effectively reduces the operation and maintenance cost, and greatly enhances the intelligent level of industrial equipment health management. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the technical schemes in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0015] Figure 1 is a schematic flowchart of a diagnostic scheme generation method based on multi-scale hypergraph and constraint optimization provided by the embodiments of the present application. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical scheme and advantages of the present application more clear, the embodiments of the present application will be further described in detail with reference to the drawings.

[0017] The embodiments of the present application provide a diagnostic scheme generation method based on multi-scale hypergraph and constraint optimization, referring to Figure 1 The method comprises the following steps: Step one: receiving fault diagnosis query information expressed by natural language, performing semantic analysis on the fault diagnosis query information to form semantic guidance information.

[0018] The user can express the fault diagnosis query information of the device by natural language, for example, describe the motor bearing vibration data, the environmental characteristics under the running condition, and the description and detection requirements for the fault, etc.

[0019] Optionally, the semantic analysis of the fault diagnosis query information to form semantic guidance information comprises the following steps: The semantic representation information is obtained by performing intent recognition and semantic decomposition on the fault diagnosis query information; the structured information is obtained by performing context semantic modeling and entity extraction processing on the semantic representation information; the potential constraint information is obtained by performing reasoning on the structured information, and the structured information and the potential constraint information form the semantic guidance information.

[0020] Among them, for the fault diagnosis query information of the user, the input can be subjected to intent recognition and semantic decomposition by using a pre-trained language model. Through context semantic modeling and entity extraction, the core diagnostic target, related components and key working condition characteristics are identified. Further, based on the extraction results, potential constraint information such as function sequence, state triggering, execution dependency, etc. is reasoned out, and it is constructed as a structured demand tensor and a potential constraint information set to form semantic representation information, providing semantic guidance for subsequent graph structure modeling.

[0021] This step can accurately convert fuzzy and unstructured natural language queries into structured semantic guidance information that can be understood and reasoned by machines. Through intent recognition and semantic decomposition, the user's core diagnostic target, related components and key working conditions are accurately captured to ensure the accuracy of the diagnostic direction. Through context reasoning, the implicit logical relationships and potential constraints in the description are excavated and made explicit, and structured information rich in logic is constructed. This not only provides a high-quality semantic basis for the construction of subsequent graph networks, but also enables the entire diagnostic system to intelligently guide based on a deep understanding of the user's intent, ensuring the accuracy and logical rigor of the generated diagnostic scheme from the source.

[0022] Step two: based on the semantic guidance information, task nodes are extracted from the technical information set, and multi-scale nested subgraphs are constructed according to the relationships between the task nodes.

[0023] Optionally, this step can include: based on the semantic guidance information and the technical information set, performing structured matching processing to extract task nodes, the task nodes including core diagnostic targets, related components and key working condition characteristics; based on the parent-child relationship and dependency relationship between the task nodes, multi-scale nested subgraphs are constructed, wherein each multi-scale nested subgraph corresponds to a task structure at a different abstraction level.

[0024] Among them, the technical information set can include multi-source heterogeneous data, including but not limited to: technical roadmap: describing the hierarchical structure and dependency relationship of task execution; fault tree analysis graph: representing fault modes and propagation paths; diagnostic manual and document: containing detailed description and constraint rules of task nodes; sensor data and topology graph: providing device running state and connection relationship between functional modules.

[0025] The matching of the semantic guidance information and the technical information set can be achieved by the following steps: The core diagnostic target, related components, and key working condition features in the semantic guidance information are encoded into vector representations using a pre-trained language model, and similarity scores are calculated with the task node vectors in the technical information set.

[0026] According to the similarity scores, task nodes related to the semantic guidance information are screened out. For example, if the query contains "abnormal vibration of motor bearing", the "bearing fault diagnosis" node in the technical information set is matched.

[0027] Align the potential constraint information with the constraint rules in the technical information set to ensure that the extracted task nodes meet the constraint rules.

[0028] After matching, the task nodes are extracted, which can include: core diagnostic target: such as "bearing fault diagnosis" "motor fault positioning"; related components: such as "bearing" "motor" "sensor"; Key working condition features: such as "high load" "impact" "sand dust". These nodes serve as the basis for subsequent multi-scale nested subgraph construction.

[0029] For the parent-child relationship and dependency relationship between the task nodes, the step of constructing a multi-scale nested subgraph is constructed. The core goal of this step is to organize the task nodes extracted in the previous step into multiple nested subgraph structures according to their inherent parent-child relationship and dependency relationship, each subgraph corresponding to a different abstract level of task structure. This process provides a structured foundation for subsequent dynamic reconstruction of task paths and construction of multi-scale hypergraphs.

[0030] This step can include: Based on the parent-child relationship between task nodes, the task nodes are divided into several abstract levels, where the top-level nodes represent the overall diagnostic target, the middle-level nodes represent sub-tasks or modules, and the bottom-level nodes represent specific operations or detection units, forming a hierarchical task structure.

[0031] Based on the dependency relationship between task nodes, the task nodes with logical association or constraint conditions are connected through the construction of hyperedges to form a dependency network, where each hyperedge represents the cooperative execution or constraint relationship between multiple task nodes, capturing the complex interaction between tasks.

[0032] Combine the hierarchical task structure with the dependency network to construct a multi-scale nested subgraph, where each nested subgraph corresponds to a task structure at an abstract level, and the nested subgraphs at different levels are connected through parent-child relationships and logical transmission across levels through hyperedges of dependency relationships, forming a multi-scale, nested, high-order associated task graph structure.

[0033] The parent-child relationship describes the hierarchical nested structure between task nodes and is the basis for building multi-scale nested subgraphs. The hierarchy can include a top layer, a middle layer, and a bottom layer. The top layer represents system-level or function-level tasks, such as "system fault diagnosis" and "device health management". These nodes are usually the entry point of the entire diagnostic task. The middle layer represents module-level or subsystem-level tasks, such as "motor fault positioning" and "sensor fault diagnosis". These nodes are subtasks of the top-level tasks. The bottom layer represents operation-level or atomic-level tasks, such as "bearing fault diagnosis" and "motor control unit fault diagnosis". These nodes are further decompositions of the middle-level tasks.

[0034] The parent-child relationship is mapped. The parent-child relationship is represented by a set of cross-layer connection edges. For example, "bearing fault diagnosis" is a child node of "motor fault positioning". "Motor fault positioning" is a child node of "system fault diagnosis".

[0035] The hierarchy is aggregated, and the fine-grained task nodes are aggregated into coarse-grained semantic units using aggregation functions. For example, "bearing fault diagnosis" and "motor control unit fault diagnosis" at the bottom layer are aggregated into "motor fault positioning" at the middle layer. "Motor fault positioning" and "sensor fault diagnosis" at the middle layer are aggregated into "system fault diagnosis" at the top layer.

[0036] Based on the dependency relationship between task nodes, a hyperedge is constructed to connect task nodes with logical associations or constraint conditions, and then a dependency network is formed. In this process, the "dependency relationship" refers to the logical relationship between task nodes in the execution process, such as the order, condition constraints, or data transmission. These relationships determine the coordination mode and execution order between task nodes. The "hyperedge" is a special data structure that can connect multiple task nodes and represent the collaborative execution or constraint relationship between these nodes. Each hyperedge not only records the connected task nodes, but also contains the specific dependency type and constraint conditions between these nodes. In building the dependency network, first, the dependency relationship between task nodes is analyzed and extracted, and then according to the extraction result, the task nodes with logical association or constraint condition are connected through the hyperedge. In this way, the originally scattered task nodes are organized into an ordered and constrained dependency network, so that the complex interaction between tasks can be more accurately captured.

[0037] Step three: based on the multi-scale nested subgraph, dynamically reconstruct the task path, represent the task nodes as hypergraph nodes, and build a multi-scale hypergraph.

[0038] Optionally, this step includes: According to the relationship between the semantic guidance information and the task nodes in the multi-scale nested subgraph, the node sequence of the task path is dynamically adjusted to obtain a reconstructed task path; the task nodes in the multi-scale nested subgraph are mapped into supergraph nodes, and the attribute vector of the supergraph node includes semantic features, structural features and constraint features; and a multi-scale supergraph is constructed based on the reconstructed task path and the supergraph nodes.

[0039] In an optional embodiment, according to the relationship between the semantic guidance information and the task nodes in the multi-scale nested subgraph, the node sequence of the task path is dynamically adjusted to obtain a reconstructed task path, which can include: Based on the semantic guidance information, the task nodes in the multi-scale nested subgraph are matched and weighted to determine the relevance of each node to the user query. Preferably, the semantic guidance information is matched with the task nodes in the multi-scale nested subgraph, and the relevance score of each node to the query is calculated. According to the score, each node is assigned a weight. This process not only improves the accuracy of task node screening, but also provides a quantitative basis for subsequent dynamic path adjustment.

[0040] The nodes are prioritized and the node execution order is dynamically adjusted in combination with the parent-child relationship and dependency relationship between the task nodes. Preferably, on the basis of node weight assignment, the system further combines the parent-child relationship and dependency relationship between the task nodes, and can perform sorting through the construction of a task priority sorting model. The model uses a topological sorting algorithm to ensure that the parent node is executed before the child node, while considering the influence of the dependency relationship on the node order. For example, if a node depends on the output result of another node, the dependent node must be placed in the pre-order position. By dynamically adjusting the node execution order, the system can generate a preliminary task path that is logically reasonable and efficient. This step combines static node relationships with dynamic priority adjustment, significantly improving the flexibility and adaptability of the task path.

[0041] The task path is reconstructed and optimized through a dynamic programming and feedback mechanism to generate an optimal execution sequence. Preferably, in order to further improve the execution efficiency and accuracy of the task path, a dynamic programming algorithm is introduced in this step to find the optimal execution sequence and minimize the execution time or resource consumption. At the same time, through a feedback mechanism, the path execution effect is monitored in real time, and if a node is found to be executing abnormally or inefficiently, the subsequent node order or the node is automatically adjusted or replaced. This dynamic reconstruction mechanism not only improves the robustness of the path, but also adapts to changes in task requirements in different scenarios, reflecting the adaptive ability of the system in complex industrial environments.

[0042] Preferably, the task nodes in the multi-scale nested subgraph are mapped into supergraph nodes, and the attribute vector of the supergraph node includes semantic features, structural features and constraint features, which can include: The task nodes in the multi-scale nested subgraph are mapped to the hypergraph nodes, and the original structure and semantic information are retained. For example, the parent nodes still maintain the control relationship to the child nodes in the hypergraph, and the dependency relationship is connected through the hyperedge. This mapping process not only simplifies the expression of complex subgraphs, but also lays the foundation for efficient calculation and reasoning of the hypergraph, embodying the innovation of multi-scale structure transformation into a hypergraph.

[0043] An attribute vector is constructed for each hypergraph node, including semantic features, structural features, and constraint features, to comprehensively represent the node characteristics.

[0044] The attribute vector expression of the hypergraph node is optimized through feature fusion and normalization processing. This step uses feature fusion and normalization technology to optimize the vector. Feature fusion effectively integrates semantic, structural, and constraint features through weighted averaging or neural network models to form a comprehensive feature vector. Normalization processing can eliminate the dimensional differences between different features, improving the consistency and comparability of the vector. The optimized attribute vector not only enhances the discrimination of the node in the hypergraph, but also improves the learning efficiency and reasoning accuracy of the subsequent graph neural network, embodying the important role of feature engineering in hypergraph construction.

[0045] Preferably, based on the reconstructed task path and the hypergraph node, a multi-scale hypergraph is constructed, which can include: According to the dependency relationship between the reconstructed task path and the hypergraph node, a hyperedge is constructed to connect nodes with cooperative or constraint relationship. The hyperedge not only connects two nodes, but also connects multiple nodes at the same time, forming a high-order dependency relationship.

[0046] The hypergraph nodes are divided into multi-scale levels, and the hypergraph nodes and hyperedges are optimized through a graph neural network to construct a multi-scale hypergraph. After the hyperedge is constructed, the system performs multi-scale level division on the hypergraph to form task structures at different abstraction levels. The nodes at different levels can be identified and optimized through a graph neural network to form a nested hypergraph structure. The graph neural network learns the features of the nodes and hyperedges on the hypergraph through a message passing mechanism, optimizing the structure and parameters of the hypergraph.

[0047] Step 4: Perform structure-aware regularization processing on the multi-scale hypergraph to obtain a regularized multi-scale hypergraph.

[0048] Preferably, this step can include: calculating semantic similarity scores of nodes and hyper-edges of the multi-scale hypergraph, for each node, selecting hyper-edges with top semantic similarity scores, and constructing a sparse connection matrix of the multi-scale hypergraph; based on the sparse connection matrix, retaining hyper-edges with connection strength greater than or equal to a preset threshold, to form a pruned sparse connection matrix; performing logical verification and conflict reconciliation on the pruned sparse connection matrix by constructing constraint conditions, to obtain the regularized multi-scale hypergraph.

[0049] Optionally, the calculating semantic similarity scores of nodes and hyper-edges of the multi-scale hypergraph, for each node, selecting hyper-edges with top semantic similarity scores, and constructing a sparse connection matrix of the multi-scale hypergraph can include: For each node, based on the semantic embedding vector of the node and the semantic embedding vectors of all hyper-edges, calculating the semantic similarity scores between the node and the hyper-edges; for each node, sorting the semantic similarity scores from high to low, and selecting a number of hyper-edges with top scores as candidate connection hyper-edges of the node; based on the correspondence between each node and its candidate connection hyper-edges, constructing a bipartite graph structure from nodes to hyper-edges, to generate an initial sparse connection matrix.

[0050] Optionally, the based on the sparse connection matrix, retaining hyper-edges with connection strength greater than or equal to a preset threshold, to form a pruned sparse connection matrix can include: Traversing each hyper-edge in the sparse connection matrix, calculating the connection strength of the hyper-edge, the connection strength being the weighted sum of the semantic similarity scores between the nodes and the hyper-edges; comparing the connection strength of each hyper-edge with a preset threshold, if the connection strength is greater than or equal to the threshold, retaining the hyper-edge, otherwise, eliminating the hyper-edge; updating the sparse connection matrix according to the retained hyper-edges, to form a pruned sparse connection matrix.

[0051] Optionally, the performing logical verification and conflict reconciliation on the pruned sparse connection matrix by constructing constraint conditions, to obtain the regularized multi-scale hypergraph can include: constructing a structure constraint hypergraph, the structure constraint hypergraph including constraint conditions for nodes and hyper-edges of the multi-scale hypergraph; modeling the structure constraint hypergraph using a hypergraph neural network, learning node selection weights and hyper-edge constraint tensions, to obtain a structure consistency optimization model; based on the structure consistency optimization model, performing logical verification and conflict reconciliation on the pruned sparse connection matrix, to obtain the regularized multi-scale hypergraph.

[0052] Optionally, the constructing a structure constraint hypergraph, the structure constraint hypergraph including constraint conditions for nodes and hyper-edges of the multi-scale hypergraph can include: Defining logical dependency between nodes, semantic coverage of hyper-edges, and topological constraint rules between nodes and hyper-edges based on domain knowledge or historical diagnosis data; Mapping the logical dependency, semantic coverage, and topological constraint rules into constraint edges in the hypergraph to form an initial structural constraint hypergraph; Performing redundancy constraint elimination and priority sorting on the initial structural constraint hypergraph to obtain a simplified structural constraint hypergraph.

[0053] By introducing domain knowledge and historical diagnosis data, the simplified structural constraint hypergraph constructed by the scheme provides high signal-to-noise ratio input for subsequent modeling, ensuring that the system has stronger reasoning ability and robustness in complex diagnosis scenarios.

[0054] The modeling of the structural constraint hypergraph using the hypergraph neural network to learn node selection weights and hyper-edge constraint tension to obtain a structural consistency optimization model can include: Inputting the structural constraint hypergraph into the hypergraph neural network to encode the interaction between nodes and hyper-edges through a message passing mechanism to generate joint embedding representations of nodes and hyper-edges; Based on the joint embedding representations, calculating the selection weight of each node and the constraint tension of each hyper-edge, wherein the node selection weight reflects its importance in the diagnosis path, and the hyper-edge constraint tension reflects its constraint strength on node connection; Jointly optimizing the node selection weight and the hyper-edge constraint tension using a loss function to obtain a structural consistency optimization model for guiding subsequent logical verification and conflict reconciliation.

[0055] The structural consistency optimization model obtained by the scheme not only improves the modeling accuracy, but also provides a reliable mathematical foundation for subsequent logical verification and conflict reconciliation, significantly enhancing the generalization ability and stability of the model.

[0056] The logical verification and conflict reconciliation of the pruned sparse connection matrix based on the structural consistency optimization model to obtain a regularized multi-scale hypergraph can include: Using the structural consistency optimization model to perform logical verification on the pruned sparse connection matrix to identify node-to-hyper-edge connection pairs that violate constraint conditions; Performing conflict reconciliation on the identified conflict connection pairs, including reallocating node selection weights, adjusting hyper-edge constraint tension, or eliminating redundant connections to eliminate logical conflicts; Updating the sparse connection matrix according to the reconciliation result and outputting a regularized multi-scale hypergraph.

[0057] This solution, based on a structural consistency optimization model, performs logical verification and conflict reconciliation on the pruned sparse connection matrix, efficiently identifying and handling node-to-hyperedge connection pairs that violate constraints. By reallocating node weights, adjusting hyperedge constraint tension, or eliminating redundant connections, the system effectively eliminates logical conflicts, ensuring the structural consistency and semantic rationality of the multi-scale hypergraph. The final output regularized multi-scale hypergraph not only improves the accuracy and reliability of inference paths but also provides high-quality structured support for intelligent decision-making in complex scenarios, possessing significant engineering application value.

[0058] In a preferred embodiment, during the construction of the structural constraint hypergraph, this application utilizes a structure-aware constraint regularization mechanism to enhance the model's ability to cluster similar semantic structures and to distinguish structural conflicts in paths. This mechanism first defines the logical dependencies between nodes, the semantic coverage of hyperedges, and the topological constraint rules between nodes and hyperedges based on domain knowledge or historical diagnostic data, mapping these to constraint edges in the hypergraph to form an initial structural constraint hypergraph. Subsequently, the initial structural constraint hypergraph is simplified through redundant constraint removal and priority ranking mechanisms.

[0059] When constructing a structurally constrained hypergraph, node consistency constraints are used to ensure semantic consistency between nodes and hyperedges. The node consistency constraint is defined as follows: , in, This represents node consistency constraints. Let represent the embedding representation of the i-th task node in layer s. This represents the embedding representation of the j-th superedge. Indicates the relationship with the hyperedge The set of all connected nodes This represents the total number of superedges in the s-th layer. This represents semantic distance. This constraint encourages task nodes connected by the same hyperedge to have similar semantic representations, preventing semantically heterogeneous nodes from being incorrectly aggregated.

[0060] In logic verification and conflict reconciliation, hyperedge heterogeneity constraints are used to enhance the similarity and differences between hyperedges. The hyperedge heterogeneity constraint is defined as follows: , in, For hyperedge heterogeneity constraints, This represents the total number of hyperedges at the s-th scale. Used to represent the relationship between hyperedge i and hyperedge j; It is the Euclidean distance between hyperedge i and hyperedge j at the s-th scale; is a truncated linear function to limit the upper bound of the distance difference, is a threshold value.

[0061] In the structure self-updating regularization term construction stage, a feedback adjustment mechanism is introduced to dynamically adjust the connection weights of nodes and hyper-edges to adapt to the changes in data distribution. The specific steps include: first, based on the historical connection state and the current input data, the connection stability score of nodes and hyper-edges is calculated; second, using the connection stability score, a structure self-updating regularization term is constructed to constrain the change amplitude of the connection weights of nodes and hyper-edges; finally, the structure self-updating regularization term is combined with the node constraint term and the hyper-edge constraint term to form a structure consistency optimization model to prevent frequent oscillation of the graph structure.

[0062] Step five: performing cross-scale attention processing on the regularized multi-scale hypergraph to obtain a multi-scale fusion hypergraph.

[0063] Optionally, this step can include: For each structural scale, calculating the semantic attention weight between each node of the multi-scale hypergraph and the hyper-edge connected to the node, aggregating node information based on the semantic attention weight and updating the node representation to obtain a fused hypergraph node; Performing linear mapping on the hyper-edges of the multi-scale hypergraph under each structural scale to fuse the hyper-edges of different structural scales to obtain a fused hypergraph hyper-edge; Splicing the fused hypergraph node and the fused hypergraph hyper-edge to obtain the multi-scale fusion hypergraph.

[0064] This scheme realizes the deep fusion of nodes and hyper-edges in the multi-scale hypergraph through cross-scale attention processing, and the specific technical effects can include: through the attention mechanism, the node can adaptively capture the semantic association between the hyper-edges, enhancing the modeling ability of complex structural relationships; through linear mapping and weighted fusion, the hyper-edge information of different structural scales is effectively integrated, eliminating the semantic deviation between scales and improving the generalization ability of the model; through the splicing and integration of nodes and hyper-edges, a multi-scale fusion hypergraph with complete structure and consistent semantics is formed, providing high-quality structured input for subsequent tasks.

[0065] Wherein, for each structural scale, the semantic attention weight between each node of the multi-scale hypergraph and the hyper-edge connected to the node is calculated, the node information is aggregated based on the semantic attention weight, and the node representation is updated to obtain a fused hypergraph node, which can include: For each structural scale, initialize the attention weight calculation parameters between nodes and hyper-edges.

[0066] Based on the node embedding vector and the hyperedge feature vector, the semantic correlation score between each node and the hyperedge connected thereto is calculated by using dot product attention or multi-head attention mechanism, and the semantic attention weight is obtained through normalization processing.

[0067] The semantic attention weight is calculated, the hyperedge information connected to the node is weighted and aggregated, the aggregated information is spliced or weighted with the original representation of the node, the node representation is updated through a nonlinear activation function, and the fused hypergraph node is obtained.

[0068] The linear mapping model is constructed for the hyperedge of the multi-scale hypergraph at each structural scale, and the high-dimensional hyperedge feature is mapped to a unified low-dimensional semantic space.

[0069] The linear mapping model is constructed for the hyperedge of the multi-scale hypergraph at each structural scale, and the high-dimensional hyperedge feature is mapped to a unified low-dimensional semantic space.

[0070] The linear mapping model is constructed for the hyperedge of the multi-scale hypergraph at each structural scale, and the high-dimensional hyperedge feature is mapped to a unified low-dimensional semantic space.

[0071] The linear mapping model is constructed for the hyperedge of the multi-scale hypergraph at each structural scale, and the high-dimensional hyperedge feature is mapped to a unified low-dimensional semantic space.

[0072] The linear mapping model is constructed for the hyperedge of the multi-scale hypergraph at each structural scale, and the high-dimensional hyperedge feature is mapped to a unified low-dimensional semantic space. The linear mapping model is constructed for the hyperedge of the multi-scale hypergraph at each structural scale, and the high-dimensional hyperedge feature is mapped to a unified low-dimensional semantic space.

[0073] Step six: generating a fault diagnosis path based on the multi-scale fusion hypergraph, and sorting the fault diagnosis path, and outputting the top-ranked fault diagnosis path as the diagnosis scheme.

[0074] Preferably, the step can include: The fault diagnosis query information is matched with the supergraph nodes of the multi-scale fusion supergraph based on the multi-scale fusion supergraph, and candidate diagnosis paths are generated. Optionally, the input fault diagnosis query information is semantically parsed, key semantic features are extracted, and vectorized representations are obtained. The semantic similarity between the query information vectorized representation and each node in the multi-scale fusion supergraph is calculated, and nodes with a similarity higher than a preset threshold are selected as starting nodes of candidate paths. Starting from the starting nodes, path expansion is performed according to the superedge connection relationship, and multiple possible candidate diagnosis paths are generated. Each path contains a plurality of nodes and connection relationships, forming a complete diagnosis reasoning chain.

[0075] The structure consistency and semantic rationality of each candidate diagnosis path are evaluated, and the candidate diagnosis paths are sorted according to the evaluation results. Optionally, the structure topology of each candidate diagnosis path is analyzed to determine whether the connection between the nodes and the superedges conforms to the preset structure constraint rules, ensuring the structural rationality of the path. The semantic coherence and logical rationality of each candidate path are evaluated in combination with the node semantic features and path context information, to avoid semantic conflicts or logical breaks. According to the evaluation results of the structure consistency and semantic rationality, a comprehensive score is assigned to each candidate path, and the paths are sorted in descending order of the score. A plurality of paths with high rankings are selected.

[0076] The candidate diagnosis paths with high rankings are output as diagnosis schemes. Optionally, according to the sorting results, a plurality of candidate paths with the highest comprehensive scores are selected as optimal diagnosis paths, and are confirmed manually or automatically. The confirmed optimal diagnosis paths are output in a structured form. Optionally, feedback analysis is performed on the output diagnosis schemes, actual diagnosis effects are combined to provide optimization suggestions, which are used for iterative improvement of subsequent models and improvement of diagnosis accuracy.

[0077] The scheme generates and sorts fault diagnosis paths based on a multi-scale fusion supergraph, achieving efficient construction and accurate output of diagnosis paths. The candidate paths are quickly generated using semantic matching and superedge connection relationships, and dual evaluation is performed in combination with structure consistency and semantic rationality, ensuring that the diagnosis schemes are both structurally reasonable and semantically coherent, significantly improving the accuracy and explainability of diagnosis. The method supports intelligent decision-making in complex scenarios and provides reliable and understandable high-quality diagnosis schemes for users. At the same time, a closed-loop mechanism is formed through feedback and optimization suggestions, promoting continuous iterative optimization of the model and enhancing the adaptability and robustness of the system in actual applications.

[0078] The present application realizes the intelligent processing of the whole process from the natural language fault query to the accurate diagnosis scheme. First, the user query is converted into structured semantic guidance information through semantic analysis, effectively capturing the intention and constraints; second, the task path is dynamically constructed based on multi-scale nested sub-graphs, and the complex task nodes and their relationships are expressed in multiple dimensions using hypergraph modeling technology, which improves the model's representation ability of complex systems. Subsequently, through structure perception regularization and cross-scale attention mechanism, the hypergraph structure and semantic fusion are optimized to ensure the structural consistency and semantic rationality of the diagnosis path. Finally, based on the multi-scale fusion hypergraph, the fault diagnosis path is generated and sorted, and the high-quality diagnosis scheme is output. The method significantly improves the accuracy, robustness and explainability of diagnosis, is suitable for intelligent decision support in complex industrial scenarios, and has good scalability and self-optimization ability.

[0079] In another aspect, the embodiments of the present application provide a system, comprising a storage medium and one or more processors; the storage medium stores a computer program, and the computer program is called by the one or more processors to implement any of the above methods.

[0080] In an exemplary embodiment, a computer-readable storage medium is also provided, the storage medium storing at least one program code, the at least one program code being loaded and executed by a processor to enable a computer to implement any of the above methods.

[0081] Optionally, the above computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0082] In an exemplary embodiment, a computer program or computer program product is also provided, the computer program or computer program product storing at least one computer instruction, the at least one computer instruction being loaded and executed by a processor to enable a computer to implement any of the above methods.

[0083] It should be noted that all actions of obtaining signals, information or data in the present application are carried out in accordance with the data protection regulation policy of the country where the device is located and with the authorization given by the owner of the corresponding device.

[0084] It should be understood that the "multiple" mentioned herein refers to two or more than two. The "and / or" describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent: A exists alone, A and B exist together, and B exists alone. The character " / " generally represents that the front and rear associated objects are in an "or" relationship.

[0085] The above-mentioned sequence numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0086] The above-mentioned only for the exemplary embodiments of the present application, and does not limit the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application, should be included in the protection scope of the present application.

Claims

1. A method for diagnosis protocol generation based on multi-scale hypergraphs and constrained optimization, characterized in that, The method comprises: receiving fault diagnosis query information expressed in natural language, performing semantic analysis on the fault diagnosis query information to form semantic guidance information; based on the semantic guidance information, extracting task nodes from a technical information set, and constructing multi-scale nested subgraphs according to the relationships between the task nodes; based on the multi-scale nested subgraphs, dynamically reconstructing task paths, representing the task nodes as hypergraph nodes, and constructing multi-scale hypergraphs; performing structure-aware regularization processing on the multi-scale hypergraphs to obtain regularized multi-scale hypergraphs; performing cross-scale attention processing on the regularized multi-scale hypergraphs to obtain multi-scale fusion hypergraphs; based on the multi-scale fusion hypergraphs, generating fault diagnosis paths, and sorting the fault diagnosis paths, and outputting the fault diagnosis paths ranked at the top as diagnosis schemes.

2. The method for generating a diagnosis protocol based on multi-scale hypergraphs and constrained optimization according to claim 1, characterized in that, The semantic analysis of the fault diagnosis query information to form semantic guidance information comprises: performing intent recognition and semantic decomposition on the fault diagnosis query information to obtain semantic representation information; performing context semantic modeling and entity extraction processing on the semantic representation information to obtain structured information; performing reasoning on the structured information to obtain potential constraint information, and the structured information and the potential constraint information form the semantic guidance information.

3. The method for diagnosis protocol generation based on multi-scale hypergraphs and constrained optimization of claim 1, wherein, Based on the semantic guidance information, task nodes are extracted from a technical information set, and multi-scale nested subgraphs are constructed according to the relationships between the task nodes, comprising: based on the semantic guidance information and the structured matching processing of the technical information set, the task nodes are extracted, the task nodes include core diagnosis targets, related components and key working condition characteristics; based on the parent-child relationship and the dependency relationship between the task nodes, multi-scale nested subgraphs are constructed, wherein each multi-scale nested subgraph corresponds to a task structure at a different abstraction level.

4. The method of claim 3, wherein, Based on the multi-scale nested subgraphs, the task paths are dynamically reconstructed, the task nodes are represented as hypergraph nodes, and multi-scale hypergraphs are constructed, comprising: based on the relationship between the semantic guidance information and the task nodes in the multi-scale nested subgraphs, the node sequence of the task path is dynamically adjusted to obtain a reconstructed task path; the task nodes in the multi-scale nested subgraphs are mapped to hypergraph nodes, the attribute vectors of the hypergraph nodes include semantic features, structural features and constraint features; based on the reconstructed task path and the hypergraph nodes, a multi-scale hypergraph is constructed.

5. The method for generating a diagnostic protocol based on multi-scale hypergraphs and constrained optimization according to any one of claims 1 to 4, characterized in that, The structure-aware regularization processing on the multi-scale hypergraphs to obtain the regularized multi-scale hypergraphs comprises: calculating the semantic similarity scores of the nodes and hyperedges of the multi-scale hypergraph, for each node, selecting the hyperedges with the highest semantic similarity scores to construct a sparse connection matrix of the multi-scale hypergraph; based on the sparse connection matrix, retaining hyperedges with connection strength greater than or equal to a preset threshold to form a pruned sparse connection matrix; performing logical verification and conflict reconciliation on the pruned sparse connection matrix by constructing constraint conditions to obtain the regularized multi-scale hypergraph.

6. The method for generating a diagnostic protocol based on multi-scale hypergraphs and constrained optimization according to claim 5, characterized in that, The logical verification and conflict reconciliation of the pruned sparse connection matrix by constructing constraint conditions to obtain the regularized multi-scale hypergraph, comprising: Constructing a structural constraint hypergraph, the structural constraint hypergraph including constraint conditions for nodes and hyperedges of the multi-scale hypergraph; Modeling the structural constraint hypergraph using a hypergraph neural network to learn node selection weights and hyperedge constraint tension, obtaining a structural consistency optimization model; Based on the structural consistency optimization model, the logical verification and conflict reconciliation of the pruned sparse connection matrix are performed to obtain a regularized multi-scale hypergraph.

7. The method for generating a diagnosis protocol based on multi-scale hypergraphs and constrained optimization according to claim 5, characterized in that, The cross-scale attention processing of the regularized multi-scale hypergraph to obtain a multi-scale fusion hypergraph, comprising: For each structural scale, calculate the semantic attention weight between each node of the multi-scale hypergraph and the hyperedge connected to the node, aggregate node information based on the semantic attention weight and update the node representation to obtain a fused hypergraph node; Linearly mapping the hyperedge of the multi-scale hypergraph under each structural scale to fuse the hyperedges of different structural scales to obtain a fused hypergraph hyperedge; Splicing the fused hypergraph node and the fused hypergraph hyperedge to obtain the multi-scale fusion hypergraph.

8. The method for diagnosis protocol generation based on multi-scale hypergraphs and constrained optimization of claim 1, wherein, The multi-scale fusion hypergraph is used to generate a fault diagnosis path, and the fault diagnosis path is sorted, and the top-ranked fault diagnosis path is output as a diagnosis scheme, comprising: Based on the multi-scale fusion hypergraph, the fault diagnosis query information is matched with the hypergraph node of the multi-scale fusion hypergraph to generate a candidate diagnosis path; Each of the candidate diagnosis paths is evaluated for structural consistency and semantic rationality, and the candidate diagnosis paths are sorted according to the evaluation results; The candidate diagnosis path with high ranking is output as a diagnosis scheme.

9. A system comprising a storage medium and one or more processors; the storage medium stores a computer program, characterized in that, The computer program is called by one or more processors to implement the diagnosis scheme generation method based on multi-scale hypergraph and constraint optimization according to any one of claims 1 to 8.

Images