Intelligent planning and aided design method for ship equipment maintenance support task

Abstract

In order to solve the defects in the prior art, the invention provides a ship equipment maintenance support task intelligent planning and aided design method which comprises the following steps: constructing a ship equipment-oriented maintenance support task element system which comprises two aspects of a maintenance support operation sequence and a maintenance support operation attribute; based on a maintenance support task element system oriented to ship equipment, ship equipment maintenance technology documents are collected and marked, and a maintenance support task element graph is constructed. Based on the maintenance support task element graph, nodes of the maintenance support task element graph are subjected to data preprocessing and then serve as learning samples of the graph convolutional neural network to be input into the graph convolutional neural network for model training. And performing intelligent planning and aided design on the compilation of the ship equipment maintenance support task based on the trained model. According to the method, based on the decision normal form of knowledge driving and data driving fusion, the time sequence compliance and the resource adaptability of the operation process are effectively ensured, and technical support is provided for full-life-cycle guarantee of ship equipment.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of ship maintenance, and particularly relates to a method for intelligent planning and auxiliary design of ship equipment maintenance support tasks. BACKGROUND

[0002] The current ship equipment system presents a highly integrated feature, and various ship systems are composed of a composite architecture of multiple levels of heterogeneous components. This complex equipment system exposes significant flow coordination problems in maintenance operations, that is, the repair of a single device often needs to cross multiple levels of disassembly and reorganization operations of subsystems. Since there is a strict logical dependency relationship in the ship maintenance process, the conventional manual arrangement method is prone to process logic breakage or operation node misplacement, which seriously affects the efficiency of maintenance procedure compilation and execution reliability.

[0003] From the maintenance engineering implementation dimension, the ship equipment support activity has a multi-dimensional coupling feature: in the time sequence constraint layer, each operation unit needs to follow a strict topological execution order. In the resource coordination layer, it involves dynamic matching of multiple technical teams, special detection instruments, heterogeneous support materials and other elements. The combination optimization problem under the dual constraints of time and space and resources has exceeded the ability boundary of traditional manual experience decision-making, and there are systematic risks such as improper resource allocation and frequent process conflicts, which seriously restrict the improvement of the efficiency of comprehensive support of marine equipment. SUMMARY

[0004] The application provides a method for intelligent planning and auxiliary design of ship equipment maintenance support tasks to solve the problems in the prior art, comprising the following steps: A ship equipment-oriented maintenance support task element system is constructed, including two aspects of maintenance support operation sequence and maintenance support operation attribute.

[0005] Based on the ship equipment-oriented maintenance support task element system, ship equipment maintenance technical documents are collected and labeled, and a maintenance support task element graph is constructed.

[0006] Based on the maintenance support task element graph, the nodes of the maintenance support task element graph are preprocessed as learning samples of a graph convolutional neural network, and input into the graph convolutional neural network for model training.

[0007] Based on the trained model, the compilation of the ship equipment maintenance support task is intelligently planned and designed.

[0008] Further, the maintenance support operation sequence is a maintenance support operation step contained in the maintenance task and having an irreversible sequential relationship, and the maintenance support operation attribute is attribute information possessed by the maintenance support operation step.

[0009] Further, the maintenance support operation attribute includes: Maintenance personnel professional attribute, the maintenance personnel professional attribute refers to: the requirement for the professional skill of the maintenance personnel in the maintenance support operation step.

[0010] Maintenance personnel skill level attribute, the maintenance personnel skill level attribute refers to: the requirement for the professional skill title or level of the maintenance personnel in the maintenance support operation step.

[0011] Support equipment / tool attribute, the support equipment / tool attribute refers to: the support equipment or tool required in the maintenance support operation step.

[0012] Spare parts and consumables, spare parts and consumables refer to: spare parts and consumables required in the maintenance support operation step.

[0013] Further, the ship equipment maintenance support task element system based on the ship equipment, the method for collecting and labeling the ship equipment maintenance technical document comprises: Collecting the ship equipment maintenance technical document.

[0014] The ship equipment maintenance support task element system based on the ship equipment labels the information of the ship equipment from the six dimensions of "maintenance support operation", "maintenance support operation step", "maintenance personnel professional", "maintenance personnel skill level", "support equipment / tool", and "spare parts and consumables".

[0015] According to the original technical document order, arrange from top to bottom in turn.

[0016] Further, the method for constructing the maintenance support task element graph comprises: based on the labeled ship equipment maintenance technical document, constructing a directed graph composed of nodes and edges, and obtaining the maintenance support task element graph by connecting the directed graph based on the association relationship.

[0017] Further, the node of the directed graph represents the maintenance information.

[0018] The edge of the directed graph represents the association between the maintenance information, that is, the sequence relationship of two adjacent maintenance support operations in a maintenance task.

[0019] Further, the method for data preprocessing the nodes of the maintenance support task element graph comprises: Embedding the text information of each node in the maintenance support task element graph into the word vector feature, so as to convert the text of each node in the maintenance support task element graph into a word vector.

[0020] Based on the converted word vector data, construct the index mapping information of the node for each node and assign a digital identifier, construct the index mapping information of the relationship for the relationship between the nodes and create the mapping from the source node to the target node.

[0021] Based on the processed: node, index mapping information of node, relationship between nodes, index mapping information of relationship to create feature data set.

[0022] The feature data set is divided into training set, validation set and test set according to the proportion of 7:2:1, which is used as the input information of the subsequent model.

[0023] Further, the graph convolutional neural network comprises: Input layer: receiving node features and relationship information between nodes.

[0024] The first graph convolutional layer: the information of the input layer is converted for the first time, and the activation function is accessed.

[0025] The second graph convolutional layer: further processes the data output by the first graph convolutional layer, and keeps the data dimension unchanged.

[0026] The output layer includes: link prediction branch and attribute prediction branch.

[0027] Further, the link prediction branch is: the features of two nodes are spliced, and are mapped to 1-dimensional output through a linear layer, representing the probability of the existence of the relationship between adjacent nodes.

[0028] The attribute prediction branch is: the node data output by the second graph convolutional layer is directly mapped to a 4-dimensional output through a linear layer, representing the prediction probability of 4 maintenance attributes. The 4 maintenance attributes are: maintenance personnel professional attribute, maintenance personnel skill level attribute, support equipment / tool attribute, spare parts and consumables attribute.

[0029] Further, given a maintenance support task element graph G=(V, E, A, R), wherein V is a node set, E is an edge set, A is a node type set, and R is an edge type set. For any node v∈V, its type is denoted as τ(v)∈A. For any edge e=(s, t)∈E, its type is denoted as At this time, the graph convolutional neural network comprises: Input layer: for the node types in the maintenance support task element graph, a type-specific linear transformation layer set is constructed, and a separate linear mapping function is defined for each input node type τ using formula one: Formula one In formula one: is the initial text embedding feature of the node. is a type-specific weight matrix. is a type-specific bias vector. is the mapped unified dimension hidden feature.

[0030] The graph convolution layer includes two layers, and the calculation process can be formalized as follows: Equation II In Equation II, l represents the number of network layers. represents the type of target node v a set of neighbor nodes of is the attention weight calculated based on the edge type is the three-parameter matrix dependent on the source node type, target node type, and edge type. is the ReLU activation function.

[0031] The attribute prediction branch of the output layer includes: For any attribute node , the feature vector is obtained by encoding the node text description by the bidirectional encoder module, and the output dimension is Equation III In Equation III, the embedding of the given repair step node is then predicted by a multi-classification decoder as shown in Equation IV: Equation IV In Equation IV, is the predicted probability distribution vector of the four attributes of the given repair step node, is the weight matrix of the given repair step node, is the bias vector of the given repair step node, is a four-dimensional vector space.

[0032] The link prediction branch of the output layer includes: First, the final embedding representation of two nodes u and v is given as where u is the source node type and v is the target node type, is a 128-dimensional vector space.

[0033] Then, based on Equation V, the two are concatenated to form a joint representation: Equation V In Equation V, is a 256-dimensional vector space.

[0034] Finally, it is mapped to a scalar score by the activation function layer, and by executing Equation VI by the autoregressive decoder, the next task sequence text of the repair task can be predicted and generated: Equation VI​ wherein is the predicted next task sequence text, is the link weight matrix, is the link bias vector.

[0035] Further, the method for intelligent planning and auxiliary design of the ship equipment maintenance support task based on the trained model comprises: maintenance support task element completion and unknown maintenance support task element prediction.

[0036] The maintenance support task element completion comprises: inputting the task list as a seed node into the trained model.

[0037] The trained model performs contrastive analysis on the related information in the task list from two aspects of maintenance support operation sequence and maintenance support operation attribute, and performs element completion on the missing part and prediction probability distribution on the supplemented element information.

[0038] The supplemented element information is normalized, and the element information with the maximum probability is selected as the prediction result to complete the task list.

[0039] The unknown maintenance support task element prediction comprises: inputting the current known maintenance support operation step as a seed node into the trained model.

[0040] The trained model obtains possible steps for the next maintenance support operation as predicted steps, and predicts the probability of each predicted step as the next maintenance support operation step.

[0041] When there is a predicted step with a prediction probability greater than the threshold value, it is determined that there is a maintenance sequence association, and the predicted step is output.

[0042] The obtained predicted step is inputted as a seed node into the trained model, the above process is repeated, a maintenance task sequence is generated, and the joint probability of the whole path is calculated, until the link prediction probability is lower than a set threshold value and / or the reasoning step number reaches a preset upper limit, so as to complete the unknown maintenance support task element prediction.

[0043] The application has the advantages that the application proposes a new artificial intelligence method fusing field knowledge representation and time sequence reasoning aiming at the multi-dimensional complexity problem of ship maintenance engineering. By constructing a ship maintenance ontology knowledge base, core elements such as equipment disassembly logic and process constraints can be structuredly represented. In combination with graph embedding technology, the time and space association between maintenance elements is deeply modeled, and an intelligent deduction mechanism of the maintenance process chain can be established. This decision paradigm based on the fusion of knowledge-driven and data-driven can provide an end-to-end optimization path for maintenance scheme generation, effectively ensure the time sequence compliance and resource adaptability of the operation process, and provide technical support for the whole life cycle support of modernized ship equipment. BRIEF DESCRIPTION OF DRAWINGS

[0044] The drawings constituting a part of the specification of the application serve to provide a further understanding of the application, and the illustrative embodiments of the application and their description serve to explain the application, and do not constitute an improper limitation on the application. Obviously, for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0045] Figure 1 A basic flowchart of the model constructed for the embodiments of the application; Figure 2 Loss change graph (a) and training process accuracy change graph (b) of the HG-BART model training process constructed for the embodiments of the application; Figure 3 A maintenance manual information-replacement board card maintenance support operation execution step schematic diagram for the embodiments of the application; Figure 4 A specific structure schematic diagram of a model for the embodiments of the application. DETAILED DESCRIPTION

[0046] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as generally understood by those skilled in the art to which the application belongs.

[0047] The application exemplarily provides a ship equipment maintenance support task intelligent planning and auxiliary design method, comprising: A ship equipment-oriented maintenance support task element system is constructed, including two aspects of maintenance support operation sequence and maintenance support operation attribute.

[0048] Based on the ship equipment-oriented maintenance support task element system, ship equipment maintenance technical documents are collected and labeled, and a maintenance support task element graph is constructed.

[0049] Based on the maintenance support task element diagram, the nodes of the maintenance support task element diagram are data preprocessed and input into a graph convolutional neural network as learning sample of the graph convolutional neural network for model training.

[0050] Based on the trained model, intelligent planning and auxiliary design are performed on the compilation of the ship equipment maintenance support task.

[0051] The present application proposes a new artificial intelligence method that integrates field knowledge representation and time sequence reasoning for the multi-dimensional complexity problem of ship maintenance engineering. By constructing a ship maintenance ontology knowledge base, core elements such as equipment disassembly logic and process constraints can be structured. Combined with graph embedding technology, the spatio-temporal association between maintenance elements is deeply modeled, and an intelligent deduction mechanism of the maintenance process chain can be established. This decision-making paradigm based on the integration of knowledge-driven and data-driven can provide an end-to-end optimization path for maintenance scheme generation, effectively ensuring the time sequence compliance and resource adaptability of the operation process, and providing technical support for the whole life cycle support of modernized ship equipment.

[0052] The present application exemplarily provides a maintenance support operation sequence, which is a maintenance support operation step contained in a maintenance task and having an irreversible sequential relationship.

[0053] The present application exemplarily provides a maintenance support operation attribute, which is attribute information possessed in a maintenance support operation step.

[0054] The present application exemplarily provides specific contents of a maintenance support operation attribute, including: Maintenance personnel professional attribute, the maintenance personnel professional attribute refers to the requirement of professional skill of maintenance personnel in the maintenance support operation step. For example, mechanical professional, electronic professional, etc.

[0055] Maintenance personnel skill level attribute, the maintenance personnel skill level attribute refers to the requirement of professional skill title or level of maintenance personnel in the maintenance support operation step. For example, primary, intermediate, advanced.

[0056] Support equipment / tool attribute, the support equipment / tool attribute refers to the required support equipment or tool in the maintenance support operation step. For example, cross screwdriver, inner hexagonal wrench, explosion-proof open wrench, etc.

[0057] Spare parts and consumables, spare parts and consumables refer to spare parts and consumables required in the maintenance support operation step. For example, spare board card, stop washer, alcohol cotton ball, etc.

[0058] The application provides a maintenance support task analysis system for ship equipment maintenance activities according to relevant national (military) standards and in combination with expert experience.

[0059] The application exemplarily provides a ship equipment maintenance support task element system, and a method for collecting and labeling ship equipment maintenance technical documents, which comprises the following steps of: Collecting ship equipment maintenance technical documents.

[0060] The ship equipment maintenance technical documents are labeled based on the ship equipment maintenance support task element system from the information of six dimensions, i.e., maintenance support operation, maintenance support operation step, maintenance personnel specialty, maintenance personnel skill level, support equipment / tool, and spare parts and consumables.

[0061] The original technical documents are arranged in order from top to bottom.

[0062] The application exemplarily provides a method for constructing the maintenance support task element graph, which comprises the following steps of: based on the labeled ship equipment maintenance technical documents, constructing a directed graph composed of nodes and edges, and obtaining the maintenance support task element graph by connecting the directed graph based on the association relationship.

[0063] The application exemplarily provides a node of the directed graph, which represents maintenance information.

[0064] The application exemplarily provides an edge of the directed graph, which represents the association between maintenance information, i.e., the sequence relationship between two adjacent maintenance support operations in a maintenance task.

[0065] Compared with the prior art, by constructing the maintenance support task element graph covering the two dimensions of maintenance support operation sequence and attribute, the time sequence constraint and resource matching demand in the maintenance activity are expressed formally, and the end-to-end association modeling of the maintenance link and the attribute element is realized based on the graph data training model, thereby laying a knowledge and model foundation for the maintenance task planning.

[0066] The application exemplarily provides a method for data preprocessing of the nodes of the maintenance support task element graph, which comprises the following steps of: The text information of each node in the maintenance support task element graph is embedded into a word vector feature, so that the text of each node of the maintenance support task element graph is converted into a word vector.

[0067] Based on the converted word vector data, index mapping information of each node is constructed and a digital identifier is assigned, and index mapping information of the relationship between nodes is constructed and a mapping from the source node to the target node is created.

[0068] Based on the processed: node, index mapping information of the node, relationship between nodes, index mapping information of the relationship, a feature data set is created.

[0069] The feature data set is divided into a training set, a validation set and a test set according to a ratio of 7:2:1, which is used as input information for subsequent models.

[0070] The application exemplarily provides a graph convolutional neural network, comprising: The input layer receives node features and relationship information between nodes.

[0071] The first graph convolutional layer performs a first dimension conversion on the information of the input layer and inputs an activation function.

[0072] The second graph convolutional layer further processes the data output by the first graph convolutional layer and keeps the data dimension unchanged.

[0073] The output layer includes a link prediction branch and an attribute prediction branch.

[0074] The application exemplarily provides a link prediction branch, which splices the features of two nodes, maps them to a 1-dimensional output through a linear layer, and represents the probability of the existence of the relationship between adjacent nodes.

[0075] The application exemplarily provides an attribute prediction branch, which directly maps the node data output by the second graph convolutional layer to a 4-dimensional output through a linear layer, and represents the prediction probability of four maintenance attributes. The four maintenance attributes are: maintenance personnel professional attribute, maintenance personnel skill level attribute, support equipment / tool attribute, spare parts and consumables attribute.

[0076] The application exemplarily provides a specific model structure based on a given maintenance support task element graph G=(V, E, A, R), wherein V is a node set, E is an edge set, A is a node type set, and R is an edge type set; for any node v∈V, its type is denoted as τ(v)∈A; for any edge e=(s, t)∈E, its type is denoted as At this time, the graph convolutional neural network is as shown in Figure 4 The input layer comprises: For the node types in the maintenance support task element graph, a type-specific linear transformation layer set is constructed, and a separate linear mapping function is defined for each input node type τ using formula one: Formula one In formula one: is the initial text embedding feature of the node; is the type-specific weight matrix; is the type-specific bias vector; is the mapped unified dimension hidden feature.

[0077] The graph convolution layer includes two layers, and the calculation process can be formalized as: Formula two In formula two, l represents the number of network layers; represents that the type of target node v is the neighbor node set of is the attention weight calculated based on the edge type is the three-parameter matrix dependent on the source node type, the target node type and the edge type; is the ReLU activation function. The attribute prediction branch of the output layer includes:

[0078] For any attribute node , the feature vector is obtained by encoding the node text description by the bidirectional encoder module, and the output dimension is , and its processing manner is shown in formula three: Formula three In formula three, based on the embedding of the given repair step node , then the attribute categories are predicted by a multi-classification decoder as shown in formula four: Formula four In formula four, is the prediction probability distribution vector of the four types of attributes of the given repair step node, is the weight matrix of the given repair step node, is the bias vector of the given repair step node, is a four-dimensional vector space.

[0079] The link prediction branch of the output layer includes: First, the final embedding representation of two nodes u and v is given as , wherein u is the source node type, v is the target node type, is a 128-dimensional vector space.

[0080] Then, the two are spliced to form a joint representation based on formula five: Formula five In formula five, It is a 256-dimensional vector space.

[0081] Finally, the activation function layer maps the data to scalar scores, and the autoregressive decoder performs Equation 6 for decoding, thus predicting and generating the next task sequence text for this maintenance task. Formula Six in The text for predicting the next task sequence. This is the link weight matrix. This is the link bias vector.

[0082] The model constructed in this invention has high link prediction accuracy on the test set and also shows good inference performance in maintenance attribute matching. This indicates that the model can learn the structural constraints and semantic association information contained in the element graph of maintenance support task well, and can provide a more accurate reference scheme for actual maintenance task planning.

[0083] This invention provides an exemplary method for intelligent planning and assisted design of ship equipment maintenance and support tasks based on the trained model, including: completing maintenance and support task elements and predicting unknown maintenance and support task elements.

[0084] This invention provides an exemplary method for completing maintenance and support task elements, including...

[0085] The model is trained by using the task sheet as the seed node input.

[0086] The trained model compares and analyzes the relevant information in the task sheet from two aspects: maintenance and support operation sequence and maintenance and support operation attributes. It also fills in the missing elements and predicts the probability distribution of the supplemented element information.

[0087] The supplementary element information is normalized, and the element information with the highest probability is selected as the prediction result to complete the task list.

[0088] This invention provides, by way of example, a method for predicting unknown maintenance support task elements, including: The model is trained by using the currently known maintenance and support operation steps as seed nodes.

[0089] The trained model obtains possible steps for the next maintenance and support operation as prediction steps, and predicts the probability that each prediction step is the next maintenance and support operation step.

[0090] When there is a prediction step with a prediction probability greater than the threshold, it is determined that there is a maintenance sequence association, and the prediction step is output.

[0091] The obtained prediction steps are used as seed nodes to input into the trained model. The above process is repeated to generate a maintenance task sequence and calculate the joint probability of the entire path until the link prediction probability is lower than the set threshold and / or the number of inference steps reaches the preset upper limit, thereby completing the prediction of unknown maintenance support task elements.

[0092] Given the initial task, the model can automatically complete the subsequent task sequence and match the corresponding attribute elements. The generated maintenance plan is highly consistent with the results of manual review, indicating that this method can significantly improve the task planning efficiency of maintenance managers and realize intelligent maintenance task reasoning and planning based on similar product models.

[0093] The technical solution of the present invention will be further described below with reference to specific embodiments.

[0094] Step 1: Construct a system of maintenance and support tasks for ship equipment Ensure the standardization and normalization of maintenance support analysis processes. Based on relevant national (military) standards and expert experience, a maintenance support task element system for ship equipment is proposed. This system models the maintenance support task analysis system for ship equipment maintenance activities from two perspectives: maintenance support operation sequence and maintenance support operation attributes. Specific details are as follows: The maintenance support operation sequence refers to the maintenance support operation steps included in a maintenance task, and these maintenance steps have an irreversible sequential relationship. The maintenance support operation attribute refers to the attribute information possessed in a maintenance support operation step, including four attributes: maintenance personnel specialty, maintenance personnel skill level, support equipment / tools, spare parts and consumable materials. The specific descriptions of these four attributes are shown in Table 1.

[0095] Table 1. Description of Maintenance and Support Operation Attributes

[0096] Step 2: Construction of Task Element Diagram for Ship Equipment Maintenance and Support Based on the maintenance and support task element system for ship equipment constructed in step 1, and driven by ship equipment maintenance technical documents, a maintenance and support task element graph for ship equipment is constructed, providing the dataset input for step 3 based on a graph model. Specific details are as follows: Step 2.1: Collect and annotate ship equipment maintenance technical document data Based on the maintenance and support task element system for ship equipment constructed in step 1, ship equipment maintenance technical documents are collected and labeled. The ship equipment maintenance technical documents are labeled with information from six dimensions: "maintenance and support operations", "maintenance and support operation steps", "maintenance personnel specialties", "maintenance personnel skill levels", "support equipment / tools" and "spare parts and consumable materials". They are arranged from top to bottom in the original order of the technical documents and saved in CSV file format.

[0097] Step 2.2: Construction of Maintenance Support Task Element Diagram Based on collected and labeled technical documentation data on ship equipment maintenance, a directed graph was constructed using the Python programming language and the NetworkX library. The directed graph consists of two main parts: nodes and edges. The nodes represent maintenance information, specifically six key categories: "Maintenance Support Operations," "Maintenance Support Operation Steps," "Maintenance Personnel Specialty," "Maintenance Personnel Skill Level," "Support Equipment / Tools," and "Spare Parts and Consumables." These six categories are distinguished by six different colors. The edges represent the relationships between maintenance information. They can represent the sequential relationship between two adjacent maintenance support operations within a maintenance task. For example, in the task of "replacing the magnetic starter junction box transformer," the relationship between maintenance support operation steps ① and ② can be represented as ①→②. They can also represent the dependency relationship between a maintenance support operation step and its maintenance attribute. For instance, the maintenance support operation "replacing the magnetic starter junction box transformer" is related to the maintenance personnel specialty "Electrical Specialty," which can be represented as "Replacing the magnetic starter junction box transformer" → "Electrical Specialty."

[0098] Specifically, firstly, based on a pre-established maintenance support task analysis system for ship equipment maintenance activities, and driven by ship equipment maintenance technical documents, the Python programming language and NetworkX library are used to traverse each row of data in a CSV file, adding maintenance support operations as main nodes in the graph, and adding corresponding maintenance support operation steps as child nodes for each maintenance support operation. Simultaneously, directed edges are established from maintenance support operations to operation steps to represent the dependency relationship between them. For each maintenance support operation step, maintenance support operation attribute information (maintenance personnel specialty, skill level, support equipment, spare parts, and consumable materials) is added as a new node and connected to the corresponding maintenance support operation step node. Directed edges are established between adjacent maintenance support operation steps within the same maintenance support operation to represent the execution order of the steps. Index mapping information is constructed for the nodes and edges of the directed graph, assigning a numeric identifier to each node and creating a source node to target node mapping for each edge. Subsequently, PyG data objects are created using the processed node features, edge features, and their index information. The PyG data object is a data structure specifically designed for graph neural networks. It can encapsulate the topological structure and feature information of a graph, integrating text word vectors and their node and edge information into a 768-dimensional tensor data format, thus forming a maintenance and support task element graph dataset. Finally, the constructed feature dataset is split into training, validation, and test sets in a 7:2:1 ratio, providing feature dataset information as input for subsequent GNN model training.

[0099] Step 3: Training the Graph Convolutional Neural Network Model Since the maintenance support task element graph constructed by this invention is essentially a heterogeneous graph structure, containing various types of nodes and semantic edge relationships of different types, traditional graph neural networks such as GNN and GAT have inherent defects when processing such heterogeneous graphs. That is, they cannot effectively distinguish the semantic differences between different types of nodes and edges and adopt a unified weighting mechanism in the information aggregation process. Therefore, it is easy to weaken or lose the heterogeneous features contained in maintenance domain knowledge.

[0100] This invention addresses the heterogeneous nature of knowledge in ship equipment maintenance and support tasks. By introducing bidirectional encoders and autoregressive decoders in the processing of heterogeneous graphs and node text, an improved Transformer model for heterogeneous graphs is obtained. Independent parameterized representation spaces are designed for different types of nodes and edges. By dynamically learning the importance weights of heterogeneous neighbors using an attention mechanism, the model can accurately capture complex cross-type and cross-level dependencies in maintenance tasks while preserving node type semantics. The core objective is to explicitly model the heterogeneity of node and edge types during graph convolution.

[0101] Step 3.1: HG-BART Model Architecture Design for Maintenance and Support Tasks Given a maintenance and support task element graph G=(V,E,A,R), where V is the set of nodes, E is the set of edges, A is the set of node types, and R is the set of edge types (relations). For any node v∈V, its type is denoted as τ(v)∈A; for any edge e=(s,t)∈E, its type is denoted as... This invention, based on the PyTorch deep learning framework and the PyTorchGeometric graph neural network library, constructs an HG-BART model for element completion in maintenance and support tasks. The specific construction process is as follows: The model architecture includes the following core components: (1) Type-aware feature encoding layer: For the six node types in the maintenance and support task graph, a set of type-specific linear transformation layers is first constructed. Using the PyTorch nn.ModuleDict data structure, an independent linear mapping function is defined for each input node type τ: Formula 1 in, The initial BERT text embedding features for the nodes; A type-specific weight matrix; A type-specific bias vector; This represents the unified dimension latent feature after mapping. In this invention, ∈R768, where R768 is a 768-dimensional vector space; ∈R128×768, where R128×768 is the cross vector space of 768-dimensional and 128-dimensional vectors; ∈R128, where R128 is a 128-dimensional vector space; ∈R128, where R128 is a 128-dimensional vector space.

[0102] (2) Multi-layer heterogeneous graph convolution module: The core uses two heterogeneous graph convolutional layers (HGConv) for information propagation and feature aggregation. The calculation process of the HGConv layer can be formally represented as: Formula 2 Where l represents the number of network layers, which is set to 2 layers in this model; The type of the target node v is The set of neighboring nodes, For edge type Calculated attention weights, σ is a ternary parameter matrix that depends on the source node type, target node type, and edge type; σ is the activation function, and this model uses the ReLU activation function.

[0103] (3) Bidirectional encoder-autoregressive decoder hybrid architecture for multi-task learning: In order to simultaneously achieve the two major goals of link prediction and attribute completion of maintenance task elements, this method designs an attribute prediction branch based on the bidirectional encoder module and a link prediction branch based on the autoregressive decoder module, which are described in detail below: 1) Attribute Prediction Branch: Used to complete the four types of attribute information for maintenance steps (personnel specialty, skill level, support equipment, spare parts). For any attribute node... Its eigenvectors ∈R128 is obtained by encoding the node text description using the bidirectional encoder module, where Given a 128-dimensional vector space acceptable to the bidirectional encoder module, the output dimension is processed as follows: Formula 3 Embedded based on given maintenance step nodes The model predicts the category of each attribute using a multi-class decoder in the manner shown in Equation 4: Formula 4 in Given a maintenance step node, this is the predicted probability distribution vector of four attributes. Given the weight matrix of the maintenance step nodes, Given the bias vector of a maintenance step node, It is a four-dimensional vector space.

[0104] 2) Link Prediction Branch: Used to predict and generate potential relationships in a maintenance task sequence. Given the final embedding representation of two nodes u (source node type, such as step) and v (target node type, such as support equipment). This model first concatenates the two to form a joint representation: Formula 5 Formula 5 It is a 256-dimensional vector space.

[0105] Subsequently, the data is mapped to scalar scores through an activation function layer, and then decoded using Equation 6 via an autoregressive decoder to predict and generate the next task sequence text for this maintenance task. Formula Six in The text for predicting the next task sequence. This is the link weight matrix. This is the link bias vector.

[0106] The final HG-BART model constructed for maintenance and support task elements is as follows: Figure 4 As shown, the basic operation flow of this model is as follows: Figure 1 As shown, this design allows the model to learn both the structural information of the graph (for link prediction) and the semantic information of the nodes (for attribute prediction), thus providing comprehensive predictive capabilities in maintenance and support tasks.

[0107] Step 3.2: HG-BART Model Training Evaluation Based on the HG-BART model constructed in step 3.1, the Adam optimizer of the nn.Module class in the PyTorch library was used for training, with the learning rate set to 0.01. After training the model for 200 epochs, the final model training test results showed a training loss of 0.4403 and a training accuracy of 0.7925. The validation loss was 0.5147 and the validation accuracy was 0.7453. The training process is as follows: Figure 2 As shown. The test set divided in step 3.1 was used for testing, and the test results are as follows: the test loss is 0.4951 and the test precision is 0.7740.

[0108] To further explore the impact of each module of the HG-BART model on performance, an ablation study was conducted using a heterogeneous graph convolutional neural network (HGCNN) as the baseline model. This study included a heterogeneous transformer model (decoder only) and a heterogeneous BERT model (encoder only). The final experimental results are shown in Table 2.

[0109] Table 2 Comparison of ablation experiments

[0110] Experimental results show that the complete HG-BART model significantly outperforms the HGCNN baseline and the two ablation models across all four evaluation metrics. Specifically, the decoder-only HG-Transformer model has an F1 score that is approximately 7.76% lower and an accuracy that is approximately 7.53% lower than the complete HG-BART model, indicating that the absence of the encoder module severely limits the model's ability to extract and understand complex structural information from heterogeneous graphs. On the other hand, while the encoder-only HG-BERT model outperforms the baseline, its F1 score is still approximately 3.34% lower and its accuracy that is approximately 2.75% lower than HG-BART, reflecting the indispensable role of the decoder in inference and sequence generation tasks based on encoded information.

[0111] The encoder's role is to deeply fuse and learn representations of multi-level features of nodes and relationships in heterogeneous graphs, providing high-quality information condensation for downstream tasks. The decoder, on the other hand, uses this representation for autoregressive generation or task inference, forming a collaborative "understanding-inference" mechanism. Ablation experiments show that removing any part disrupts this information processing loop, leading to a decrease in the model's expressive power, and further verifying the necessity of the encoder-decoder joint architecture in heterogeneous graph representation learning.

[0112] Step 4: Intelligent planning and auxiliary design for ship equipment maintenance and support tasks Complete existing maintenance and support task elements: Based on the HG-BART model constructed in step 3, and within the framework established in step 1, predictions are made using existing maintenance and support tasks as the target, thereby supplementing the missing elements of existing maintenance and support tasks (maintenance personnel expertise, maintenance personnel skill levels, support equipment / tools, and spare parts and consumables). This step first inputs the currently known maintenance operation steps as seed nodes into the HG-BART model. Using the encoder function defined in step 3.1, the 768-dimensional input features are mapped to a 64-dimensional node embedding representation through two layers of graph convolution. Secondly, this step outputs a 4-dimensional vector through a 64-dimensional node embedding linear mapping layer, corresponding to the predicted probability distributions of the four attributes: maintenance personnel expertise, maintenance personnel skill levels, support equipment / tools, spare parts, and consumables. The Softmax function is used to normalize the outputs of each attribute category, selecting the category with the highest probability as the prediction result, thus ultimately completing the element supplementation for existing maintenance and support tasks.

[0113] Prediction of unknown maintenance and support task elements: Based on the HG-BART model constructed in step 3, and within the architecture established in step 1, the next unknown maintenance and support task and its corresponding elements are predicted using existing maintenance and support tasks as the object. This step first inputs the currently known maintenance operation steps as seed nodes into the HG-BART model. Using the encoder function defined in step 3.1, the 768-dimensional input features are mapped to a 64-dimensional node embedding representation through two layers of graph convolution. Next, the link prediction branch is used to infer the next maintenance operation step. The 64-dimensional embedding of the current node is concatenated with the 64-dimensional embedding of the candidate node to form a 128-dimensional edge feature vector. This vector is compressed into a 1-dimensional output through a linear mapping layer, and after passing through a Sigmoid activation function, the link existence probability value in the [0,1] interval is obtained. A probability threshold of 0.5 is set; when the predicted probability is greater than this threshold, a maintenance sequence association is determined, thus achieving the prediction of unknown maintenance and support tasks. Subsequently, for newly added maintenance operation steps, the attribute prediction branch completes their four types of maintenance attribute information. Based on the predicted probability distribution, the most matching specific value is selected for each attribute, such as "Electrical Specialty," "Junior," and "Phillips Screwdriver." Finally, the newly generated maintenance task steps and their attributes are used as known information and re-inputted into the HG-BART model for the next round of inference. For the generated maintenance task sequence, the joint probability of the entire path is calculated. Let the predicted probability of the link from step i to step i+1 be... The confidence score of the entire maintenance sequence path containing n steps is:

[0114] Repeat the link prediction and attribute prediction process until the termination conditions are met. The termination conditions include: 1. The link prediction probability falls below a set threshold, indicating that the end of the maintenance sequence has been reached. 2. The number of inference steps reaches a preset upper limit to prevent infinite loops.

[0115] Furthermore, this invention provides an application process example based on a specific embodiment: Based on a maintenance support operation "replacing a circuit board" in the test set, information from the maintenance manual was consulted, and maintenance technical document data was obtained, such as... Figure 3 As shown.

[0116] Then, based on the maintenance and support task element system for ship equipment constructed in step 1, the steps in step 2, such as collecting and annotating ship equipment maintenance technical document data and constructing the maintenance and support task element diagram for ship equipment, are executed to annotate the associated maintenance tasks and corresponding elements and construct the maintenance and support task element diagram.

[0117] Execute step 3, input the maintenance support task element map established above as a learning sample into the HG-BART model for training and debugging. Then, based on the debugging results, input the first maintenance support operation step of the task, "open the frame with a screwdriver". After converting the information of this step into word vector features and loading it into the HG-BART model, the attribute reasoning information of the "replace board" step is completed, as shown in Table 3.

[0118] Table 3. Complete Existing Maintenance Support Task Elements Based on the HG-BART Model

[0119] Secondly, the unknown maintenance and support task elements were predicted based on the HG-BART model: the first maintenance operation step of the task was input as "open the frame with a screwdriver". After converting the step information into word vector features and loading it into the HG-BART model, the unknown maintenance and support task was inferred as "press the power off button to ensure that the system is in a normal shutdown state", and the inferred maintenance and support task element was "electrical profession, primary, explosion-proof Phillips screwdriver". The results are shown in Table 4.

[0120] Table 4. Prediction of unknown maintenance support task elements based on the HG-BART model

[0121] Based on the predicted next operation step and its attribute information, the data is loaded into the HG-BART model for prediction and reasoning of subsequent steps. This process continues iteratively, and after repeated inferences, the reasoning results for the entire maintenance operation "replacing the circuit board" are shown in Table 5.

[0122] Table 5. Reasoning Results for “Replacing the Board”

[0123] After manual review, the final results are shown in Table 6. All the predictions for maintenance and support operation steps were correct, with only slight deviations in steps ② and ⑤ regarding the reasoning for support equipment and spare parts. This indicates that the model's reasoning effect is good and can assist maintenance designers in quickly formulating large-scale equipment maintenance and support tasks.

[0124] Table 6 Results after manual review

[0125] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A method for intelligent planning and auxiliary design of ship equipment maintenance and support tasks, characterized in that, include: Construct a system of maintenance and support task elements for ship equipment, including two aspects: maintenance and support operation sequence and maintenance and support operation attributes; Based on the maintenance and support task element system for ship equipment, we collect and annotate ship equipment maintenance technical documents and construct a maintenance and support task element diagram. Based on the maintenance and support task element graph, the nodes of the maintenance and support task element graph are preprocessed and then used as learning samples for graph convolutional neural network (GCNN) to train the model. The trained model enables intelligent planning and assisted design of ship equipment maintenance and support tasks.

2. The intelligent planning and auxiliary design method for ship equipment maintenance and support tasks according to claim 1, characterized in that, The maintenance support operation sequence is a maintenance operation step that has an irreversible sequential relationship within a maintenance task, and the maintenance support operation attribute is the attribute information possessed by the maintenance operation step.

3. The intelligent planning and auxiliary design method for ship equipment maintenance and support tasks according to claim 2, characterized in that, The maintenance and support operation attributes include: The professional attributes of maintenance personnel refer to the professional skills required of maintenance personnel in this maintenance operation step. The skill level attribute of maintenance personnel refers to the requirements of the professional skill title or level of maintenance personnel in this maintenance operation step. Support equipment / tool ​​attributes: Support equipment / tool ​​attributes refer to the support equipment or tools required in this maintenance operation step; Spare parts and consumables refer to the spare parts and consumables required in this maintenance operation.

4. The intelligent planning and auxiliary design method for ship equipment maintenance and support tasks according to claim 1, characterized in that, The method for collecting and annotating ship equipment repair technical documents based on the ship equipment maintenance and support task element system includes: Collect technical documents on ship equipment maintenance; Based on the task element system for ship equipment maintenance support, the ship equipment maintenance technical documents are labeled with information from six dimensions: "maintenance operation", "maintenance operation steps", "maintenance personnel specialty", "maintenance personnel skill level", "support equipment / tools" and "spare parts and consumables". Arranged in the order of the original technical documents from top to bottom.

5. The intelligent planning and auxiliary design method for ship equipment maintenance and support tasks according to claim 1, characterized in that, The method for constructing the maintenance support task element graph includes: constructing a directed graph consisting of nodes and edges based on the labeled ship equipment maintenance technical documents; and connecting the directed graph based on the association relationship to obtain the maintenance support task element graph. The nodes of the directed graph represent maintenance information; The edges of the directed graph represent the associations between maintenance information, that is, the sequence relationship between two adjacent maintenance operations in a maintenance task.

6. The intelligent planning and auxiliary design method for ship equipment maintenance and support tasks according to claim 1, characterized in that, The method for preprocessing the nodes of the maintenance and support task element map includes: Embed word vector features into the text information of each node in the maintenance and support task element diagram, thereby converting the text of each node in the maintenance and support task element diagram into word vectors. Based on the word vector data obtained from the transformation, index mapping information of each node is constructed and a numerical identifier is assigned to each node. Index mapping information of the relationship between nodes is constructed and a mapping from source node to target node is created. A feature dataset is created based on the processed information: nodes, node index mapping information, relationships between nodes, and relationship index mapping information. The feature dataset is divided into training, validation, and test sets in a 7:2:1 ratio, which will serve as input information for subsequent models.

7. The intelligent planning and auxiliary design method for ship equipment maintenance and support tasks according to claim 1, characterized in that, The graph convolutional neural network includes: Input layer: Receives node features and information about relationships between nodes; The first convolutional layer performs the first dimensionality transformation on the information from the input layer and then connects it to the activation function; The second convolutional layer further processes the data output from the first convolutional layer while maintaining the same data dimensions. The output layer includes: link prediction branch and attribute prediction branch.

8. The intelligent planning and auxiliary design method for ship equipment maintenance and support tasks according to claim 7, characterized in that, The link prediction branch is as follows: the 3D features of two nodes are concatenated and mapped to a 1D output through a linear layer, representing the probability of the existence of a relationship between adjacent nodes; The attribute prediction branch is as follows: the node data output by the convolutional layer of the second graph is directly mapped to the 4-dimensional output through a linear layer, representing the prediction probability of four maintenance attributes; the four maintenance attributes are: maintenance personnel professional attributes, maintenance personnel skill level attributes, support equipment / tool ​​attributes, and spare parts and consumable material attributes.

9. The intelligent planning and auxiliary design method for ship equipment maintenance and support tasks according to claim 8, characterized in that, Given a maintenance and support task element graph G=(V, E, A, R), where V is the set of nodes, E is the set of edges, A is the set of node types, and R is the set of edge types; for any node v∈V, its type is denoted as τ(v)∈A; for any edge e=(s,t)∈E, its type is denoted as... ; At this point, the graph convolutional neural network includes: Input Layer: For each node type in the maintenance and support task element graph, a set of type-specific linear transformation layers is constructed, using Equation 1 to define an independent linear mapping function for each input node type τ. Formula 1 In formula 1: The initial text embedding features for the nodes; A type-specific weight matrix; A type-specific bias vector; The mapped unified dimension latent features; A graph convolutional layer consists of two layers, and its computation process can be formally represented as follows: Formula 2 In Equation 2: l represents the number of network layers; The type of the target node v is The set of neighboring nodes, For edge type Calculated attention weights, A ternary parameter matrix that depends on the source node type, target node type, and edge type; It is the ReLU activation function; The output layer's attribute prediction branches include: For any attribute node Its eigenvectors The output dimension is obtained by encoding the node text description by the bidirectional encoder module, and its processing method is shown in Equation 3: Formula 3 In Equation 3, the embedding is based on a given maintenance step node. Then, the multi-class decoder predicts the category of each attribute as shown in Equation 4: Formula 4 Formula 4 Given a maintenance step node, this is the predicted probability distribution vector of four attributes. Given the weight matrix of the maintenance step nodes, Given the bias vector of a maintenance step node, It is a four-dimensional vector space; The link prediction branch of the output layer includes: First, given the final embedding representations of two nodes u and v. Where u is the source node type and v is the target node type. It is a 128-dimensional vector space; Then, based on Equation 5, the two are concatenated to form a joint representation: Formula 5 Formula 5 It is a 256-dimensional vector space; Finally, the activation function layer maps the data to scalar scores, and the autoregressive decoder performs Equation 6 for decoding, thus predicting and generating the next task sequence text for this maintenance task. Formula Six in The text for predicting the next task sequence. This is the link weight matrix. This is the link bias vector.

10. The intelligent planning and auxiliary design method for ship equipment maintenance and support tasks according to claim 1, characterized in that, The method for intelligent planning and assisted design of ship equipment maintenance and support tasks based on the trained model includes: completion of maintenance and support task elements and prediction of unknown maintenance and support task elements; The completion of maintenance and support task elements includes: The model is trained by inputting the task sheet as a seed node. The trained model compares and analyzes the relevant information in the task sheet from two aspects: maintenance and support operation sequence and maintenance and support operation attributes. It also fills in the missing elements and predicts the probability distribution of the supplemented element information. The supplementary element information is normalized, and the element information with the highest probability is selected as the prediction result to complete the task list. The prediction of unknown maintenance and support task elements includes: The model is trained by inputting the currently known maintenance operation steps as seed nodes; The trained model obtains possible steps for the next maintenance operation as prediction steps, and predicts the probability that each prediction step is the next maintenance operation step. When there is a prediction step with a prediction probability greater than the threshold, it is determined that there is a maintenance sequence association, and the prediction step is output. The obtained prediction steps are used as seed nodes to input into the trained model. The above process is repeated to generate a maintenance task sequence and calculate the joint probability of the entire path until the link prediction probability is lower than the set threshold and / or the number of inference steps reaches the preset upper limit, thereby completing the prediction of unknown maintenance support task elements.

Images