Cable processing technology generation method and system based on knowledge graph

Abstract

The invention discloses a cable processing technology generation method and system based on a knowledge graph, and the method comprises the following steps: 1, carrying out the modeling of cable processing technology knowledge, building a CPKG model, converting the relation between the sub-types and attributes of an entity into the relation between classes through extracting the technology information in a cable processing technology document and a 3D model, and carrying out the calculation of the relation between the classes; and creating a cable processing process mode layer. And 2, generating a structure diagram, embedding the structure diagram into a cable processing process, creating an ICPKG structure tree, constructing a distributed embedded reasoning model, and continuously completing CPKG through calculation of sub-diagram similarity between ICPKG and CPKGD. And 3, producing a cable processing technology planning RDF / XML document and outputting the RDF / XML document. According to the method, the knowledge graph technology is used for completing automatic generation of the cable processing technological process, repeated calculation and waste of technological knowledge in cable processing are reduced, and therefore automation and intelligence of the cable processing process are improved.

Description

Technical Field

[0001] The present invention belongs to the field of cable process production, and specifically relates to a cable processing technology generation method and system based on a knowledge graph. Background Art

[0002] The traditional cable processing process planning is cumbersome, repetitive, long, and inefficient. Most of them are designed manually, which easily limits the quality of products. At the same time, due to the different experiences and design ideas of different process personnel, it is difficult to form the inheritance, standardization and standardization of process planning. The existence of these problems makes the processing and manufacturing of cables increasingly incompatible with market demand and gradually unable to keep up with the pace of modern production. In recent years, some solutions for the intelligent generation of cable process planning have been proposed. Products can be determined by describing the characteristics of products, thereby determining the process of products. However, this process relies on process personnel to calculate cable parameters. Cables are used in many places, and there are different requirements for various parameters and process methods of cables. This makes the calculation of process personnel cumbersome, repetitive work is large, and the error rate is high. The utilization of used cable information documents is low, resulting in the slow generation of cable process flow documents, which seriously restricts the improvement of the efficiency of cable processing process preparation. How to improve the efficiency and standardization of cable processing process planning and realize the knowledge and intelligence of cable processing process planning is urgent and important in the cable processing process.

[0003] The development of artificial intelligence technology has brought new impetus to the development of manufacturing industry and promoted the transformation of manufacturing industry from automation to intelligence. This requires process planning to develop in the direction of intelligence to meet the requirements of intelligent manufacturing. In intelligent process design, how to accurately represent and utilize process knowledge is an important issue. These process knowledge comes from expert experience, process documents and design documents. Therefore, the lack of effective organization and management of process knowledge makes it impossible to establish an effective and accurate knowledge base. KG (knowledge graph), as a branch of artificial intelligence, is an effective tool for knowledge management. Process knowledge extraction, part drawing construction and process route generation are important contents of intelligent process design. PKG is a domain-specific KG. PKG mainly manages process knowledge and performs manufacturing process planning. General KG is a structured knowledge base in a general domain, which contains a large amount of common sense knowledge from the open world. The main difference between PKG and general KG is data source and object-oriented. Process knowledge is dynamic, highly associated, and describes process entities that change over time. Open world knowledge is powerful and once established, it may not change for a long time. PKG is also similar to general KG in many ways. First, many methods used in general knowledge bases can be migrated to PKG, such as knowledge extraction, entity alignment, knowledge fusion, etc. Second, some mature tools can be used for PKG, such as Neo4j database.

[0004] Computer-aided process planning (CAPP) has been widely used in the process planning of various parts and products, bringing convenience to process personnel. However, it is unable to reuse process knowledge and experience. PKG can perfectly solve this problem. Compared with traditional CAPP, PKG-based process design has the following advantages: (1) PKG can more conveniently store and process various forms of complex process knowledge, including structured process knowledge, semi-structured process knowledge and structured process knowledge; (2) PKG can reveal the explicit and implicit relationships between process knowledge, so that seemingly isolated process knowledge can be interconnected. The structure of PKG determines its ability to complete process knowledge; (3) PKG-based process generation is the global optimal result obtained by combining all process knowledge in the graph. Compared with traditional CAPP, it has stronger versatility and intelligence; (4) PKG-based process planning is flexible and can be applied to large-scale dynamic manufacturing processes. On the contrary, traditional CAPP is not suitable for large-scale dynamic manufacturing processes because it cannot truly adjust the process in real time and balance resource load; (5) PKG is a key technology for the transformation from traditional manufacturing to intelligent manufacturing. Through these advantages, PKG can reuse process knowledge and automatically generate process planning, promoting the standardization and automation of processing process planning. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a cable processing technology generation method and system based on knowledge graph, aiming to use knowledge graph technology to complete the automatic generation of cable processing process flow, reduce repetitive calculations and waste of process knowledge in cable processing, and thus improve the automation and intelligence of the cable processing process.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A method for generating a cable processing technology based on a knowledge graph comprises the following steps:

[0008] Modeling the cable processing technology knowledge, establishing a CPKG model, converting the relationship between the subtypes and attributes of entities into the relationship between classes based on the CPKG model, and creating a cable processing process model layer;

[0009] Generate a structure diagram and embed it into the cable processing process model layer, create an ICPKG structure tree, and build a distributed embedded reasoning model based on the ICPKG structure tree;

[0010] Based on the distributed embedded reasoning model, a cable processing process planning RDF / XML document is generated.

[0011] Preferably, the method of modeling the cable processing technology knowledge and establishing the CPKG model includes:

[0012] Extract the process information contained in the cable processing process documents and 3D models, and describe the CPKG model as:

[0013] CPKGM={P∪F∪G∪C∪T∪S}

[0014] Among them, P represents process steps, F represents cable processing features, G represents cable components, C represents processing requirements, T represents tooling, and S represents semantic relationships.

[0015] Preferably, based on the CPKG model, the relationship between the subtypes and attributes of the entity is converted into the relationship between classes, and the method for creating the cable processing process model layer includes:

[0016] Preset semantic relationship S;

[0017] Based on the semantic relationship S, six classes are established: process, processing feature, component, requirement, tool and relationship, and subclasses are established through inheritance;

[0018] Create the cable processing process model layer based on the six established classes and subclasses established through inheritance;

[0019] Among them, the semantic relationship S is described as:

[0020] S = {Has, Connect, Sequence, Parallel, Recycle}, where Has represents a containment relationship, Connect represents the connection mode of wires and components, Sequence represents the sequential relationship between process steps, Parallel represents the parallel relationship between process steps, and Recycle represents the cycle of a certain process step;

[0021] Based on the semantic relationship S, the methods for establishing six categories of processes, processing features, components, requirements, tools and relationships include:

[0022] (F, P) represents the inclusion relationship between cable processing features and process steps, (P, G) represents the inclusion relationship between process steps and cable components, (P, T) represents the inclusion relationship between process steps and tooling, (P, C) represents the inclusion relationship between process steps and processing requirements, (F, C) represents the inclusion relationship between cable processing features and technical requirements, and (G, F) represents the inclusion relationship between cable components and processing features;

[0023] The subclasses established through inheritance include: wire and component classes inherit from component classes, geometric features, tolerance dimensions, and material feature classes inherit from processing feature classes, dimensional deviations, heat treatment requirements, and ingredient requirements classes inherit from requirement classes, and process step classes inherit from process classes.

[0024] Preferably, the generated structure diagram is embedded into the cable processing process model layer, an ICPKG structure tree is created, and based on the ICPKG structure tree, a method for constructing a distributed embedded reasoning model includes:

[0025] Construct a cable component structure tree to describe the connection relationship between cable components, and save the cable component structure tree as a triple, and connect each in the form of a cable component structure tree to construct an initial processing technology knowledge graph model ICPKG; wherein the cable component structure tree information includes the ID number of the component, the connected components and wires, and the semantic relationship "Connect";

[0026] Based on the initial machining process knowledge graph model ICPKG, scan the machining process documents and CAD models to generate the machining process knowledge graph database CPKGD;

[0027] A distributed embedded reasoning model is established based on the machining technology knowledge graph database CPKGD.

[0028] Preferably, based on the processing technology knowledge graph database CPKGD, the method for establishing a distributed embedded reasoning model includes:

[0029] The SKGCN method is used for knowledge reasoning to extract multiple processing case graph data from the processing technology knowledge graph database CPKGD;

[0030] Use CompGCN to learn the aggregate information of each machining process case graph data, generate the embedding vector corresponding to each machining process sequence, and connect the embedding vectors corresponding to different machining process situations;

[0031] Based on the connected embedding vectors, a capsule network is used to design an evaluation function to learn the implicit association semantics between cable processing instances at the global level and establish a distributed embedding reasoning model.

[0032] Preferably, the method of using CompGCN to learn the aggregate information of each processing process case graph data respectively and generate the embedding vector corresponding to each processing process sequence includes:

[0033]

[0034] Among them, h o represents the update amount of node o, is the set of neighbors of the outgoing edge of o, is a specific parameter of a relation type, φ represents a composite operation, x s , z r Represent the initial features of node s and relationship r respectively;

[0035]

[0036] Among them, h r represents relation embedding, is a learnable transformation matrix that projects all relations into the same embedding mapping space as nodes, describes a learnable scalar weight for a relation and a specific basis, represents a set of learnable basis vectors, V b =∈V1,V2,......,V β represents a set of learnable basis vectors, and β is the number of sets of learnable basis vectors;

[0037]

[0038] in, represents the node O obtained after k layers, is a specific parameter of the kth layer, represents the node features of the kth layer, represents the relation embedding at the kth layer;

[0039]

[0040] in, Represents the representation of relation r after k layers.

[0041] Preferably, using a capsule network, a method for designing an evaluation function includes:

[0042] f(s, r, o)=||capsnet(g([v, r, u]*Ω))||

[0043] Among them, s, r, o are the output triplet information, s, o represent nodes, and r represents relationships; similarly, v, r, u represent the input triplet information, v, u represent nodes, and r represents relationships; Ω represents the weight matrix, and the g function represents the nonlinear transformation of the triplet after the weight matrix transformation; caspsnet represents the capsule neural network structure.

[0044] Preferably, based on the distributed embedded reasoning model, the method for producing a cable processing process planning RDF / XML document includes:

[0045] Store the nodes and edges in the CPKG model in XML format;

[0046] Store edge triplets by traversing the "first and last nodes" information of each edge;

[0047] The attribute value information of each node is obtained through iterative query.

[0048] The present invention also provides a cable processing technology generation system based on knowledge graph, including: a pattern layer creation module, a model building module and a generation module;

[0049] The pattern layer creation module is used to model the cable processing technology knowledge, establish a CPKG model, and based on the CPKG model, convert the relationship between the subtype and attribute of the entity into the relationship between classes and classes to create a cable processing process pattern layer;

[0050] The model building module is used to generate a structure diagram embedded in the cable processing process model layer, create an ICPKG structure tree, and build a distributed embedded reasoning model based on the ICPKG structure tree;

[0051] The generation module is used to generate a cable processing process planning RDF / XML document based on the distributed embedded reasoning model.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] The present invention discloses 1. Modeling the cable processing technology knowledge and establishing a CPKG model. By extracting the process information contained in the cable processing technology documents and 3D models, the relationship between the subtypes and attributes of the entities is converted into the relationship between classes and classes, and a cable processing process model layer is created. 2. Generate a structure diagram and embed it into the cable processing process, create an ICPKG structure tree, build a distributed embedded reasoning model, and continuously complete the CPKG by calculating the subgraph similarity between ICPKG and CPKGD. 3. Produce a cable processing process planning RDF / XML document for output. The present invention aims to use knowledge graph technology to complete the automatic generation of cable processing process flow, reduce repetitive calculations and waste of process knowledge in cable processing, and thus improve the automation and intelligence of the cable processing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0055] Figure 1 A schematic diagram showing a mapping relationship of part of the data between mode layers according to an embodiment of the present invention;

[0056] Figure 2 Schematic diagram of a method for constructing a distributed embedded reasoning model according to an embodiment of the present invention;

[0057] Figure 3It is a schematic diagram of a method for converting information in a structure tree into triples according to an embodiment of the present invention.

[0058] Figure 4 Schematic diagram of the cable processing mode layer according to an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0061] Embodiment 1

[0062] The present invention provides a method for generating a cable processing technology based on a knowledge graph, comprising the following steps:

[0063] Modeling the cable processing technology knowledge, establishing a processing technology knowledge graph (CPKG) model, based on the CPKG model, converting the relationship between the subtypes and attributes of entities into the relationship between classes, and creating a cable processing process pattern layer;

[0064] Generate a structure diagram and embed it into the cable processing process model layer, create an ICPKG structure tree, and build a distributed embedded reasoning model based on the ICPKG structure tree;

[0065] Based on the distributed embedded reasoning model, a cable processing process planning RDF / XML document is generated.

[0066] In this embodiment, the method of modeling the cable processing technology knowledge and establishing the CPKG model includes:

[0067] Extract the process information contained in the cable processing technology document and a certain type of cable (3D model) that has been saved, and describe the CPKG model as:

[0068] CPKGM={P∪F∪G∪C∪T∪S}

[0069] Among them, P represents process steps, F represents cable processing features, G represents cable components, C represents processing requirements, T represents tooling, and S represents semantic relationships. F is 3D model information.

[0070] In this embodiment, based on the CPKG model, the relationship between the subtype and the attribute of the entity is converted into the relationship between the class and the class, and the method for creating the cable processing process model layer includes:

[0071] Preset semantic relationship S;

[0072] Based on the semantic relationship S, six classes are established: process, processing feature, component, requirement, tool and relationship, and subclasses are established through inheritance;

[0073] Create the cable processing process model layer based on the six established classes and subclasses established through inheritance;

[0074] Among them, the semantic relationship S is described as:

[0075] S = {Has, Connect, Sequence, Parallel, Recycle}, where Has represents a containment relationship, Connect represents the connection mode of wires and components, Sequence represents the sequential relationship between process steps, Parallel represents the parallel relationship between process steps, and Recycle represents the cycle of a certain process step;

[0076] Based on the semantic relationship S, the methods for establishing six categories of processes, processing features, components, requirements, tools and relationships include:

[0077] Has represents a containment relationship, which can be described as:

[0078] S={(F,P),(P,G),(P,T),(P,C),(F,C),(G,F)}

[0079] In this description, (F, P) indicates the inclusive relationship between cable processing features and process steps, (P, G) indicates the inclusive relationship between process steps and cable components, (P, T) indicates the inclusive relationship between process steps and tooling, (P, C) indicates the inclusive relationship between process steps and processing requirements, (F, C) indicates the inclusive relationship between cable processing features and technical requirements, and (G, F) indicates the inclusive relationship between cable components and processing features.

[0080] Subclasses established through inheritance include: wires and components inherit from component classes, geometric features, tolerance dimensions, and material features inherit from processing feature classes, dimensional deviations, heat treatment requirements, and ingredient requirements inherit from requirement classes, and process step classes inherit from process classes. Thus, the relationship between two entities that have subtypes or attributes is converted into a specific and unique node relationship between ontology pattern classes and classes.

[0081] Example: Figure 1It shows the specific mapping from entity relationship to class and class relationship. It can be seen from the figure that the subtype relationship or attribute relationship between two entities is converted into a node relationship between classes. For example, the subtype relationship between cable assembly and cable component is mapped into a subclass relationship between assembly class and component subclass, and the attribute relationship between cable component and processing feature is mapped into a characteristic relationship between component class and feature class.

[0082] Finally, the pattern layer is generated according to the relationship mapping. In order to clearly describe the complex semantic information of the cable processing process, the cable processing process pattern layer (semantic relationship diagram, various semantic relationships connect the attributes of the six classes mentioned above) can be divided into the cable structure pattern layer, the cable processing process step pattern layer and the cable processing procedure pattern layer, such as Figure 4 As shown in the figure. The processing structure semantic relations Has and Connect are mainly reflected in the cable processing structure pattern layer. The processing sequence semantic relations include Sequence, Parallel and Recycle, which are mainly reflected in the cable processing process step pattern layer and the cable processing process pattern layer.

[0083] Example: Figure 2 A schematic diagram of the construction method of the distributed embedding reasoning model is shown.

[0084] In this embodiment, a structure diagram is generated and embedded into the cable processing process model layer, an ICPKG structure tree is created, and a method for constructing a distributed embedded reasoning model based on the ICPKG structure tree includes:

[0085] Construct a cable component structure tree to describe the connection relationship between cable components, and save the structure tree as a triple. The information of the structure tree includes the ID number of the component, the connected components and wires, and the semantic relationship "Connect". Connect each component and wire entity in the form of the structure tree to construct an ICPKG (initial processing technology knowledge graph) model.

[0086] Scan the processing technology documents and the CAD model containing the cable processing features (including the cable processing features), define and generate CPKGD (processing technology knowledge graph database).

[0087] Construct a distributed embedding inference model. The construction method is as follows Figure 2As shown in the figure, the sequential knowledge graph convolutional network (SKGCN) algorithm is used for knowledge reasoning, that is, multiple processing process case graph data are extracted from CPKGD, and the aggregation information of each processing process case graph data is learned separately using the combination-based multi-relation graph convolutional network (CompGCN), and the embedding vector corresponding to each processing process sequence is generated. The embedding vectors corresponding to different processing conditions are connected, and an evaluation function is designed based on the capsule network to learn the implicit association semantics between cable processing instances at the global level. The specific steps are as follows:

[0088] 1) Define CPKGD as in ε represents the set of nodes and edges respectively, represents the d0-dimensional input feature of each node, represents a collection of relations, Represents the initial relationship characteristics.

[0089] 2) Define h o is the node embedding, expressed as follows:

[0090]

[0091] Among them, h o represents the update amount of node o, is the set of neighbors of the outgoing edge of o, is a relation type specific parameter, Represents a compound operation, x s , z r They represent the initial features of node s and relation r respectively.

[0092] 3) Define the relation embedding h r as follows:

[0093]

[0094] is a learnable transformation matrix that projects all relations into the same embedding map space as the nodes, describes the learnable scalar weights of the relation and a specific basis, V b =∈V1,V2,......,V β represents a set of learnable basis vectors, and β is the number of sets of learnable basis vectors. In addition, h r It also allows these relations to be used in the next SKGCN layer.

[0095] 4) Definition The node o obtained after k layers is expressed as follows:

[0096]

[0097] is a specific parameter of the kth layer, represents the node features of the kth layer, represents the relation embedding at the kth layer.

[0098] 5) Definition The mathematical representation of the relation r after k layers is as follows:

[0099]

[0100] 6) The evaluation function based on the capsule network is defined as follows:

[0101] f(s, r, o)=||caosnet(g([v, r, u]*Ω))|| (5)

[0102] s, r, o are the output triplet information, s, o represent nodes, and r represents relationships; similarly, v, r, u represent the input triplet information, v, u represent nodes, and r represents relationships; Ω represents the weight matrix, and the g function represents the nonlinear transformation of the triplet after the weight matrix transformation; caspsnet represents the capsule neural network structure.

[0103] 7) Define the graph similarity method as follows:

[0104]

[0105] in, is a subgraph of CPKG, which can be described as a multi-hop sequence S i ={s, r1, e1...r i , e i ...r l ,o}, l represents the number of hops, represents the embedding vectors of nodes and edges from a multi-hop sequence with a central node, where h v and h r It can be obtained from equations (3) and (4).

[0106] 8) Define formula (7) to obtain the similar subgraph corresponding to the subgraph. If the subgraph contains multiple multi-hop sequences, multiple corresponding similar subgraphs can be obtained by formula (7), as shown below:

[0107]

[0108] 9) Based on the calculations in the above steps, the knowledge structure is learned from CPKGD to obtain multiple initial processing schemes.

[0109] 10) Reuse formula (7) to calculate the missing processing information, including processing steps, processes, etc. The specific implementation steps are: first input ICPKG containing basic nodes and edges, then complete a certain type of node, complete the edges involved in the node type, and then determine whether all types of nodes have been completed. Finally, when all types of nodes and edges are completed, output the completed ICPKG.

[0110] In this embodiment, the nodes and edges in CPKG are first stored in XML format. Different nodes have different attributes. When generating RDF / XML, they are divided into different nodes, including cable components (parts), wires (lead), process steps (ProcessStep) and processes (Process). The attributes of the edges include name (name), head node (headnode) and tail node (tailnode).

[0111] Then, the edge triplets are stored by traversing the “first and last nodes” information of each edge.

[0112] The attribute value information of each node can be obtained through iterative query.

[0113] Application example: Take the processing of a certain type of cable as an example. First, based on the step model of the cable and the scan sequence of the processing technology document, create a triple to obtain the processing technology knowledge database of the cable. In this database, the scanned entities and relationships are saved in the form of nodes and edges. The nodes should include cable components, processing features, technical requirements, process steps, and tooling. The edges (relationships) include Has, Connect, Sequence, Parallel, and Recycle. Then start to create a processing structure tree and save the structure tree information in the form of triples. The specific method is as follows: Figure 3As shown. Input the processing structure tree, create an entity list listN and a relationship listR, save the nodes in listN by traversing the structure tree, save the relationship between every two nodes in the traversal structure tree in listR, output the <component, connecting component, connecting wire> triple, build ICPKG through the structure tree of each component and wire, based on the distributed embedding reasoning model mentioned in the second step, input the triple in ICPKG. Combined with the process modeling in the first step, the model is updated through incremental training. Then generate implicit processing relationships and entities. Because ICPKG only contains three types of information. That is, component and wire nodes, "connecting wire" edges between components and wires, and "connecting components" edges between components and components. Therefore, the similarity between the subgraphs in ICPKG and CPKGD can be calculated by loading the nodes of "components, wires" and the edges of "connecting wires, connecting components", selecting the subgraph with the highest similarity from multiple subgraphs, completing the pattern layer of the cable structure, and using it as the CPKG of the period, and calculating the similarity by loading the multi-hop sequence between the subgraphs in the current CPKG and CPKGD, including the edges of "Has, Connect" and the nodes of "cable wires, cable components, tooling, processing features", and also selecting the subgraph with the highest similarity to complete the CPKG processing step pattern layer, and then similarly calculating the similarity by loading the multi-hop sequence between the subgraphs in the current CPKG and CPKGD, including the nodes "cable wires, cable components, tooling, processing features, process steps" and the edges "Has, Connect, Sequence, Parallel, Recycle", and finally selecting the subgraph with the highest similarity to obtain the cable processing process plan. Finally, by completing the CPKG loading and the triples related to the process steps and procedures, the semantic information of "Process-parts, Process-leads, Process-Process, ProcessStep-parts, ProcessStep-leads, Process-ProcessStep" is obtained, and the processing sequence information is converted into RDF / XML format output.

[0114] Embodiment 2

[0115] The present invention also provides a cable processing technology generation system based on knowledge graph, including: a pattern layer creation module, a model building module and a generation module;

[0116] The pattern layer creation module is used to model the cable processing technology knowledge, establish a CPKG model, and based on the CPKG model, convert the relationship between the subtype and attribute of the entity into the relationship between classes and classes, so as to create a cable processing process pattern layer;

[0117] The model building module is used to generate a structure diagram embedded in the cable processing process model layer, create an ICPKG structure tree, and build a distributed embedded reasoning model based on the ICPKG structure tree;

[0118] The generation module is used to generate a cable processing process planning RDF / XML document based on the distributed embedded reasoning model.

[0119] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for generating cable processing technology based on knowledge graph, characterized in that: The following steps are involved: Modeling the cable processing technology knowledge, establishing a CPKG model, converting the relationship between the subtypes and attributes of entities into the relationship between classes based on the CPKG model, and creating a cable processing process model layer; Generate a structure diagram and embed it into the cable processing process model layer, create an ICPKG structure tree, and build a distributed embedded reasoning model based on the ICPKG structure tree; Based on the distributed embedded reasoning model, a cable processing process planning RDF / XML document is generated.

2. The cable processing technology generation method based on knowledge graph according to claim 1 is characterized in that: The methods for modeling cable processing technology knowledge and establishing CPKG model include: Extract the process information contained in the cable processing process documents and 3D models, and describe the CPKG model as: CPKGM={P∪F∪G∪C∪T∪S} Among them, P represents process steps, F represents cable processing features, G represents cable components, C represents processing requirements, T represents tooling, and S represents semantic relationships.

3. The cable processing technology generation method based on knowledge graph according to claim 2 is characterized in that: Based on the CPKG model, the relationship between the subtypes and attributes of entities is converted into the relationship between classes, and the method for creating the cable processing process model layer includes: Preset semantic relationship S; Based on the semantic relationship S, six classes are established: process, processing feature, component, requirement, tool and relationship, and subclasses are established through inheritance; Create the cable processing process model layer based on the six established classes and subclasses established through inheritance; Among them, the semantic relationship S is described as: S={Has, Connect, sequence, Parallel, Recycle} Among them, Has indicates the inclusion relationship, Connect indicates the connection mode of wires and components, Sequence indicates the sequential relationship between process steps, Parallel indicates the parallel relationship between process steps, and Recycle indicates the cycle of a process step; Based on the semantic relationship S, the methods for establishing six categories of processes, processing features, components, requirements, tools and relationships include: (F, P) represents the inclusion relationship between cable processing features and process steps, (P, G) represents the inclusion relationship between process steps and cable components, (P, T) represents the inclusion relationship between process steps and tooling, (P, C) represents the inclusion relationship between process steps and processing requirements, (F, C) represents the inclusion relationship between cable processing features and technical requirements, and (G, F) represents the inclusion relationship between cable components and processing features; The subclasses established through inheritance include: wire and component classes inherit from component classes, geometric features, tolerance dimensions, and material feature classes inherit from processing feature classes, dimensional deviations, heat treatment requirements, and ingredient requirements classes inherit from requirement classes, and process step classes inherit from process classes.

4. The cable processing technology generation method based on knowledge graph according to claim 1 is characterized in that: The generated structure diagram is embedded into the cable processing process model layer, and an ICPKG structure tree is created. Based on the ICPKG structure tree, a method for constructing a distributed embedded reasoning model includes: Construct a cable component structure tree to describe the connection relationship between cable components, and save the cable component structure tree as a triple, and connect each in the form of a cable component structure tree to construct an initial processing technology knowledge graph model ICPKG; wherein the cable component structure tree information includes the ID number of the component, the connected components and wires, and the semantic relationship "Connect"; Based on the initial machining process knowledge graph model ICPKG, scan the machining process documents and CAD models to generate the machining process knowledge graph database CPKGD; A distributed embedded reasoning model is established based on the machining technology knowledge graph database CPKGD.

5. The method for generating cable processing technology based on knowledge graph according to claim 4 is characterized in that: Based on the machining process knowledge graph database CPKGD, the method of establishing a distributed embedded reasoning model includes: The SKGCN method is used for knowledge reasoning to extract multiple processing case graph data from the processing technology knowledge graph database CPKGD; Use CompGCN to learn the aggregate information of each machining process case graph data, generate the embedding vector corresponding to each machining process sequence, and connect the embedding vectors corresponding to different machining process situations; Based on the connected embedding vectors, a capsule network is used to design an evaluation function to learn the implicit association semantics between cable processing instances at the global level and establish a distributed embedding reasoning model.

6. The method for generating cable processing technology based on knowledge graph according to claim 5 is characterized in that: The method of using CompGCN to learn the aggregate information of each processing process case graph data and generate the embedding vector corresponding to each processing process sequence includes: Among them, h o represents the update amount of node o, is the set of neighbors of o’s outgoing edge, is a specific parameter of a relation type, φ represents a composite operation, x s , z r Represent the initial features of node s and relationship r respectively; Among them, h r represents relation embedding, is a learnable transformation matrix that projects all relations into the same embedding mapping space as nodes, describes a learnable scalar weight for a relation and a specific basis, represents a set of learnable basis vectors, represents a set of learnable basis vectors, V b =∈V1,V2,......,V β represents a set of learnable basis vectors, and β is the number of sets of learnable basis vectors; in, represents the node O obtained after k layers, is a specific parameter of the kth layer, represents the node features of the kth layer, represents the relation embedding at the kth layer; in, Represents the representation of relation r after k layers.

7. The method for generating cable processing technology based on knowledge graph according to claim 5 is characterized in that: Using capsule networks, methods for designing an evaluation function include: f(s, r, o)=||capsnet(g([v, r, u]*Ω))|| Among them, s, r, o are the output triplet information, s, o represent nodes, and r represents relationships; similarly, v, r, u represent the input triplet information, v, u represent nodes, and r represents relationships; Ω represents the weight matrix, and the g function represents the nonlinear transformation of the triplet after the weight matrix transformation; caspsnet represents the capsule neural network structure.

8. The method for generating cable processing technology based on knowledge graph according to claim 1, characterized in that: Based on the distributed embedded reasoning model, the method for producing a cable processing process planning RDF / XML document includes: Store the nodes and edges in the CPKG model in XML format; Store edge triplets by traversing the "first and last nodes" information of each edge; The attribute value information of each node is obtained through iterative query.

9. A cable processing technology generation system based on knowledge graph, characterized in that: include: Pattern layer creation module, model building module and generation module; The pattern layer creation module is used to model the cable processing technology knowledge, establish a CPKG model, and based on the CPKG model, convert the relationship between the subtype and attribute of the entity into the relationship between classes and classes to create a cable processing process pattern layer; The model building module is used to generate a structure diagram embedded in the cable processing process model layer, create an ICPKG structure tree, and build a distributed embedded reasoning model based on the ICPKG structure tree; The generation module is used to generate a cable processing process planning RDF / XML document based on the distributed embedded reasoning model.

Images