Knowledge graph-based assembly process autonomous planning method

Abstract

The invention discloses an assembly process autonomous planning method based on a knowledge graph, and relates to the field of intelligent assembly.The method comprises the steps that a product ontology model, an assembly process library and an assembly equipment library are constructed through the knowledge graph, a knowledge graph reasoning query method is adopted, and the assembly process library is constructed by analyzing the geometric feature structure of the assembly process; independent selection of an assembly process is completed; and importing the assembly equipment as an example into the knowledge graph, selecting the assembly equipment capable of completing the assembly task in combination with related parameters of the assembly process, and finally realizing autonomous planning of the assembly process. The method can solve the assembly planning problem of long time consumption in the complex product assembly process planning process, reduces the dependence of the assembly planning process on personnel experience, and greatly improves the assembly planning efficiency.

Description

Technical Field

[0001] The present invention relates to the field of intelligent assembly, and in particular to an autonomous assembly process planning method based on a knowledge graph. Background Art

[0002] The assembly of complex parts is an important stage in the manufacturing of industrial products. According to relevant statistics, in modern manufacturing, assembly hours account for 40% to 60% of the total hours of product research and development and production. As products develop in the direction of personalization and customization, and at the same time have timeliness requirements, traditional automation and information technology have gradually shown their limitations in the assembly process planning stage. In these modes, assembly process planning requires different engineers and systems to design processes, calculate parameters, and select assembly tools. Assembly process planning accounts for a large proportion of the total assembly time. It is not easy to quickly generate assembly process planning for customized products. Summary of the invention

[0003] In order to solve the above technical problems, the present invention proposes an autonomous planning method for assembly process based on knowledge graph. By modeling the assembly product ontology, a product model, a knowledge base and an assembly equipment library are constructed in the knowledge graph. Combined with the knowledge graph reasoning method, the assembly process is autonomously planned and the corresponding assembly equipment is obtained, thereby realizing the automation of assembly process planning, improving the speed of the assembly process, and quickly generating the assembly process plan for customized products.

[0004] The technical solution adopted by the present invention to achieve the above-mentioned purpose is: an assembly process autonomous planning method based on knowledge graph, comprising the following steps:

[0005] Step S1: Based on the knowledge graph, construct a product ontology model to express product geometric structure information;

[0006] Step S2: Building an assembly process library and an assembly equipment library based on the knowledge graph to infer assembly process and equipment matching;

[0007] Step S3: Construct the knowledge graph reasoning rules, use the constructed product ontology model, and combine the assembly process library to infer the product assembly process;

[0008] Step S4: In combination with the product assembly sequence, according to the product body geometric attribute parameters and the assembly equipment library, matching the assembly equipment that can complete the assembly process.

[0009] The step S1 comprises the following steps:

[0010] Step S1.1: Construct a product ontology model in the knowledge graph. The model is divided into three levels, namely product structure layer, assembly process layer and assembly step layer. The product structure layer is used to describe the product geometric information, which is divided into three levels: product component surface, local assembly features, and overall assembly features. The assembly process layer is used to describe the assembly process between products. The assembly step layer is used to describe the steps included in the assembly process and the assembly resources required for each step.

[0011] Step S1.2: Extract the surface information of each part of the product from the three-dimensional model of the product, including surface shape, characteristic parameters, etc., and fill it into the component surface hierarchy in the product structure layer of the product body model as an instance.

[0012] The step S2 comprises the following steps:

[0013] Step S2.1: Build an assembly process library. According to the specific task background, clarify the types of assembly processes that the assembly process library needs to cover, determine the various attributes and component steps involved in each assembly process, and define the relationship between assembly processes and their attributes in the knowledge graph.

[0014] Step S2.2: Build an assembly equipment library, determine the types of equipment and the parameter attributes of each equipment, and import each assembly equipment as an instance into the knowledge graph.

[0015] The S3 step includes the following steps:

[0016] Step S3.1: Determine the relationship between the component surfaces of the parts. According to the position information and geometric parameters of the component surfaces of the parts, determine the spatial position relationship between the component surfaces of the parts, including parallel, overlapping, coaxial, adjacent, etc., and import them as the relationship attributes between the component surfaces into the knowledge graph product component surface layer.

[0017] Step S3.2: Determine the local assembly features of the part. The local assembly features of the part are composed of multiple component surfaces. The component surfaces constitute the local assembly features of the part according to a specific geometric topological structure. Analyze the geometric structure of each assembly feature and describe it with a graph structure. Use the knowledge graph SPARQL to query each graph structure to obtain the local assembly feature information of the part, and import it into the local assembly feature layer of the knowledge graph as an instance.

[0018] Step S3.3: Determine the overall assembly features of the parts. The overall assembly features of the parts are composed of multiple local assembly features according to a specific geometric topological structure. Analyze each assembly process and use a graph structure to describe the spatial position relationship between the local assembly features. Use the knowledge graph SPARQL to query each graph structure to obtain the overall assembly feature information of the parts. Corresponding to the assembly process between the parts, import the overall assembly feature layer and assembly process layer of the knowledge graph as an example

[0019] Step S3.4: Based on the determined assembly process layer instance, combined with the process steps and attributes of each assembly process defined in the assembly process library, the knowledge graph assembly process step layer is obtained, and the attribute parameters involved in each process step are obtained through the knowledge graph product geometry structure layer.

[0020] The step S4 comprises the following steps:

[0021] Step S4.1: According to the assembly sequence of product parts, the parts are gradually assembled into sub-assemblies. According to the assembly process between parts inferred in step S3, the assembly process between the sub-assemblies composed of the parts is determined and updated in the assembly process layer.

[0022] Step S4.2: According to the changes of each assembly process instance in the assembly process layer, the parameter attributes of the corresponding step in the assembly step layer are updated, and finally the assembly equipment that can complete the assembly task of each step is searched through the assembly equipment library.

[0023] The assembly process autonomous planning system based on knowledge graph includes:

[0024] Product ontology model building module, which is used to build a product ontology model based on the knowledge graph to express product geometric structure information;

[0025] Assembly process library and assembly equipment library construction module, which is used to build assembly process library and assembly equipment library based on knowledge graph, and is used to infer assembly process and equipment matching;

[0026] The product assembly process reasoning module is used to construct the knowledge graph reasoning rules, use the built product ontology model and the assembly process library to reason out the product assembly process;

[0027] The assembly equipment matching module is used to match the assembly equipment that can complete the assembly process according to the product assembly sequence, the product body set attribute parameters and the assembly equipment library.

[0028] A device for autonomous planning of assembly processes based on a knowledge graph comprises a memory and a processor; the memory is used to store a computer program; the processor is used to implement the method for autonomous planning of assembly processes based on a knowledge graph when executing the computer program.

[0029] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for autonomous assembly process planning based on a knowledge graph is implemented.

[0030] The present invention has the following beneficial effects and advantages:

[0031] 1. The present invention utilizes the knowledge graph to construct a product ontology model, an assembly process library, and an assembly equipment library, which can provide an accurate model basis for autonomous planning of assembly processes.

[0032] 2. The present invention realizes autonomous planning of product assembly process through the knowledge graph reasoning query function. Compared with traditional manual assembly process planning, the efficiency of assembly process planning can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is the overall flow chart of the present invention;

[0034] Figure 2 This is a schematic diagram of the knowledge graph hierarchy;

[0035] Figure 3 Flowchart of the assembly process reasoning algorithm;

[0036] Figure 4 Update the schematic for assembly process parameters. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments, but the scope of protection claimed in the present invention is not limited to the following specific embodiments.

[0038] This embodiment relates to an assembly process autonomous planning method based on knowledge graph. The overall flow chart is shown in the attached figure. Figure 1 As shown. The following steps are included:

[0039] Step S1: Based on the knowledge graph, construct a product ontology model to express product geometric structure information;

[0040] Step S2: Building an assembly process library and an assembly equipment library based on the knowledge graph to infer assembly process and equipment matching;

[0041] Step S3: Construct the knowledge graph reasoning rules, use the constructed product ontology model, and combine the assembly process library to infer the product assembly process;

[0042] Step S4: In combination with the product assembly sequence, according to the product body geometric attribute parameters and the assembly equipment library, matching the assembly equipment that can complete the assembly process.

[0043] Specifically, step S1 includes the following steps:

[0044] Step S1.1: Construct a product ontology model in the knowledge graph construction software Protégé. The model is divided into three levels, namely product structure layer, assembly process layer and assembly step layer. The product structure layer is used to describe the product geometry information, including three levels: product component surface, local assembly features and overall assembly features. The assembly process layer is used to describe the assembly process between product parts. The assembly step layer is used to describe the steps included in each assembly process and the assembly resources required for each step, such as the attached Figure 2 As shown;

[0045] Step S1.2: Read the product 3D model file, such as the STEP format file, and fill the product surface description information in it as an instance into the established knowledge graph ontology. The surface information belongs to the product composition surface level in the product structure layer.

[0046] Specifically, step S2 includes the following steps:

[0047] Step S2.1: Build an assembly process library, analyze the specific assembly task background, and clarify the types of assembly processes that need to be included in the assembly process library. For example, in this example, radial screw assembly, internal thread assembly, and other assembly processes need to be included. Determine which specific steps each assembly process consists of in the assembly process, the relevant attribute parameters involved in each step, and establish the relationship between assembly processes, assembly steps, and their attributes through Protégé.

[0048] Step S2.2: Build an assembly equipment library, determine the types of equipment and the parameter attributes of each equipment, including basic equipment parameters, assembly capacity parameters, assembly time parameters, etc. Use Protégé to import each equipment as an instance into the knowledge graph ontology.

[0049] The assembly process and equipment knowledge graph (P-DKGM) can be expressed as follows:

[0050] P-DKGM=C∪AC∪R∪I

[0051] Among them, P-DKGM represents the entire assembly process and equipment knowledge graph; C represents the set of related concepts in the field; AC represents the attributes of the concepts in set C; R represents the relationships and constraints between the concepts in set C; I represents the set of instances in the knowledge graph.

[0052] Specifically, step S3 includes the following steps:

[0053] Step S3.1: Determine the relationship between the component surfaces. According to the geometric parameters and spatial position information of the product component surfaces stored in the knowledge graph in step S1.2, determine the positional relationship between the component surfaces, including parallel, overlapping, coaxial, adjacent, etc., and import them into the knowledge graph, which belongs to the product component surface level in the product structure layer.

[0054] Step S3.2: Determine the local assembly features of the parts. The local assembly features of the parts are composed of multiple component surfaces according to a specific geometric topological structure. The geometric structure of each local assembly feature is analyzed and described with a graph structure. For example, a screw connection can be composed of two adjacent cylindrical surfaces and two overlapping planes. Use the knowledge graph SPARQL to query the geometric structure corresponding to each assembly feature, obtain the local assembly feature information of the parts, and import it as an instance into the local assembly information hierarchy of the parts in the product structure layer.

[0055] Step S3.3: Determine the overall assembly features of the part. The overall assembly features of the part are composed of multiple local assembly features according to a specific geometric topological structure. For example, for radial screw assembly, its overall assembly features include multiple coplanar screw assembly features. Analyze the geometric structure of each assembly process and describe it with a graph structure. Use the knowledge graph SPARQL to query the graph structure corresponding to each assembly process to obtain the overall assembly features of the part. Corresponding to the assembly process between parts, import the overall assembly feature layer and assembly process layer of the knowledge graph product structure layer as an example, as shown in the attached figure. Figure 3 shown.

[0056] Step S3.4: Based on the instances in the assembly process layer and the definition of each assembly process attribute in the assembly process library, the relevant information of the assembly step layer is obtained. The parameters involved in each step are obtained through the knowledge graph product structure layer data.

[0057] Specifically, step S4 includes the following steps:

[0058] S4.1: According to the assembly sequence of product parts, the parts are gradually assembled into sub-assemblies. The assembly processes between sub-assemblies correspond to the assembly processes between parts inferred in step S3. Each assembly process instance is updated at the assembly process level, as shown in the attached figure. Figure 4 shown.

[0059] Step S4.2: According to the changes of each assembly process instance in the assembly process layer, update the parameters involved in each step in the assembly common step layer, and use SPARQL query statements to search for matching assembly equipment that can complete the assembly task in combination with the assembly equipment library.

[0060] The above specific implementations can be partially adjusted in different ways by those skilled in the art without departing from the principles and purposes of the present invention, and some modifications and changes to the invention should also fall within the scope of protection of the claims of the present invention. The scope of protection of the present invention shall be subject to the claims and shall not be limited by the above specific implementations, and each implementation within its scope shall be subject to the constraints of the present invention.

Claims

1. The method of autonomous assembly process planning based on knowledge graph is characterized by: The following steps are involved: Step S1: Based on the knowledge graph, a product ontology model is constructed to express product geometric structure information; Step S2: Building an assembly process library and an assembly equipment library based on the knowledge graph to infer assembly process and equipment matching; Step S3: Construct the knowledge graph reasoning rules, use the constructed product ontology model, and combine the assembly process library to infer the product assembly process; Step S4: In combination with the product assembly sequence, according to the product body set attribute parameters and the assembly equipment library, matching the assembly equipment that can complete the assembly process.

2. The method for autonomous planning of product assembly process based on knowledge graph according to claim 1, characterized in that: The step S1 comprises the following steps: Step S1.1: construct a product ontology model in the knowledge graph; the product ontology model is divided into three levels, namely, product structure layer, assembly process layer and assembly step layer; the product structure layer is used to describe product geometric information, which is divided into three levels: product component surface, local assembly features, and overall assembly features; the assembly process layer is used to describe the assembly process between products; the assembly step layer is used to describe the steps included in the assembly process and the assembly resources required for each step; Step S1.2: extracting surface information of each part of the product from the three-dimensional model of the product, the surface information including surface shape and characteristic parameters, and filling the surface information as an instance into the product component surface hierarchy in the product structure layer of the product body model.

3. The method for autonomous assembly process planning based on knowledge graph according to claim 1 is characterized in that: The step S2 comprises the following steps: Step S2.1: Build an assembly process library: According to the specific task background, clarify the types of assembly processes that the assembly process library needs to cover, determine the various attributes and component steps involved in the assembly process for each assembly process, and define the relationship between the assembly process and its attributes in the knowledge graph; Step S2.2: Build an assembly equipment library, determine the types of equipment and the parameter attributes of each equipment, and import each assembly equipment as an instance into the knowledge graph.

4. The method for autonomous assembly process planning based on knowledge graph according to claim 1 is characterized in that: The step S3 comprises the following steps: Step S3.1: Determine the relationship between the component surfaces: According to the position information and geometric parameters of the component surfaces, determine the spatial position relationship between the component surfaces, and import it as the relationship attribute between the component surfaces into the product component surface level in the knowledge graph product structure layer; Step S3.2: Determine the local assembly features of the part: The local assembly features of the part are composed of multiple component surfaces, and the component surfaces form the local assembly features of the part according to a specific geometric topological structure; analyze the geometric structure of each assembly feature and describe it with a graph structure, use the knowledge graph SPARQL to query each graph structure, obtain the local assembly feature information of the part, and import it into the knowledge graph local assembly feature layer as an example; Step S3.3: Determine the overall assembly features of the parts: The overall assembly features of the parts are composed of multiple local assembly features according to a specific geometric topological structure; analyze each assembly process, use a graph structure to describe the spatial position relationship between the local assembly features, use the knowledge graph SPARQL to query each graph structure, obtain the overall assembly feature information of the parts, corresponding to the assembly process between the parts, and import it into the overall assembly feature layer and assembly process layer in the product structure layer of the knowledge graph as an example; Step S3.4: Based on the determined assembly process layer instance, combined with the process steps and attributes of each assembly process defined in the assembly process library, the knowledge graph assembly process step layer is obtained, and the attribute parameters involved in each process step are obtained through the knowledge graph product geometry structure layer.

5. The method for autonomous assembly process planning based on knowledge graph according to claim 1 is characterized in that: The step S4 comprises the following steps: Step S4.1: According to the assembly sequence of product parts, the parts are gradually assembled into sub-assemblies. According to the assembly process between the parts inferred in step S3, the assembly process between the sub-assemblies composed of the parts is determined, and updated in the assembly process layer; Step S4.2: According to the changes of each assembly process instance in the assembly process layer, the parameter attributes of the corresponding step in the assembly step layer are updated, and finally the assembly equipment that can complete the assembly task of each step is searched through the assembly equipment library.

6. The assembly process autonomous planning system based on knowledge graph is characterized by: include: Product ontology model building module, which is used to build a product ontology model based on the knowledge graph to express product geometric structure information; Assembly process library and assembly equipment library construction module, which is used to build assembly process library and assembly equipment library based on knowledge graph, and is used to infer assembly process and equipment matching; The product assembly process reasoning module is used to construct the knowledge graph reasoning rules, use the built product ontology model and the assembly process library to reason out the product assembly process; The assembly equipment matching module is used to match the assembly equipment that can complete the assembly process according to the product assembly sequence, the product body set attribute parameters and the assembly equipment library.

7. An assembly process autonomous planning device based on knowledge graph, characterized in that: It comprises a memory and a processor; the memory is used to store a computer program; the processor is used to implement the assembly process autonomous planning method based on knowledge graph as described in any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method for autonomous assembly process planning based on a knowledge graph as described in any one of claims 1 to 5 is implemented.

Images