Automatic configuration method for discrete manufacturing production line equipment based on knowledge graph

Abstract

The invention discloses a discrete manufacturing production line equipment automatic configuration method based on a knowledge graph, and the method comprises the following steps: 1, collection and fusion of production line data: dividing the type of a discrete manufacturing production line, determining key parameters of equipment, refining the hierarchical structure of the production line, and constructing an equipment database; 2, defining production line resource requirements: extracting product requirement keywords, analyzing a mapping relation between equipment parameters and product requirements, and creating a requirement conversion system; step 3, construction and storage of a knowledge graph: constructing a production line equipment configuration knowledge graph containing a hierarchical structure and an association relationship; 4, defining a production line equipment selection evaluation standard: dividing parameter grades and assigning scores according to equipment key parameters and cost, and calculating and generating a comprehensive score of an equipment configuration scheme; 5, production line equipment configuration based on the knowledge graph: outputting equipment nodes matched with the knowledge graph, and selecting an optimal scheme in combination with an evaluation standard; and 6, outputting and displaying a configuration result.

Description

Technical Field

[0001] The present invention belongs to the technical field of discrete manufacturing production lines, and specifically provides a method for automatically configuring discrete manufacturing production line equipment based on a knowledge graph. Background Art

[0002] Discrete manufacturing industries, such as machinery manufacturing, automobile manufacturing, and aerospace, occupy a vital position in my country's economic system and are an integral part of the real economy. With the development of the global economy and the improvement of people's living standards, the demand for discrete manufacturing products is increasing. The rapid development of the automotive, machinery and equipment, electronic equipment, and aerospace industries has further promoted the development of the discrete manufacturing industry.

[0003] Equipment configuration for discrete manufacturing production lines is crucial for ensuring production efficiency, product quality, and cost control. Traditional discrete manufacturing production line equipment configuration requires demand analysis, process flow analysis, technical parameter evaluation, cost analysis, supplier evaluation, equipment compatibility analysis, quality control, on-site inspections, and consideration of future developments. This involves numerous factors, a wide range of knowledge, and a complex system.

[0004] The difficulties in configuring equipment for discrete manufacturing production lines are mainly reflected in the following aspects:

[0005] (1) Information asymmetry leads to incomplete configuration considerations. The information provided by equipment suppliers may not be comprehensive, making it difficult for users to obtain the true performance and applicability of the equipment.

[0006] (2) Complex demand analysis leads to a complex selection process. The production requirements of different products vary, involving a variety of processes and technical parameters.

[0007] (3) Trade-offs among multiple options lead to difficult decision-making. There are many types of equipment on the market, and users need to make trade-offs in multiple dimensions, such as performance, cost, and reliability.

[0008] (4) Rapid technological updates make it difficult for users to keep up with the latest market trends. With the rapid development of technology, new equipment and new technologies emerge in an endless stream.

[0009] To address the above problems, modern tool knowledge graphs are used for automatic configuration of discrete manufacturing production line equipment. The automatic configuration method of discrete manufacturing production line equipment based on knowledge graphs can improve the efficiency of production line equipment configuration and select the most suitable equipment, thereby improving production efficiency, reducing costs and ensuring product quality to adapt to complex production environments and rapidly changing market demands.

[0010] The concept of knowledge graphs was proposed by Google in 2012, primarily for use in search engines. Knowledge graphs describe the semantic network of relationships between physical entities from a relational perspective, and are used to store and represent knowledge about a system's structure, attributes, relationships, and behavioral rules. There are four advantages to using knowledge graphs for production line equipment configuration:

[0011] (1) Information integration and visualization. Knowledge graphs can integrate information from different sources, including device performance, user feedback, and industry standards, to provide a comprehensive view of device information and help users make more informed decisions. This solves the problem of incomplete configuration considerations due to information asymmetry.

[0012] (2) Relationship analysis and reasoning. Knowledge graphs can analyze the relationships and mutual influences between devices, helping users understand the role of different devices in the production process and thus optimize selection. Query instructions can be used to match the most appropriate device selection based on the user's specific needs and historical data, reducing the bias of manual judgment. This solves problems such as complex selection processes and difficult decision-making.

[0013] (3) Dynamic update and adaptability. The knowledge graph can be updated in real time to reflect changes in new equipment and new technologies on the market, helping users obtain the latest information in a timely manner and reducing the risks brought by technological updates.

[0014] (4) Standardization and normalization. Through the knowledge graph, a standardized model for equipment selection can be established, providing unified evaluation criteria and simplifying the comparison and selection process.

[0015] Discrete manufacturing production line equipment configuration faces multiple challenges. Knowledge graphs can effectively integrate information, provide intelligent recommendations, analyze relationships, and dynamically update, significantly improving the efficiency and accuracy of equipment selection. Knowledge graph-based discrete manufacturing production line equipment configuration methods not only help manufacturers make better decisions in complex market environments but also improve the overall efficiency and flexibility of production lines. However, this method is not mature in the market and its application is less than ideal.

[0016] In summary, there is an urgent need to propose an automatic configuration method for discrete manufacturing production line equipment based on knowledge graph to optimize the configuration and evaluation methods of production line equipment. Summary of the Invention

[0017] In view of this, the purpose of the present invention is to provide a method for automatic configuration of discrete manufacturing production line equipment based on knowledge graph, so as to optimize the configuration and evaluation method of production line equipment.

[0018] In order to achieve the above object, the present invention provides the following technical solutions:

[0019] A method for automatic configuration of discrete manufacturing production line equipment based on knowledge graph, comprising the following steps:

[0020] Step 1: Production line data collection and integration: By classifying discrete manufacturing production lines, determining key equipment parameters, and refining the production line hierarchy, an equipment database is constructed that includes equipment type, technical parameters, cost, compatibility, and supplier information.

[0021] Step 2: Define production line resource requirements: Extract product requirement keywords, analyze the mapping relationship between equipment parameters and product requirements, and create a demand conversion system that automatically converts product requirements into equipment parameter requirements;

[0022] Step 3: Knowledge graph construction and storage: Using equipment as nodes, equipment attributes as node parameters, and inter-equipment compatibility and process associations as edges, a production line equipment configuration knowledge graph with a hierarchical structure and associations is constructed and stored in a graph database and / or RDF format.

[0023] Step 4: Define the evaluation criteria for production line equipment selection: Divide the equipment into parameter levels and assign scores based on key parameters and costs, and generate a comprehensive score for the equipment configuration plan through normalized calculation;

[0024] Step 5: Production line equipment configuration based on the knowledge graph: According to the output of the demand conversion system, the equipment nodes in the knowledge graph are matched, and the production line is hierarchically configured according to workstations, conveying systems, and signal transmission systems. The optimal solution is selected based on the evaluation criteria.

[0025] Step 6: Output and display configuration results: Output the configuration results in the form of knowledge graph visualization and device parameter list.

[0026] Furthermore, in step 1, the method steps for collecting and integrating production line data are as follows:

[0027] 11) Classification of discrete manufacturing production line types

[0028] Based on the production line implementation plan, discrete manufacturing production lines are divided into basic types according to their functions;

[0029] 12) Determine key equipment parameters

[0030] Considering the equipment's technical parameters, cost, compatibility, and suppliers, analyze the production process based on the product characteristics of a specific production line and determine the type and function of the equipment;

[0031] 13) Refine the hierarchical structure of the production line

[0032] The typical structure of each type is extracted as a class, and its equipment or devices as a subclass. Multi-level attribute parameters including equipment type, technical parameters, cost, compatibility and supplier are set for each class and subclass.

[0033] Furthermore, in step 2, the method steps for defining production line resource requirements are as follows:

[0034] 21) Product requirement keyword extraction

[0035] Extract dimensional accuracy, defect detection indicators, and budget constraint keywords from product requirements through natural language processing;

[0036] 22) Equipment parameter requirement analysis

[0037] Establish a mathematical mapping model between equipment parameters and product requirements, including calculating sensor resolution based on detection accuracy requirements and deriving equipment processing speed based on production cycle requirements;

[0038] 23) Create a product requirements conversion system

[0039] The matching rule library between product requirement keywords and equipment parameters is optimized through machine learning algorithms to create a product requirement conversion system.

[0040] Furthermore, in step 3, the method steps for constructing and storing the knowledge graph are as follows:

[0041] 31) Build a production line knowledge graph ontology

[0042] Based on the equipment database constructed in step 1, the production line knowledge graph ontology is constructed with the equipment as the ontology and the equipment type, technical parameters, cost, compatibility and supplier information as the attributes;

[0043] 32) Define the production line knowledge graph data structure

[0044] The ontology is used as a node and a triple storage mode is adopted, including two structures: "node-attribute-attribute value" and "node-relationship-node". The relationship between entities is set as the same-level relationship and cross-level relationship. The same-level relationship represents the flow relationship between equipment in the process of production, and the cross-level relationship represents the inclusion relationship between different levels. The relationship edge contains the attribute values of process sequence, compatibility matching degree and cost association weight.

[0045] 33) Knowledge graph storage and update

[0046] The constructed production line knowledge graph is stored in a graph database and / or RDF format, and the production line knowledge graph is synchronously updated after the production data is updated; the graph database is used to provide graph traversal and path query functions to quickly retrieve and analyze data in the knowledge graph.

[0047] Furthermore, in step 4, the method steps for defining the production line equipment selection evaluation criteria are as follows:

[0048] 41) Determine equipment evaluation objectives

[0049] Evaluate the equipment parameters and costs obtained from the system conversion based on product requirements;

[0050] 42) Implement production line equipment selection evaluation

[0051] The key parameters and cost information of the equipment are obtained through the knowledge graph query language. The key parameters and costs are classified into intervals and graded. The normalized calculation is used to obtain the comprehensive score of the production line equipment configuration plan.

[0052] 43) Select production line equipment

[0053] The production line equipment configuration plan with the highest score is selected for production line configuration.

[0054] Furthermore, in step 5, the method steps for configuring production line equipment based on the knowledge graph are as follows:

[0055] 51) Determine configuration goals

[0056] The production line is hierarchically configured according to the workstation, conveying system, and signal transmission system. The product requirement conversion system constructed in step 2 is used to obtain the production line type. The production line knowledge graph constructed in step 3 is used to query the basic composition of the production line and the composition of the workstation subsystem, and the equipment or components of the workstation subsystem are evaluated.

[0057] 52) Layered Configuration

[0058] Using the production line equipment selection evaluation criteria described in step 4, select the workstation subsystem configuration solution with the highest score for workstation subsystem configuration, select the conveyor system configuration solution with the highest score for conveyor system configuration, and select the signal transmission system configuration solution with the highest score for signal transmission system configuration;

[0059] 53) Production line configuration

[0060] Based on the workstation subsystem configuration, conveying system configuration and signal transmission system configuration obtained in step 52), the discrete manufacturing production line configuration is preliminarily completed.

[0061] Furthermore, in step 6, the configuration result output and display includes graph database display and device list output; the graph database display displays the knowledge graph database of the configured discrete manufacturing production line through a query language, and displays the production line system composition and attribute parameters in the knowledge graph database;

[0062] The equipment list output will include the equipment model, manufacturer, technical parameters, cost of the equipment parameters to form a list and output it in list form to facilitate procurement and configuration.

[0063] The beneficial effects of the present invention are:

[0064] The present invention is an automatic configuration method for discrete manufacturing production line equipment based on knowledge graph, including production line data collection and fusion, production line resource demand definition, knowledge graph construction and storage, production line equipment selection and evaluation standard definition based on knowledge graph, production line configuration based on knowledge graph and configuration result specification output and display, etc. In view of the problems of incomplete configuration considerations, complex selection process, difficult decision-making, and difficulty in keeping up with the latest market trends in traditional methods of discrete manufacturing production line equipment configuration, the present invention can clearly and accurately configure and evaluate production lines based on knowledge graph, improve production line configuration efficiency, enhance production line stability and product quality, and thus improve production efficiency and market change responsiveness.

[0065] Specifically, the present invention associates the entity knowledge of each link of production line production in different application scenarios by constructing a knowledge graph, stores the types and functions of existing production line equipment, equipment technical parameters, equipment costs, equipment compatibility, equipment suppliers, etc., and can be updated and expanded. It provides a realistic basis for the automated configuration of new production line equipment, solves problems such as incomplete consideration of production line configuration, complex selection process, difficult decision-making, and difficulty in keeping up with the latest market trends, and greatly shortens the preliminary work of production line configuration. The comprehensive performance of production line configuration is ensured by defining the resource requirements of the production line and the evaluation criteria for the selection of production line equipment. Most importantly, the method of the present invention is universal and has a wide range of applications, which greatly reduces the technical threshold and cost of production line configuration. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to make the purpose, technical solutions and beneficial effects of the present invention more clear, the present invention provides the following drawings for illustration:

[0067] Figure 1 This is a flow chart of the method for automatic configuration of discrete manufacturing production line equipment based on knowledge graph of the present invention;

[0068] Figure 2 Result graphs defining production line resource requirements for a vision inspection system;

[0069] Figure 3 It is the composition of the welding production line;

[0070] Figure 4 An example of a configuration for a signal transmission system;

[0071] Figure 5 An example of a configuration for a conveying system;

[0072] Figure 6 This is the composition tree diagram of the arc welding robot workstation;

[0073] Figure 7 Visualize the knowledge graph for arc welding robot workstation;

[0074] Figure 8 Graphic database display for selecting the best solution for arc welding robot workstation;

[0075] Figure 9 A list showing the best options for selecting an arc welding robot workstation. DETAILED DESCRIPTION

[0076] The present invention will be further described below with reference to the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it. However, the embodiments are not intended to limit the present invention.

[0077] like Figure 1 As shown, the automatic configuration method of discrete manufacturing production line equipment based on knowledge graph in this embodiment includes the following steps.

[0078] Step 1: Production line data collection and integration: By classifying discrete manufacturing production lines, determining key equipment parameters, and refining the production line hierarchy, an equipment database is constructed that includes equipment type, technical parameters, cost, compatibility, and supplier information.

[0079] In this embodiment, the method steps for collecting and fusing production line data are as follows.

[0080] 11) Classification of discrete manufacturing production line types

[0081] Consult the information to obtain the currently commonly used production line implementation plan. Based on the production line implementation plan, the discrete manufacturing production line is divided into basic types according to its function. Specifically, the currently commonly used production line implementation plan can be obtained through the Internet, books and existing production knowledge, and the discrete manufacturing production line is divided into basic types according to its function.

[0082] 12) Determine key equipment parameters

[0083] Consider the equipment's technical parameters (such as processing accuracy, speed, stability, reliability, etc.), costs (including price, shipping, installation costs, equipment energy consumption, maintenance costs, labor costs, etc.), compatibility, and suppliers. Analyze the production process based on the product characteristics of the specific production line to determine the type and function of the equipment. Specifically, you can analyze the production process based on the product characteristics of the specific production line to determine the type and function of the equipment, and consider the equipment's technical parameters, such as processing accuracy, speed, stability, reliability, etc. Also consider the equipment's cost, including price, shipping, installation costs, equipment energy consumption, maintenance costs, labor costs, etc. Finally, consider the equipment's compatibility and supplier. Collect user reviews and feedback on the equipment from e-commerce platforms, professional forums, and social media. Obtain standards and specifications for relevant industries to understand the compliance requirements of the equipment.

[0084] 13) Refine the hierarchical structure of the production line

[0085] Each typical structure is extracted as a class, and its equipment or components as subclasses. Each class and subclass is then assigned multi-level attribute parameters, including equipment type, technical parameters, cost, compatibility, and supplier. Specifically, each basic production line can be divided into several typical systems. Each typical system is extracted to form a class, and the corresponding equipment or components are each a subclass. These classes and subclasses are each assigned parameters, such as equipment type and function parameters, equipment technical parameters, equipment cost parameters, equipment compatibility parameters, and equipment supplier parameters. All of this information can be stored as input into the knowledge graph.

[0086] Step 2: Define production line resource requirements: Extract product requirement keywords, analyze the mapping relationship between equipment parameters and product requirements, and create a demand conversion system that automatically converts product requirements into equipment parameter requirements.

[0087] Discrete manufacturing production line configuration involves multiple factors, including demand analysis, process flow analysis, technical parameter analysis, cost analysis, and supplier analysis. Each factor can yield the optimal equipment configuration for that factor, but this often requires considering global factors and making trade-offs. Furthermore, demand analysis and process flow analysis often cannot be directly linked to equipment parameters, requiring designers to leverage their on-the-job experience and translate them into equipment parameter requirements. Based on these considerations, product requirement keyword extraction and equipment parameter requirement analysis are necessary, along with the creation of a product requirement conversion system. First, based on the product requirements to be produced by the production line, product requirement keywords are summarized. Second, the specific equipment parameters determine the quality of the product produced. Based on these specific equipment parameters, the achievable product requirements are analyzed. Finally, based on the obtained product requirement keywords and their corresponding equipment parameters, a requirements conversion system is created. This system automatically extracts product requirement keywords from the input product requirements and generates matching equipment parameter requirements, providing a basis for production line configuration and equipment selection.

[0088] Specifically, in this embodiment, the method steps for defining production line resource requirements are as follows.

[0089] 21) Product requirement keyword extraction

[0090] Natural language processing is used to extract keywords such as dimensional accuracy, defect detection indicators, and budget constraints from product requirements.

[0091] 22) Equipment parameter requirement analysis

[0092] The specific parameters of a device determine the quality of the product produced. Based on these specific device parameters, the achievable product requirements are analyzed. This embodiment establishes a mathematical mapping model between device parameters and product requirements, including calculating sensor resolution based on detection accuracy requirements and deriving device processing speed based on production cycle requirements.

[0093] 23) Create a product requirements conversion system

[0094] By optimizing the matching rule base between product requirement keywords and equipment parameters through machine learning algorithms, a product requirement conversion system was created. The product requirement conversion system can automatically extract product requirement keywords and obtain matching equipment parameter requirements based on the input of product requirements, providing a basis for production line configuration and equipment selection.

[0095] Specifically, such as Figure 2As shown, this embodiment uses the example of a production line requiring a visual inspection system. The visual inspection system is used to inspect finished products measuring 10x12cm (with a field of view of 12x10mm). Common defects are known to be approximately 0.02mm diameter holes or 0.01mm wide gaps. With a budget of 100,000 yuan, a machine vision system is required. This requires programming and machine learning to build a product requirement translation system. The system extracts keywords based on user input, such as: visual inspection, product size 10x12cm, diameter 2mm, width 1mm, budget 10,000 yuan. Through built-in settings and machine learning, the system can infer specific equipment parameter requirements from the required keywords. For example, based on the minimum defect size of 0.01mm, the system infers a resolution of 0.01mm. Based on the field of view, the system infers that the resolution must be greater than (12 / 0.01) x (10 / 0.01) = 1,200,000. Based on the budget, the system infers that the total equipment cost must be at least less than 100,000 yuan.

[0096] Step 3: Knowledge graph construction and storage: Using equipment as nodes, equipment attributes as node parameters, and compatibility and process association relationships between equipment as edges, a production line equipment configuration knowledge graph containing hierarchical structures and association relationships is constructed and stored in a graph database and / or RDF format.

[0097] Specifically, we build a knowledge graph for production line equipment configuration based on the equipment configuration of traditional discrete manufacturing production lines in the market. This knowledge graph stores parameters and values, including equipment type and function, technical specifications, cost, compatibility, and supplier. Some attributes have sub-attributes and sub-attribute values, and devices are interconnected and matched.

[0098] To build a device knowledge graph, you can use the knowledge graph software available on the market. First, you need to clarify the ontology and attributes, with the device as the ontology, and the type, function, technical parameters, cost parameters, and supplier information of the device as the attributes. The ontology has attributes, and similarly, attributes can also have sub-attributes. During the specific construction, it is necessary to add relationships between various attributes, such as mutually exclusive relationships, independent relationships, inclusion relationships, reversible relationships, etc. The hierarchical logic must be clear and unambiguous to facilitate subsequent query operations. After a single device knowledge graph is constructed, other types of device knowledge graphs can be constructed, and relationships can be defined between these devices. For example, in a machine vision system, the lens and camera are used in conjunction with each other. The two are independent and work together. The two can be associated together in the knowledge graph to form a device chain, and the association condition is that the interface types of the two are the same.

[0099] Specifically, in this embodiment, the method steps for constructing and storing the knowledge graph are as follows.

[0100] 31) Build a production line knowledge graph ontology

[0101] Starting from demand analysis, process flow analysis, technical parameter analysis, cost analysis, supplier analysis and other aspects, the equipment database constructed according to step one takes the equipment as the ontology and the equipment type, technical parameters, cost, compatibility and supplier information as attributes. Each attribute has sub-attributes to construct the production line knowledge graph ontology.

[0102] 32) Define the production line knowledge graph data structure

[0103] The model uses ontologies as nodes and employs a triple storage model, encompassing two structures: "node-attribute-attribute value" and "node-relationship-node." Entity relationships are structured as either intra-level or inter-level. Intra-level relationships represent the flow between equipment during the production process, while inter-level relationships represent inclusion relationships between different levels. Relationship edges include attribute values such as process sequence, compatibility matching, and cost-related weights.

[0104] 33) Knowledge graph storage and update

[0105] The constructed production line knowledge graph is stored in a graph database and / or RDF format, and is updated synchronously with production data updates to ensure that the proposed method can adapt to new technologies and requirements. The graph database is used to provide graph traversal and path query capabilities to quickly retrieve and analyze data in the knowledge graph.

[0106] Specifically, such as Figure 3 As shown, this embodiment uses the configuration of a welding production line as an example. A welding production line generally consists of a workpiece preparation workstation, a welding workstation, an inspection workstation, a shipping workstation, and a workstation-related system. Simply configuring these workstations and the associated equipment between them completes the production line configuration. Using the most typical arc welding robot workstation as an example, equipment configuration based on the knowledge graph can be performed similarly for other workstations and their associated equipment, and will not be further elaborated here.

[0107] An arc welding robot workstation primarily consists of a robot system, welding power supply system, welding fixture system, positioner, sensors, safety system, host computer, control system, and fume and dust removal system. The robot system includes the robot body, robot control cabinet, and teach pendant. The welding power supply system includes the welding machine, wire feeder, welding gun, wire spool holder, and gun cleaner. Sensors include proximity sensors, vision sensors, and welding sensors. The safety system includes fencing, safety light barriers, and safety locks. The control system includes a PLC control cabinet, HMI touch screen, and operating console. The fume and dust removal system includes self-cleaning dust removal equipment, fume hoods, and piping. To maximize efficiency, the manufacturer does not configure the entire production line. Instead, they configure key equipment. The arc welding robot workstation's robot system, welding power supply system, positioner, and sensors are configured based on product requirements. The safety system, control system, and fume and dust removal system are relatively mature and versatile and can be purchased separately. This article focuses on the configuration of the robot system, welding power supply system, positioner, and sensors for the arc welding robot workstation. Based on the composition of the arc welding robot workstation, including the specific components of the robot system, welding power supply system, positioner, and sensors, multiple ontologies can be constructed, with a hierarchical relationship between ontologies. The next step is to determine the ontologies' attributes. For example, the robot ontologies have the following attributes: robotic arm structure: parallelogram (suitable for spot welding and handling robots) or pendulum (suitable for arc welding, flame cutting, plasma arc welding, laser welding and cutting); repeatability: 0.05-0.1mm (better than 0.1mm); payload: 3-8kg; and motion radius: 1.4-1.6m. A specific example has parameters such as: model IRB1410, payload: 5kg, motion radius: 1444mm, repeatability: 0.05mm. Similarly, the arc welding power supply typically consists of a 500A power supply with a 100% duty cycle or a 600A or larger battery with a 60% duty cycle. Wire feeders are available in integrated and detachable types (suitable for automatic welding gun replacement). Welding guns include ordinary gooseneck welding guns and wire drawing welding guns. Sensors include proximity sensors, visual sensors, and welding sensors. Positioners are available in single-axis, two-axis (the most common), three-axis, and multi-axis types. Examples include: model IRBP-A, with a load capacity of 250kg, a height of 900mm, and a diameter of 1000mm.

[0108] Each ontology is stored as a triple, consisting of two main types: "node, attribute, attribute value" and "node, relationship, node." Entity relationships can be either intra-level or inter-level. Intra-level relationships represent the flow between equipment in a production process, while inter-level relationships represent inclusion relationships between different levels. Relationships can also be configured with attributes and attribute values.

[0109] For example, an ontology is a node, each with attributes such as equipment specifications, cost, and model. Nodes are connected by relationships, such as "has" and "contains." Based on the key information of an arc welding robot workstation and its equipment specifications, cost, and model, a knowledge graph can be constructed. The more comprehensive and accurate the known information, the larger the knowledge graph.

[0110] Step 4: Definition of Evaluation Criteria for Production Line Equipment Selection: When the knowledge graph data volume is sufficiently large, multiple options for satisfying user requirements for production line equipment configurations may be generated, requiring evaluation of each option. In this embodiment, equipment configuration options are graded and scored based on key parameters and costs, and a comprehensive score is generated through normalized calculation.

[0111] When the amount of knowledge graph data is large enough, there will be multiple solutions for the production line configuration that meets the requirements based on user needs. At this time, each solution needs to be evaluated. The evaluation is mainly based on the key equipment parameters and costs obtained from the product requirements analysis. Key parameters such as range, accuracy, service life, etc. The key parameters and costs are graded according to certain intervals, assigned points according to the levels, and normalized. The key parameters and cost information of the equipment are obtained through the knowledge graph query language, the corresponding scores are matched, and the total score of the solution is calculated. The higher the score, the better the overall performance of the production line. Based on the equipment configuration solution obtained through the knowledge graph query language, according to the scores of each solution obtained, the equipment with the highest score among the alternative solutions is selected to configure the production line.

[0112] Specifically, in this embodiment, the method steps for defining the production line equipment selection evaluation criteria are as follows.

[0113] 41) Determine equipment evaluation objectives

[0114] Evaluate the equipment parameters and costs obtained based on the product requirements conversion system, including key parameters such as range, accuracy, service life, etc.

[0115] 42) Implement production line equipment selection evaluation

[0116] The key parameters and cost information of the equipment are obtained through the knowledge graph query language, the key parameters and costs are classified into intervals and graded, and the normalized calculation is used to obtain the comprehensive score of the production line equipment configuration plan. The higher the score, the better the overall performance of the production line.

[0117] 43) Select production line equipment

[0118] Based on the equipment configuration scheme obtained through the knowledge graph query language, according to the scores of each scheme obtained in step 42), the production line equipment configuration scheme with the highest score is selected for production line configuration.

[0119] Step 5: Knowledge-Graph-Based Production Line Equipment Configuration: Based on the output of the requirements conversion system, the equipment nodes in the knowledge graph are matched. The production line is configured hierarchically by workstation, conveyor system, and signal transmission system. The optimal solution is selected based on evaluation criteria. Specifically, the product requirements conversion system first determines the desired production line type. The knowledge graph's query language is used to query the production line to determine its basic components. Specific workstations are then queried to determine the subsystem components of the workstation. The equipment or components of these subsystems are the targets to be configured.

[0120] In this embodiment, the method steps for configuring production line equipment based on knowledge graph are as follows.

[0121] 51) Determine configuration goals

[0122] The production line is configured hierarchically based on workstations, conveying systems, and signal transmission systems. The product requirements conversion system constructed in step 2 is used to determine the production line type. The production line knowledge graph constructed in step 3 is used to query the basic components of the production line and the workstation subsystems. The equipment or components of the workstation subsystems are then evaluated. For example, an arc welding robot workstation includes an arc welding robot system, a welding power supply system, a control system, and a safety system. The subsystem configuration is performed based on the equipment or components defined in step 4 as part of the production line equipment selection and evaluation criteria.

[0123] 52) Layered Configuration

[0124] Use the production line equipment selection evaluation criteria in step 4 and select the workstation subsystem configuration solution with the highest score to configure the workstation subsystem. Figure 4 As shown in , the transport system configuration scheme with the highest score is selected for transport system configuration. Figure 5 As shown in , the signal transmission system configuration scheme with the highest score is selected for signal transmission system configuration. Figure 6-7 As shown, the arc welding robot workstation includes an arc welding robot system, a welding power supply system, a control system, a safety system, etc., and the subsystem configuration is performed based on the equipment or components obtained in step four.

[0125] 53) Production line configuration

[0126] Based on the workstation subsystem configurations, conveyor system configurations, and signal transmission system configurations obtained in step 52), the discrete manufacturing production line configuration is initially completed. Specifically, each subsystem is composed of various equipment components. According to step 4, multiple subsystems may meet the requirements. Similarly, the subsystems are evaluated and selected to complete the workstation configuration. Similarly, the configuration of each workstation component, as well as the conveyor system and signal transmission system configuration between workstations, is completed to initially complete the discrete manufacturing production line configuration.

[0127] Specifically, by analyzing product requirements using the conversion system created in step 2, specific equipment parameter requirements can be obtained, which can then be queried and reasoned with the help of the knowledge graph language (SPARQL, Cypher).

[0128] Use Cypher query language to extract information from the knowledge graph. For example, to query all devices with a load capacity greater than 100kg:

[0129] MATCH(d:Device)

[0130] WHERE d.loadCapacity>100

[0131] RETURNd.name,d.loadCapacity,d.workingRange

[0132] The knowledge graph query language is powerful and can query data with labels, attributes, specific attribute parameters, and node relationships.

[0133] By analyzing the similarities between devices, we recommend devices with similar performance or complementary functions. For example, if a user searches for "welding robots with a load capacity of 150kg", the system can recommend other robots with a load capacity between 120kg and 180kg, such as:

[0134] The KUKA KR 30 has a load capacity of 120 kg and the FANUC M-30iA has a load capacity of 180 kg.

[0135] Recommend equipment based on user needs. For example, if a user wants equipment with a load capacity between 10kg and 200kg:

[0136] MATCH(d:Device)

[0137] WHERE d.loadCapacity>=10AND d.loadCapacity<=200

[0138] RETURNd.name,d.loadCapacity,d.workingRange

[0139] Similarly, after querying and obtaining equipment that meets the requirements, you can view other parameters of the equipment, as well as the matching relationship between the equipment and other equipment. Users can use this as a basis to configure the production line equipment. In addition, you can further query and obtain the equipment configuration plan for the entire production line by organizing the query language.

[0140] Step 6: Output and display configuration results: Output the configuration results in the form of knowledge graph visualization and device parameter list.

[0141] In this embodiment, the configuration result output and display include graph database display and equipment list output; the graph database display uses query language to display the knowledge graph database of the configured discrete manufacturing production line, and displays the production line system composition and attribute parameters in the knowledge graph database. The equipment list output will include the equipment model, manufacturer, technical parameters, cost, and equipment parameters to form a list and output it in list form to facilitate procurement and configuration. Figure 8 As shown in the figure, it is the graph database display of the optimal solution for selecting the welding robot workstation; Figure 9 As shown in the figure, the optimal solution list for arc welding robot workstation selection is displayed.

[0142] The above embodiments are merely preferred embodiments for the purpose of fully illustrating the present invention, and the scope of protection of the present invention is not limited thereto. Equivalent substitutions or modifications made by those skilled in the art based on the present invention are within the scope of protection of the present invention. The scope of protection of the present invention shall be subject to the claims.

Claims

1. A method for automatic configuration of discrete manufacturing production line equipment based on knowledge graph, characterized by: The steps include: Step 1: Production line data collection and integration: By classifying discrete manufacturing production lines, determining key equipment parameters, and refining the production line hierarchy, an equipment database is constructed that includes equipment type, technical parameters, cost, compatibility, and supplier information. Step 2: Define production line resource requirements: Extract product requirement keywords, analyze the mapping relationship between equipment parameters and product requirements, and create a demand conversion system that automatically converts product requirements into equipment parameter requirements; Step 3: Knowledge graph construction and storage: Using equipment as nodes, equipment attributes as node parameters, and inter-equipment compatibility and process associations as edges, a production line equipment configuration knowledge graph with a hierarchical structure and associations is constructed and stored in a graph database and / or RDF format. Step 4: Define the evaluation criteria for production line equipment selection: Divide the equipment into parameter levels and assign scores based on key parameters and costs, and generate a comprehensive score for the equipment configuration plan through normalized calculation; Step 5: Production line equipment configuration based on the knowledge graph: According to the output of the demand conversion system, the equipment nodes in the knowledge graph are matched, and the production line is hierarchically configured according to workstations, conveying systems, and signal transmission systems. The optimal solution is selected based on the evaluation criteria. Step 6: Output and display configuration results: Output the configuration results in the form of knowledge graph visualization and device parameter list.

2. The method for automatic configuration of discrete manufacturing production line equipment based on knowledge graph according to claim 1 is characterized in that: In step 1, the method steps for collecting and integrating production line data are as follows: 11) Classification of discrete manufacturing production line types Based on the production line implementation plan, discrete manufacturing production lines are divided into basic types according to their functions; 12) Determine key equipment parameters Considering the equipment's technical parameters, cost, compatibility, and suppliers, analyze the production process based on the product characteristics of a specific production line and determine the type and function of the equipment; 13) Refine the hierarchical structure of the production line The typical structure of each type is extracted as a class, and its equipment or devices as a subclass. Multi-level attribute parameters including equipment type, technical parameters, cost, compatibility and supplier are set for each class and subclass.

3. The method for automatic configuration of discrete manufacturing production line equipment based on knowledge graph according to claim 1 is characterized in that: In step 2, the method steps for defining production line resource requirements are as follows: 21) Product requirement keyword extraction Extract dimensional accuracy, defect detection indicators, and budget constraint keywords from product requirements through natural language processing; 22) Equipment parameter requirement analysis Establish a mathematical mapping model between equipment parameters and product requirements, including calculating sensor resolution based on detection accuracy requirements and deriving equipment processing speed based on production cycle requirements; 23) Create a product requirements conversion system The matching rule library between product requirement keywords and equipment parameters is optimized through machine learning algorithms to create a product requirement conversion system.

4. The method for automatic configuration of discrete manufacturing production line equipment based on knowledge graph according to claim 1 is characterized in that: In step 3, the method steps for constructing and storing the knowledge graph are as follows: 31) Build a production line knowledge graph ontology Based on the equipment database constructed in step 1, the production line knowledge graph ontology is constructed with the equipment as the ontology and the equipment type, technical parameters, cost, compatibility and supplier information as the attributes; 32) Define the production line knowledge graph data structure The ontology is used as a node and a triple storage model is adopted, including two structures: "node-attribute-attribute value" and "node-relationship-node". The relationships between entities are set as same-level relationships and cross-level relationships. The same-level relationship represents the flow relationship between equipment in the process production process, and the cross-level relationship represents the inclusion relationship between different levels. The relationship edge contains the attribute values of process sequence, compatibility matching degree and cost association weight. 33) Knowledge graph storage and update The constructed production line knowledge graph is stored in a graph database and / or RDF format, and the production line knowledge graph is synchronously updated after the production data is updated; the graph database is used to provide graph traversal and path query functions to quickly retrieve and analyze data in the knowledge graph.

5. The method for automatic configuration of discrete manufacturing production line equipment based on knowledge graph according to claim 1 is characterized in that: In step 4, the method steps for defining the production line equipment selection evaluation criteria are as follows: 41) Determine equipment evaluation objectives Evaluate the equipment parameters and costs obtained from the system conversion based on product requirements; 42) Implement production line equipment selection evaluation The key parameters and cost information of the equipment are obtained through the knowledge graph query language. The key parameters and costs are classified into intervals and graded. The normalized calculation is used to obtain the comprehensive score of the production line equipment configuration plan. 43) Select production line equipment The production line equipment configuration plan with the highest score is selected for production line configuration.

6. The method for automatic configuration of discrete manufacturing production line equipment based on knowledge graph according to claim 1 is characterized in that: In step 5, the method steps for configuring production line equipment based on the knowledge graph are as follows: 51) Determine configuration goals The production line is hierarchically configured according to the workstation, conveying system, and signal transmission system. The product requirement conversion system constructed in step 2 is used to obtain the production line type. The production line knowledge graph constructed in step 3 is used to query the basic composition of the production line and the composition of the workstation subsystem, and the equipment or components of the workstation subsystem are evaluated. 52) Layered Configuration Using the production line equipment selection evaluation criteria described in step 4, select the workstation subsystem configuration solution with the highest score for workstation subsystem configuration, select the conveyor system configuration solution with the highest score for conveyor system configuration, and select the signal transmission system configuration solution with the highest score for signal transmission system configuration; 53) Production line configuration Based on the workstation subsystem configuration, conveying system configuration and signal transmission system configuration obtained in step 52), the discrete manufacturing production line configuration is preliminarily completed.

7. The method for automatic configuration of discrete manufacturing production line equipment based on knowledge graph according to claim 1 is characterized by: In step 6, the configuration result output and display include graph database display and device list output; the graph database display displays the knowledge graph database of the configured discrete manufacturing production line through a query language, and displays the production line system composition and attribute parameters in the knowledge graph database; The equipment list output will include the equipment model, manufacturer, technical parameters, cost of the equipment parameters to form a list and output it in list form to facilitate procurement and configuration.

Images