Robotic hole-making process knowledge planning method and system based on digital twin

Abstract

The present invention relates to the technical field of digital twin transformation of process equipment, and solves the technical problems of lack of management of robot hole-making process knowledge system, unclear data interaction mode of digital twin, and insufficient robot hole-making process planning capability in hole-making processing of large structural parts. In particular, it relates to a robot hole-making process knowledge planning method and system based on digital twin. The modeling method includes: constructing a knowledge graph database of processing technology, constructing a digital twin system of robot hole-making, updating the indicator element layer in the knowledge graph database, and constructing a robot hole-making process planning link based on digital twin. The present invention obtains processing technology data and knowledge through the combination of twins and graphs, and combines the twin model to perform offline and online verification of process planning, thereby optimizing the traditional robot hole-making process flow, improving the accuracy of the processing technology, and effectively avoiding failure problems in the processing process.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twin transformation of process equipment, and in particular to a robot hole-making process knowledge planning method and system based on digital twin. Background Art

[0002] Industrial robot hole-making equipment can perform large-scale hole-making operations on large aviation parts, fully demonstrating the advantages of efficient processing. The current robot hole-making process is generally carried out in the form of digital simulation design and experiments. The general motion path planning uses simulation software to adjust the posture offline, and then transmits the simulation information to the actual equipment for motion teaching. The uncertain process parameters in hole-making are generally judged through actual manual experience or a large number of repetitive experiments, lacking theoretical basis. At the same time, the offline method is inefficient in interaction and lacks real-time control of the actual operating status.

[0003] As the development of robotic hole-making technology and equipment matures, and considering the process integration of digital twins and knowledge graphs, large structural parts hole-making processing requires a set of robotic hole-making systems with virtual-real synchronous mapping and high-efficiency data management to improve the robot hole-making process planning capabilities. However, this field lacks a way to integrate digital twins and knowledge graphs on robotic hole-making equipment, lacks management of the robotic hole-making process knowledge system, and the data interaction method of digital twins is unclear. The robotic hole-making process planning capabilities are insufficient, which leads to the problem of difficulty in promoting the digital transformation of the process and twin management planning during robotic hole-making. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a robot hole-making process knowledge planning method and system based on digital twins, which solves the technical problems of lack of management of the robot hole-making process knowledge system in the hole-making processing of large structural parts, unclear data interaction mode of digital twins, and insufficient robot hole-making process planning capabilities.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a robot hole making process knowledge planning method based on digital twin, the method comprising the following steps:

[0006] Constructing a knowledge graph database of machining technology: According to the induction of machining technology knowledge information, the process knowledge information elements are divided into object element layer, process element layer and indicator element layer, and a triple knowledge graph related to the robot hole making process is constructed based on the object element layer, process element layer and indicator element layer, and stored in the knowledge graph database;

[0007] Build a digital twin system for robot hole making: Build a digital twin virtual model of the robot hole making platform, update the calibration information in the physical space to the twin space of the digital twin virtual model, and then establish real-time data interaction between the physical space and the twin virtual model to build a digital twin system for robot hole making;

[0008] Update the indicator element layer in the knowledge graph database: associate the comprehensive quality evaluation valuation with the feature hole body in the part feature element layer to form effective indicator element layer information, update the indicator element layer according to the indicator element layer information and store it in the knowledge graph database;

[0009] Construct a robot hole-making process planning link based on digital twins: Generate a hole-making process machine code that can be recognized by the robot hole-making equipment control system in the physical space based on the processing process information generated by the digital twin virtual model and the processing parameters recommended by the knowledge graph database after retrieval decision, and perform process simulation in the twin space of the robot hole-making. After the process simulation information is confirmed to be correct, the hole-making process is demonstrated synchronously through the digital twin system to generate a robot hole-making process planning link.

[0010] Furthermore, the calibration information is: calibration of the position and movement of each component of the robotic hole-making equipment in the physical space by tracking and measuring equipment such as a laser tracker.

[0011] Furthermore, in the process of constructing the knowledge graph database of the processing technology, the following processes are specifically included:

[0012] The object element layer is used as the initial target node, the process element layer and the indicator element layer are used as branches to associate with the object element layer, and the indicator element layer is used as the decision information source of the process element layer;

[0013] The elements of each layer are extracted to form an ontology, and the attached and attached relationships between ontologies are constructed. Corresponding labels and attributes are generated between ontologies according to process knowledge. The relationship between each ontology generates corresponding types and labels, and then a triple knowledge graph related to the robot hole-making process is constructed and stored in the Neo4j knowledge graph database.

[0014] Furthermore, the object element layer extracts relevant feature information from the parts according to the forming state of the processed parts, and uses it as the starting node of the robot hole making processing knowledge graph to serve the association of the knowledge graph process information;

[0015] The process element layer extracts knowledge similarity of the processing process and the process steps according to the conventional robot hole making process, generates processing process related information for the features of the pre-processed holes, and establishes a robot hole making process similarity template;

[0016] The indicator element layer classifies the machining hole accuracy and quality indicators in the robot hole making process under the interaction of digital twins to form a comprehensive quality evaluation valuation, and summarizes the machining parameters and auxiliary tools corresponding to each type of indicator to form triples and store them in the knowledge graph database.

[0017] Furthermore, in the process of building a digital twin system for robot hole making, the following processes are specifically included:

[0018] A digital twin virtual model of the robot hole-making platform is built through QT and OpenGL, and the motion, mechanics, and physics rules of the digital twin virtual model are developed. A pre-processed hole feature recognition mechanism is set in the digital twin virtual model to obtain the position and shape information of the feature hole in the twin space.

[0019] Calibrate the position and motion of each component of the robotic hole-making equipment in the physical space, and update the calibration information to the twin space in a timely manner to achieve high-fidelity and high-precision mapping between the digital twin virtual model and the physical model;

[0020] Establish real-time data interaction between physical space and virtual models, and build a digital twin system for robot hole making.

[0021] Furthermore, the specific process of establishing real-time data interaction between the physical space and the virtual model is: based on the EtherCAT bus, the real-time data of the drive devices and external expansion sensor devices in the physical space are collected, the data in the bus is uniformly managed through the TwinCAT industrial control software system, and the interaction between the industrial control software and the digital twin virtual model is established through ADS communication.

[0022] Furthermore, in the process of updating the indicator element layer in the knowledge graph database, the following processes are specifically included:

[0023] Comprehensively process the machining parameters and real-time data during the robot hole-making process, evaluate and analyze the key quality feature data, and generate a comprehensive quality evaluation estimate;

[0024] The historical data consisting of machining parameters and comprehensive quality evaluation estimates are summarized to form a sample set, the comprehensive quality evaluation estimates of part hole machining are classified, and the range of machining parameter sets corresponding to the comprehensive quality evaluation estimates and the corresponding auxiliary tools are obtained;

[0025] The comprehensive quality evaluation valuation is associated with the feature hole entity in the part feature element layer, and then the comprehensive quality evaluation valuation, processing parameters and auxiliary tools are constructed to build the relationship between the entity and the entity to form effective indicator element layer information, which is updated and stored in the knowledge graph database.

[0026] Furthermore, in the process of building a robot hole-making process planning link based on digital twins, the following processes are specifically included:

[0027] Constrain the conditions of the hole-making process, extract the feature information of the pre-processed hole through the feature recognition mechanism in the digital twin system, set the processing task, determine the processing indicators of the pre-processed hole, simulate the motion processing process of the robot hole-making equipment in the motion space according to the position information of the processed hole through the digital twin virtual model, and retrieve the indicator element layer through the Neo4j knowledge graph database to make decisions and obtain the recommendation of processing parameters;

[0028] Update the processing information and processing parameters to the process element layer in the knowledge graph database, generate complete hole-making process information, and update the external environment of auxiliary tools and processing methods in the twin space;

[0029] Generate a hole-making process machine code that can be recognized by the robot hole-making equipment control system in the physical space through the processing process information and processing parameters;

[0030] Combine the environment in the twin space with the hole-making process machine code to perform process simulation in the twin space of robot hole-making;

[0031] After the process simulation information is confirmed to be correct, auxiliary tools are configured in the physical space, and holes are made through the hole-making process machine code, and the hole-making process is demonstrated synchronously through the digital twin system.

[0032] The technology also provides a digital twin system for implementing the above digital twin modeling method, which can realize the planning of the robot hole-making process, guide the assembly site in a visual way, and improve the accuracy and efficiency of the hole-making process. The system includes:

[0033] The data layer contains real-time data of the interaction between the knowledge graph database and the digital twin system. The data of the twin interaction is processed and summarized into the knowledge graph database for data acquisition, storage, processing and retrieval of data in the digital twin system.

[0034] The service layer is a service engine developed for process knowledge in the robot hole-making process, including the path and optimal planning of the movement during the movement, the prediction of quality-related factors in the processing process, intelligent decision-making and optimal recommendation of processing parameters, twin recognition in the twin scene, auxiliary tool configuration and update of the twin scene;

[0035] The application layer includes a human-machine interactive interface to realize the robot hole-making process planning process, set condition constraints according to the robot processing process, retrieve the model information of the processing material from the knowledge graph database, generate a twin model in the twin space, set the pre-processing hole features in the twin model, and set the processing assembly tasks and processing indicators according to the location of the processing feature hole and the processing indicator information;

[0036] Combined with the constraints of the machining process, the process is planned, demonstrated, verified, adjusted and output in the twin space; the results of the process planning are output to guide the actual machining.

[0037] Through the above technical solution, the present invention provides a robot hole making process knowledge planning method and system based on digital twin, which has at least the following beneficial effects:

[0038] 1. The present invention integrates digital twins and knowledge graphs to realize knowledge and twin modeling of the robot hole-making process, abstracts the knowledge elements of the robot hole-making process, and then constructs a digital twin with knowledge graph as the interactive medium. The twin process evaluation and planning are realized through the digitization, visualization, state perception and decision-making technology of the twin interface, forming an efficient and convenient process guidance method, which improves the process production factor mapping capability of existing equipment and the efficiency of robot machining process planning.

[0039] 2. The present invention can realize the planning of robot hole-making processing technology, guide the assembly site in a visual manner, and improve the accuracy and efficiency of the hole-making process.

[0040] 3. The present invention obtains processing technology data and knowledge through the combination of twins and maps, and combines the twin model to perform offline and online verification of process planning, thereby optimizing the traditional robot hole-making process, improving the accuracy of the processing technology, and effectively avoiding failure problems in the processing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0042] Figure 1 It is a flow chart of the digital twin modeling method of the present invention;

[0043] Figure 2 It is a relationship diagram between each ontology and the ontology in the object element layer, process element layer and indicator element layer of the present invention;

[0044] Figure 3 This is a principle structural block diagram of the digital twin system of the present invention. DETAILED DESCRIPTION

[0045] In order to make the above-mentioned purposes, 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 implementation methods, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0046] The assembly process of an aircraft involves a large number of connection hole features, most of which are riveted connections. The processing quality of the hole wall seriously affects the connection performance of the aircraft, and thus affects the overall strength of the aircraft. In addition, the large-scale processing surface and curved surface structure of the fuselage cause inconvenience in actual processing and production operations. The processing method is generally based on manual work, with low production efficiency and easy to cause product failure. Industrial robot hole-making equipment can perform large-scale hole-making operations on large aviation components, fully demonstrating the advantages of efficient processing. The current robot hole-making process is generally carried out in the form of digital simulation design and experiments. The general motion path planning uses simulation software to adjust the posture offline, and then transmits the simulation information to the actual equipment for motion teaching. The uncertain process parameters in hole-making are generally judged by actual manual experience or a large number of repetitive experiments, lacking theoretical basis. At the same time, the offline method is inefficient in interaction and lacks real-time control of the actual operating status.

[0047] At present, digital twin technology is booming and is widely used in various fields of society. It brings more convenience and speed to people, and makes the aviation industry develop in the direction of intercommunication. Digital twins make full use of historical data, perception information data, and mechanism data to achieve cross-fusion of multiple physical quantities and multiple scales, and realize real-time monitoring and control technology of equipment status. The knowledge graph is based on ontology and triples as the basic form of expression. It can be used to abstract and summarize processing and assembly knowledge, reveal the semantic relationship of processing technology, and visualize the graph database through visual expression. It realizes key technologies such as knowledge storage, knowledge extraction, knowledge reasoning, knowledge fusion, and knowledge representation. It can be used as an interactive medium for robot hole making process and twin data information management.

[0048] As the development of robotic hole-making technology and equipment matures, and considering the process integration of digital twins and knowledge graphs, large structural parts hole-making processing requires a set of robotic hole-making systems with virtual-real synchronous mapping and high-efficiency data management to improve the robot hole-making process planning capabilities. However, this field lacks a way to integrate digital twins and knowledge graphs on robotic hole-making equipment, lacks management of the robotic hole-making process knowledge system, and the data interaction method of digital twins is unclear. The robotic hole-making process planning capabilities are insufficient, which leads to the problem of difficulty in promoting the digital transformation of the process and twin management planning during robotic hole-making.

[0049] Based on the above shortcomings of the prior art, please refer to Figure 1-Figure 3 , shows a specific implementation of this embodiment. This embodiment integrates digital twins and knowledge graphs to model the robot hole-making process, and implements a planning method for the robot hole-making process that combines twins and graphs, aiming to form a complete and reliable processing and assembly process to guide the simulation and driving interaction capabilities of digital twins. By abstracting the knowledge elements of the robot hole-making process, a digital twin with knowledge graphs as the interactive driving method is constructed, the mapping capability of process production elements of existing equipment is improved, and an efficient and convenient process guidance method is realized. The twin process evaluation and planning are formed through the digitization, visualization, state perception and decision-making technology of the twin interface, and the interaction between the twin model and the knowledge service platform is realized.

[0050] Please refer to Figure 1 , this embodiment proposes a robot hole making process knowledge planning method based on digital twin, which includes the following process:

[0051] Construct a knowledge graph database of machining technology: According to the induction of machining technology knowledge information, the process knowledge information elements are divided into object element layer, process element layer and indicator element layer, and a triple knowledge graph related to the robot hole making process is constructed based on the object element layer, process element layer and indicator element layer, and stored in the knowledge graph database. The specific implementation method is the following process:

[0052] The object element layer is taken as the initial target node, the process element layer and the indicator element layer are taken as branches and associated with the object element layer. The indicator element layer is taken as the decision information source of the process element layer. The elements of each layer are extracted to form an ontology, and the attached and attached relationships between ontologies are constructed. The corresponding labels and attributes are generated between ontologies according to process knowledge, and the corresponding types and labels are generated between each ontology. Then, a triple knowledge graph related to the robot hole-making process is constructed and stored in the Neo4j knowledge graph database.

[0053] The relationship between each ontology in the object element layer, process element layer and indicator element layer of the present invention is as follows: Figure 2 The specific contents are as follows:

[0054] The object element layer in the present invention extracts relevant feature information from the parts according to the forming state of the processed parts, and uses it as the starting node of the robot hole-making processing knowledge graph to serve the association of the knowledge graph process information. The processed part is used as the starting node Part, and each processed hole feature in the part is used as the second sequence node Hole. Among them, the description of the processed part feature information includes the part number, name, material, geometric shape, etc.; the information of the pre-processed holes in the parts is used as the information to be confirmed, and the processing task is determined through the digital twin virtual model in the digital twin system, from which the name, size, position, depth, etc. of the pre-processed holes are extracted.

[0055] The process element layer in the present invention is to extract knowledge similarity of the processing process and process steps according to the conventional robot hole making process flow, generate processing process related information for the features of the pre-processed holes, and establish a robot hole making process similarity template. Among them, the process similarity template stipulates the processing flow for the characteristic holes, the process steps are each node T, each process step is followed by the process P, the process M mainly includes the processing method, the processing method of robot hole making is drilling, and the processing parameters D related to hole making and auxiliary tools W such as cutting tools in each process S can be obtained by simulating the corresponding working environment in the digital twin.

[0056] The indicator element layer in the present invention classifies indicators such as machining hole accuracy and quality in the robot hole making process under digital twin interaction to form a comprehensive quality evaluation estimate F, and summarizes the machining parameters D and auxiliary tools W corresponding to each type of indicator to form a triple and store it in the knowledge graph, which provides guidance for the selection of machining parameters and auxiliary tools when planning each robot hole making process.

[0057] The present invention constructs a triple knowledge graph related to the robot hole making process, which is conducive to similarity analysis and matching of the robot hole making processing technology, and improves the efficiency of twin process planning for robot hole making.

[0058] Constructing a digital twin system for robot hole making: Build a digital twin virtual model of the robot hole making platform, and update the calibration information in the physical space to the twin space of the digital twin virtual model. The calibration information is to calibrate the position and movement of each component of the robot hole making equipment in the physical space through tracking and measuring equipment such as laser trackers, and then establish real-time data interaction between the physical space and the twin virtual model to build a digital twin system for robot hole making. The specific implementation method is the following process:

[0059] A digital twin virtual model of the robot hole-making platform is built through QT and OpenGL, and the motion, mechanics, and physics rules are developed for the digital twin virtual model. A pre-processed hole feature recognition mechanism is set up in the digital twin virtual model to obtain the position and shape information of the feature holes in the twin space. The position and motion of each component of the robot hole-making equipment in the physical space are calibrated, and the calibration information is updated to the twin space in a timely manner to achieve high-fidelity and high-precision mapping between the digital twin virtual model and the physical model.

[0060] Establish real-time data interaction between physical space and virtual model, and build a digital twin system for robot hole making. Collect real-time data of drive devices and external expansion sensor devices in physical space based on EtherCAT bus, manage data in the bus uniformly through TwinCAT industrial control software system, and establish interaction between industrial control software and digital twin virtual model through ADS communication.

[0061] In this process, the detailed operation method is as follows: a digital twin virtual model of the robot hole-making platform is built through QT and OpenGL, and according to the data objects monitored and controlled by the sensors in the physical space of the actual robot hole-making platform, the motion, mechanics, physics and other rules are developed for the digital twin virtual model. A pre-processed hole feature recognition mechanism is set in the digital twin virtual model, and specific pre-processed holes can be identified according to the digital twin virtual model to obtain the position, shape and other information of the feature holes in the twin space; the position and motion of each component of the robot hole-making equipment in the physical space are calibrated through tracking and measuring equipment such as laser trackers, and the calibration information is updated to the twin space in a timely manner to achieve high-fidelity and high-precision mapping between the digital twin virtual model and the physical model; based on the EtherCAT bus, the real-time data of the drive equipment and external expansion sensor equipment in the physical space are collected, and the data in the bus is uniformly managed through the TwinCAT industrial control software system, and the interaction between the industrial control software and the digital twin virtual model is established through ADS communication, and the real-time data in the platform is mapped to the variables formed by the rules in the virtual model, so as to build a robot hole-making digital twin interactive system.

[0062] The present invention establishes a digital twin model of robot hole making, which can monitor and control the twin state of robot hole making, process real-time data, construct an indicator element layer in the knowledge graph, improve the dynamic data interaction capability of the hole making process, effectively optimize the decision-making processing parameters, and greatly improve the processing quality.

[0063] Update the index element layer in the knowledge graph database: associate the comprehensive quality evaluation and the feature hole body in the part feature element layer to form effective index element layer information, update the index element layer according to the index element layer information and store it in the knowledge graph database. The specific implementation method is the following process:

[0064] The processing parameters and real-time data in the robot hole-making process are comprehensively processed, and the key quality feature data are evaluated and analyzed to generate a comprehensive quality evaluation estimate. The historical data consisting of the processing parameters and the comprehensive quality evaluation estimates are summarized to form a sample set, and the comprehensive quality evaluation estimates of the part hole processing are classified to obtain the processing parameter set range and the corresponding auxiliary tools for the comprehensive quality evaluation estimates. The comprehensive quality evaluation estimates are associated with the feature hole ontology in the part feature element layer, and then the comprehensive quality evaluation estimates, processing parameters and auxiliary tools are constructed into entities and the association relationship between the entities is formed to form effective indicator element layer information, which is updated and stored in the knowledge graph database.

[0065] In this process, the detailed operation method is as follows: the processing parameters and real-time data in each robot hole making process are extracted and converted through the ETL data processing method, and the real-time data related to the hole making quality such as drilling force and drilling vibration are reduced in dimension and feature data is extracted; the key quality feature data are evaluated and analyzed through the grey correlation evaluation analysis method to generate feature data weights, and then the feature data and weights are weighted to score and obtain the quality comprehensive evaluation valuation; the historical data consisting of the processing parameters and quality comprehensive evaluation valuation obtained in the digital twin virtual model of robot hole making are summarized to form a sample set, and the quality comprehensive evaluation valuation of the part hole processing is classified based on the SVM support vector machine algorithm to obtain the range of the processing parameter set corresponding to the quality comprehensive evaluation valuation, and the processing parameters associated with the quality comprehensive evaluation valuation and the corresponding auxiliary tools are matched; the quality comprehensive evaluation valuation is associated with the feature hole ontology in the part feature element layer, and then the quality comprehensive evaluation valuation, processing parameters and auxiliary tools are constructed into an ontology and an association relationship between the ontology to form an effective indicator element layer information, which is updated and stored in the Neo4j knowledge graph database to provide processing indicator guidance for the next hole making and form a closed loop.

[0066] Construct a robot hole-making process planning link based on digital twins: Generate processing information based on pre-processed hole feature information through the digital twin virtual model, retrieve the index element layer through the knowledge graph database to determine the recommended processing parameters, generate the hole-making process machine code that can be recognized by the robot hole-making equipment control system in the physical space through the processing process information and processing parameters, and perform process simulation in the twin space of the robot hole-making. After the process simulation information is confirmed to be correct, the digital twin system is used to perform a twin synchronous demonstration of the hole-making process to generate a robot hole-making process planning link. The specific implementation method is the following process:

[0067] The conditions of the hole-making process are constrained, the feature information of the pre-processed holes is extracted through the feature recognition mechanism in the digital twin system, the processing tasks are set, and the processing indicators of the pre-processed holes are determined. The motion processing process of the robot hole-making equipment is simulated in the motion space according to the position information of the processed holes through the digital twin virtual model, and the indicator element layer is retrieved through the Neo4j knowledge graph database to make decisions and obtain recommendations for processing parameters.

[0068] The processing process information and processing parameters are updated to the process element layer in the knowledge graph database to generate complete hole-making process information, and the external environment of the auxiliary tools and processing methods in the twin space is updated; the hole-making process machine code that can be recognized by the robot hole-making equipment control system in the physical space is generated through the processing process information and processing parameters; combined with the environment and hole-making process machine code in the twin space, process simulation is performed in the twin space of robot hole-making.

[0069] After the process simulation information is confirmed to be correct, auxiliary tools are configured in the physical space, and holes are made through the hole-making process machine code, and the hole-making process is demonstrated synchronously through the digital twin system. On the one hand, the actual robot hole-making process is monitored in real time through the digital twin system, and on the other hand, the actual robot hole-making process is compared and verified with the twin hole-making process. The data threshold in the comparison and verification process is set. When the motion data and drilling force data in the hole-making process exceed the threshold with the twin simulation data, an alarm is issued.

[0070] The robot hole-making process planning of the present invention acquires processing technology data and knowledge through the combination of twin and atlas, and performs offline and online verification of process planning in combination with the twin model, thereby optimizing the traditional robot hole-making process flow, improving the accuracy of the processing technology, and effectively avoiding failure problems in the processing process.

[0071] This embodiment integrates digital twins and knowledge graphs to realize knowledge and twin modeling of the robot hole-making process, abstracts the knowledge elements of the robot hole-making process, and then constructs a digital twin with knowledge graph as the interactive medium. The twin process evaluation and planning are realized through the digitization, visualization, state perception and decision-making technology of the twin interface, forming an efficient and convenient process guidance method, which improves the process production factor mapping capability of existing equipment and the efficiency of robot machining process planning.

[0072] Please refer to Figure 3 This embodiment also proposes a robot hole making process knowledge planning digital twin system, which includes:

[0073] The data layer contains real-time data of the interaction between the knowledge graph database and the digital twin system. The data of the twin interaction is processed and summarized into the knowledge graph database, which can acquire, store, process and retrieve data in the digital twin system.

[0074] The service layer is mainly a service engine developed for the process knowledge of robot hole making, and is a functional module developed for the key methods in this invention. It includes the optimal planning of the path and movement during the movement process; the prediction of quality-related factors during the processing process, intelligent decision-making and optimal recommendation of processing parameters; twin recognition in the twin scene, auxiliary tool configuration and update of the twin scene.

[0075] At the application layer, according to the digital twin modeling method of the present invention, and in combination with the functions of the digital twin system, a human-computer interaction interface is developed to realize the process of robot hole-making process planning. Condition constraints are set according to the robot processing process, and the processing material model information is retrieved from the knowledge graph database. A twin model is generated in the twin space, and pre-processing hole features are set in the twin model. According to the location of the processing feature hole and the processing index information, the processing assembly tasks and processing indicators are set; combined with the processing process condition constraints, the process is planned, demonstrated, verified, adjusted and output in the twin space; the results of the process planning are output to guide the actual processing. The specific output results are shown in the following table:

[0076] Serial number Twin planning results Function 1 Process Machine Code Robotic hole making equipment in physical space to identify and perform integrated hole making processing 2 3D process video Record the twin demonstration process during the processing and generate animation 3 Parameter decision information Make decisions on the pre-processing indicators for hole formation and obtain better processing parameters 4 Process route Generate processing technology and process routes to control the actual processing process and configure auxiliary tools for the processing process

[0077] The present invention acquires processing technology data and knowledge through the combination of twins and maps, and uses the twin model to perform offline and online verification of process planning, thereby optimizing the traditional robot hole-making process, improving the accuracy of the processing technology, and effectively avoiding failure problems in the processing process.

[0078] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the above embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0079] The above implementation methods have been described in detail. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A robot hole-making process knowledge planning method based on digital twin, characterized in that: The method comprises the following steps: Constructing a knowledge graph database of machining technology: According to the induction of machining technology knowledge information, the process knowledge information elements are divided into object element layer, process element layer and indicator element layer, and a triple knowledge graph related to the robot hole making process is constructed based on the object element layer, process element layer and indicator element layer, and stored in the knowledge graph database; Build a digital twin system for robot hole making: Build a digital twin virtual model of the robot hole making platform, update the calibration information in the physical space to the twin space of the digital twin virtual model, and then establish real-time data interaction between the physical space and the twin virtual model to build a digital twin system for robot hole making; Update the indicator element layer in the knowledge graph database: associate the comprehensive quality evaluation valuation with the characteristic hole ontology in the object element layer to form effective indicator element layer information, update the indicator element layer according to the indicator element layer information and store it in the knowledge graph database; Construct a robot hole-making process planning link based on digital twins: Generate a hole-making process machine code that can be recognized by the robot hole-making equipment control system in the physical space based on the processing process information generated by the digital twin virtual model and the processing parameters recommended by the knowledge graph database after retrieval decision, and perform process simulation in the twin space of the robot hole-making. After the process simulation information is confirmed to be correct, the hole-making process is demonstrated synchronously through the digital twin system to generate a robot hole-making process planning link.

2. The robot hole making process knowledge planning method according to claim 1 is characterized in that: The calibration information is: the position and movement of each component of the robotic hole-making equipment in the physical space are calibrated by a laser tracker.

3. The robot hole making process knowledge planning method according to claim 1 is characterized in that: In the process of building the knowledge graph database of processing technology, The process includes: The object element layer is used as the initial target node, the process element layer and the indicator element layer are used as branches to associate with the object element layer, and the indicator element layer is used as the decision information source of the process element layer; The elements of each layer are extracted to form an ontology, and the attached and attached relationships between ontologies are constructed. Corresponding labels and attributes are generated between ontologies according to process knowledge. The relationship between each ontology generates corresponding types and labels, and then a triple knowledge graph related to the robot hole-making process is constructed and stored in the Neo4j knowledge graph database.

4. The robot hole making process knowledge planning method according to claim 1 or 3, characterized in that: The object element layer extracts relevant feature information from the parts according to the forming state of the processed parts, and uses it as the starting node of the robot hole making processing knowledge graph to serve the association of the knowledge graph process information; The process element layer extracts knowledge similarity of the processing process and the process steps according to the conventional robot hole making process, generates processing process related information for the features of the pre-processed holes, and establishes a robot hole making process similarity template; The indicator element layer classifies the robot hole-making accuracy and quality indicators under the interaction of digital twins to form a comprehensive quality evaluation valuation, and summarizes the processing parameters and auxiliary tools corresponding to each type of indicator to form triples and store them in the knowledge graph database.

5. The robot hole making process knowledge planning method according to claim 1 is characterized in that: In the process of building a digital twin system for robot hole making, The process includes: A digital twin virtual model of the robot hole-making platform is built through QT and OpenGL, and the motion, mechanics, and physics rules of the digital twin virtual model are developed. A pre-processed hole feature recognition mechanism is set in the digital twin virtual model to obtain the position and shape information of the feature hole in the twin space. Calibrate the position and motion of each component of the robotic hole-making equipment in the physical space, and update the calibration information to the twin space in a timely manner; Establish real-time data interaction between physical space and virtual models, and build a digital twin system for robot hole making.

6. The robot hole making process knowledge planning method according to claim 5 is characterized in that: The specific process of establishing real-time data interaction between the physical space and the virtual model is as follows: based on the EtherCAT bus, real-time data of drive devices and external expansion sensor devices in the physical space are collected, the data in the bus is uniformly managed through the TwinCAT industrial control software system, and the interaction between the industrial control software and the digital twin virtual model is established through ADS communication.

7. The robot hole making process knowledge planning method according to claim 1 is characterized in that: In the process of updating the indicator element layer in the knowledge graph database, The process includes: Comprehensively process the machining parameters and real-time data during the robot hole-making process, evaluate and analyze the key quality feature data, and generate a comprehensive quality evaluation estimate; The historical data consisting of machining parameters and comprehensive quality evaluation estimates are summarized to form a sample set, the comprehensive quality evaluation estimates of part hole machining are classified, and the range of machining parameter sets corresponding to the comprehensive quality evaluation estimates and the corresponding auxiliary tools are obtained; The comprehensive quality evaluation valuation is associated with the characteristic hole ontology in the object element layer, and then the comprehensive quality evaluation valuation, processing parameters and auxiliary tools are constructed into the ontology and the association relationship between the ontologies to form effective indicator element layer information, which is updated and stored in the knowledge graph database.

8. The robot hole making process knowledge planning method according to claim 1, characterized in that: In the process of building a robot hole-making process planning link based on digital twins, The process includes: Constrain the conditions of the hole-making process, extract the feature information of the pre-processed hole through the feature recognition mechanism in the digital twin system, set the processing task, determine the processing indicators of the pre-processed hole, simulate the motion processing process of the robot hole-making equipment in the motion space according to the position information of the processed hole through the digital twin virtual model, and retrieve the indicator element layer through the Neo4j knowledge graph database to make decisions and obtain the recommendation of processing parameters; Update the processing information and processing parameters to the process element layer in the knowledge graph database, generate complete hole-making process information, and update the external environment of auxiliary tools and processing methods in the twin space; Generate a hole-making process machine code that can be recognized by the robot hole-making equipment control system in the physical space through the processing process information and processing parameters; Combine the environment in the twin space with the hole-making process machine code to perform process simulation in the twin space of robot hole-making; After the process simulation information is confirmed to be correct, auxiliary tools are configured in the physical space, and holes are made through the hole-making process machine code, and the hole-making process is demonstrated synchronously through the digital twin system.

9. A digital twin system for implementing the robot hole making process knowledge planning method according to any one of claims 1 to 8, characterized in that: The system includes: The data layer contains real-time data of the interaction between the knowledge graph database and the digital twin system. The data of the twin interaction is processed and summarized into the knowledge graph database for data acquisition, storage, processing and retrieval of data in the digital twin system. The service layer is a service engine developed for process knowledge in the robot hole-making process, including the path and optimal planning of the movement during the movement, the prediction of quality-related factors in the processing process, intelligent decision-making and optimal recommendation of processing parameters, twin recognition in the twin scene, auxiliary tool configuration and update of the twin scene; Application layer, the human-machine interaction interface for robot hole-making process planning, including setting condition constraints according to the robot processing process, retrieving model information of processing materials from the knowledge graph database, generating twin models in the twin space, setting pre-processing hole features in the twin models, and setting processing and assembly tasks according to the location of the processing feature holes and processing index information; Combined with the constraints of the machining process, the process is planned, demonstrated, verified, adjusted and output in the twin space; the results of the process planning are output to guide the actual machining.

Images