CN113741216A - Production equipment virtual combined simulation system and method based on artificial intelligence optimization algorithm - Google Patents
Production equipment virtual combined simulation system and method based on artificial intelligence optimization algorithm Download PDFInfo
- Publication number
- CN113741216A CN113741216A CN202111107683.7A CN202111107683A CN113741216A CN 113741216 A CN113741216 A CN 113741216A CN 202111107683 A CN202111107683 A CN 202111107683A CN 113741216 A CN113741216 A CN 113741216A
- Authority
- CN
- China
- Prior art keywords
- virtual
- digital twin
- simulation
- unit
- production
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B17/00—Systems involving the use of models or simulators of said systems
- G05B17/02—Systems involving the use of models or simulators of said systems electric
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to an artificial intelligence optimization algorithm-based production equipment offline virtual combined simulation system and method. The system comprises a scheme management system, an enterprise resource planning system, a field operation system and a digital twin system; the scheme management system is responsible for making a scheme and controlling and analyzing; the enterprise resource planning system provides actual manufacturing feedback and virtual simulation feedback; the field operating system provides a real production environment model, and optimizes the effect of actual production; the digital twin system provides the analysis results. The method comprises a geometric model lightweight processing method and a customized whole line and unit design method. The method can effectively solve the problems of complex optimization and insufficient solving capability, has excellent simulation effect, and is suitable for popularization and application.
Description
Technical Field
The invention provides a production equipment virtual combination simulation system and method based on an artificial intelligence optimization algorithm. The problems of complex optimization and insufficient solving capability are effectively solved.
Background
The ship general assembly and matched product manufacturing process relates to a large amount of structured and unstructured real-time data, tens of thousands of parts need to be processed, assembled and installed, and the process also relates to a large amount of graphic and text data such as patterns, schedules, part purchase orders, material lists, dispatching orders, change orders and the like of the part processing and assembling. Through the information modeling, the data are all converted into corresponding data in a digital twin model, and then a large data model of the manufacturing process is constructed by using an artificial intelligent algorithm according to the data characteristics of multiple elements, isomerism, complex relation and the like of a ship manufacturing enterprise, so that the simulation optimization of the workshop manufacturing process, the preventive maintenance of equipment and the intelligent aid decision are realized.
In process system engineering, reinforcement learning techniques in artificial intelligence have been used to solve some challenging optimization control problems, especially nonlinear stochastic optimization control problems. The digital twin system researched by the project is applied to the management and scheduling problem of factory workshop equipment, is influenced by massive environment variable factors, is a complex nonlinear optimization problem, and is difficult to process and analyze by common technical means. Among the control and scheduling management problems in the engineering world, reinforcement learning, which is an optimization technology simulating the human learning process, has shown the capability of being different from the traditional control optimization technology when dealing with some real-world complex problems, and is very potentially applicable to optimizing factory workshops, especially production process systems involving mechanical equipment and robot automation equipment.
The method is limited by the problem that the complex Optimization problem expression and the solution capability are insufficient in the operation process of the whole line or unit by the traditional Simulation party, and the Optimization design idea of Integrated Simulation (OIS) is utilized, so that the production line customization method research based on the virtual combination of production equipment can be carried out, and the design steps of the whole line or unit are customized, therefore, the problem of how to solve the complex Optimization problem and the solution capability are insufficient needs to be solved urgently.
Disclosure of Invention
The invention aims to provide a production equipment off-line virtual combined simulation system and method based on an artificial intelligence optimization algorithm aiming at the defects in the prior art, and the system and method can effectively solve the problems of complex optimization and insufficient solving capability.
In order to achieve the purpose, the invention adopts the following technical scheme:
a production equipment virtual combination simulation system based on an artificial intelligence optimization algorithm comprises coordination and cooperation among four systems, namely a scheme management system, a digital twin system, an enterprise resource planning system and a field operation system;
the scheme management system is responsible for making a scheme and controlling and analyzing, and plays a role in predicting and regulating the virtual combined simulation technology of the production equipment; interaction exists between the scheme management system and the digital twin system, and the scheme management system and the digital twin system are responsible for reasonably and effectively analyzing a planning scheme manufactured by the digital twin system and finally providing an analysis result for the digital twin system for adjustment;
the enterprise resource planning system is responsible for providing actual manufacturing feedback and virtual simulation feedback to the engineering leader, so that the leader judges the difference and feasibility between the obtained virtual simulation result and actual manufacturing, and then provides production planning feedback to the digital twin system again to continuously solve the optimal solution;
the application occasion of the field operating system is in a physical workshop, the field operating system provides a real production environment model for the digital twin system, and is responsible for the control and execution of the physical workshop, and the field operating system applies the optimal mode obtained from the digital twin system to the actual production life, so that the effect of optimizing the actual production is achieved; the digital twin system is used as a core system and is responsible for communication with other three systems, information interchange and scheme adjustment; the digital twin system can generate data provided by a field operating system to construct a virtual environment, collect various state and motion signals of equipment in the processing process, and the signals are transmitted to the digital twin system after noise reduction and clear preprocessing operations.
Preferably, the digital twin system is responsible for generating a virtual environment, then performing virtual simulation in the environment and analyzing and optimizing the result, then providing the production planning and execution scheme to the scheme management system, and obtaining feedback from the scheme management system to perform iterative optimization, thereby achieving optimization under the existing production equipment and technology.
Preferably, the enterprise resource planning system is a management platform which is established on the basis of information technology, integrates the information technology and the advanced management thought, and provides decision means for enterprise employees and decision layers by using a systematized management thought. The system is a new generation of integrated management information system developed from a material demand plan (MRP), expands the functions of the MRP and has the core idea of supply chain management. The method breaks out of the traditional enterprise boundary, optimizes the resources of the enterprise from the supply chain range, optimizes the operation mode of the modern enterprise, and reflects the requirement of the market on the reasonable allocation of the resources of the enterprise. The method has obvious effects on improving the business process of enterprises and improving the core competitiveness of the enterprises. In the simulation technology, the method is responsible for providing actual manufacturing feedback and virtual simulation feedback to an engineering leader, so that the leader judges the difference and feasibility between the obtained virtual simulation result and actual manufacturing, and then provides production planning feedback to the digital twin system again to continuously solve the optimal solution.
Preferably, the field operating system provides a real production environment model for the digital twin system, and is responsible for the control and execution of the physical workshop, and the field operating system applies the optimal mode obtained from the digital twin system to the actual production life, so as to achieve the effect of optimizing the actual production.
Preferably, the digital twin system comprises a field acquisition system, a signal interface system, a geometric model lightweight processing and a customized whole line and unit design system; the on-site acquisition system comprises three basic units, namely an artificial intelligence unit, a physical unit and a digital twin virtual unit;
the artificial intelligence unit is responsible for planning and coordinating a series of operation processes of workshop environment simulation, equipment interaction communication, data recording and prediction for reasonably assigning tasks; the machine learning big data model module can analyze and calculate by using the received information, update an internal algorithm and output action decision information to be carried out to the scheduler; the scheduler is responsible for converting information transmitted from the big data model of the machine learning algorithm into a virtual manufacturing strategy which can be identified by the virtual environment and transmitting the virtual manufacturing strategy to the virtual simulation environment, and the virtual production process carries out virtual strategy execution; the reinforced learning strategy optimization process of one-time virtual simulation is realized;
the physical unit refers to a series of production equipment such as a sensor, a PLC (programmable logic controller), an execution unit and the like, and provides a practical reference template for the simulation of the digital twin module; the sensors in the physical unit are arranged at each part of the workshop manufacturing and processing equipment and are used for collecting various states and motion signals of the equipment in the processing process; after the sensor signal is subjected to preprocessing operations such as noise reduction and definition, information data and signal data of a physical entity of the equipment are transmitted into the digital twin virtual unit.
The digital twin unit can provide a digital environment, the signal generation and feedback rates can be greatly improved through the fictitious digital environment, the real production process is not influenced, and more possibilities of an optimization strategy can be safely explored; the digital twin unit updates and utilizes industrial simulation software and the like to establish a corresponding three-dimensional digital model in real time according to the received physical unit information; meanwhile, the simulation feedback signal generated by the digital twin unit can be sent to the physical unit in real time to realize real-time unification of the virtual environment and the physical environment.
Preferably, the artificial intelligence unit comprises a computer, a virtual environment, a model database, platform data and an intelligent scheduler; the computer is used as a core and is responsible for coordinating the operation and logic of the rest parts; simulating the virtual environment by the computer according to the real environment and the production equipment, and performing multiple iterative optimization calculation in the virtual environment by the computer; the model database is used for storing and providing model data of the production equipment; the intelligent scheduler is responsible for providing solutions to the production process and providing timely feedback to the computer.
Preferably, the digital twin virtual unit comprises a client-oriented form, a tool, a function, a database and an integrated interface, and the client-oriented form comprises three forms, namely a network Web, a client/server C/S and a browser/server B/S, according to the field requirements; a production line planning simulation, a production process simulation, a manufacturing process digital twin, production data visualization, three-dimensional scene management and a three-dimensional rendering engine are used as digital twin unit tools, so that the tools have the functions of service component management, message management, session management, authorization management and service registration management; the digital twin virtual unit exchanges and feeds back information with other modules through an ESB (enterprise service bus), an OPC (optical proximity correction) and a network server, the used databases comprise a real-time database, a business database, an attribute database and a geometric model database, the databases provide a basis for the function of the unit, and the data are stored in the databases after the function of the digital twin unit is operated; the expansion integration interface of the digital twin virtual unit comprises a secondary development interface, a business system integration interface, a real-time data integration interface and a model data exchange interface, and the interfaces are integrated interfaces prepared for redevelopment, data exchange and information communication.
Preferably, the integrated interface refers to an enterprise information service bus, a network server and industrial communication, wherein the interface between the simulation analysis system and the digital twin system is the enterprise information service bus and the network server, and 2 interfaces exist between the actual workshop and the digital twin system: one is an enterprise information service bus between the enterprise resource planning management system and the product development system to the digital twin system, and the other is industrial communication between the manufacturing execution system and the field operating system to the digital twin system.
The three main units complete the training Process of the machine learning big data model through OPC (OLE for Process Control), which is an automatic Control protocol based on Microsoft OLE and COM/DCOM technologies, including specifications such as OPC data access (DataAccess), OPC data exchange (DataeXchange), OPC Security (Security), OPC XML-DA and the like, and interface, attribute and method standard sets used for Process Control and industrial automation systems. And continuously optimizing and iterating the simulation result in the training process to obtain a gradually optimized mathematical model, and outputting the strategy to the strategy execution intelligent agent. The data in the workshop processing process are monitored in real time through the digital twin body and the sensor of the physical unit, so that the state monitoring of equipment and the prediction of unknown factors are realized, and in the prediction process, on one hand, the real-time monitored data can be tested and corrected according to historical accumulated data, and on the other hand, the historical data can be updated and expanded through the real-time monitored data. The physical unit dynamically tracks and reflects the latest state of the equipment entity through the digital twin body, generates corresponding decision information through analog simulation, and evaluates and optimizes the execution unit by utilizing the constantly set decision information.
The invention relates to a production equipment virtual combined simulation system based on an artificial intelligence optimization algorithm, which comprises four systems: the system comprises a scheme management system, a digital twin system, an enterprise resource planning system and a field operation system. The four systems are mainly related and interact with each other. The scheme management system is mainly responsible for making a scheme and controlling and analyzing, and predicting and regulating the virtual combined simulation technology of the production equipment; the digital twin system is responsible for generating a virtual environment and providing a virtual simulation scheme and results for other systems so as to achieve an optimal solution; the enterprise resource planning system integrates information technology and advanced management ideas, provides decision means for enterprise employees and decision layers through a systematized management idea, and is mainly responsible for providing actual manufacturing feedback and virtual simulation feedback to production responsible persons to judge the feasibility and optimization of the plan. And the field operating system is responsible for providing actual environment data and actual production result feedback. Of these four systems, the most important is the digital twin system, which introduces the field acquisition technology, artificial intelligence unit and its integrated interface technology, the way of lightweight processing geometric model and the design steps of customized whole line and unit in the digital twin system.
The invention relates to a production equipment virtual combined simulation method based on an artificial intelligence optimization algorithm, which is used for implementing the virtual combined simulation of production equipment by utilizing a production equipment virtual combined simulation system based on the artificial intelligence optimization algorithm, and comprises a geometric model lightweight processing method and a design method of a customized whole line and unit, and is characterized in that the geometric model lightweight method comprises the following operation steps:
a. after a model is established through three-dimensional modeling software, exporting the model into an STL format file;
b. importing the format file into Unity 3D software for analysis, judging whether redundant point lines exist or not, and deleting redundant point lines if the redundant point lines exist; if no redundant point line exists, whether the model needs to be subjected to surface reduction treatment is judged;
c. then judging whether the model has redundant surfaces, if so, performing surface reduction operation by using a Polygon Cruncher tool, and performing shell extraction treatment on the complex model;
d. correcting the simplified UV chartlet according to the UV chartlet position before the surface patch is simplified so as to ensure the chartlet consistency before and after the surface patch is simplified;
e. and finally, exporting the model in an FBX format through a plug-in the Unity 3D software, wherein export setting parameters comprise geometric parameter setting, scale factor parameter setting, FBX file format setting, animation setting and light setting.
Preferably, the operation steps of the method for designing customized whole lines and cells are as follows:
firstly, in the initial design stage, factors such as target product characteristics, factory site space, process characteristics, expected productivity and the like are researched, and the overall layout design is carried out according to the equipment requirements and configuration of an automatic production line;
secondly, three-dimensional modeling is carried out on the special machine equipment and the intermediate equipment by using a three-dimensional modeling tool, whole-line layout is carried out in simulation software according to a planned layout drawing, whole-line static planning is completed, and action design and action planning are carried out on an equipment model led into the simulation software;
abstracting and digitizing the whole line or unit model, and establishing an optimized objective function and constraint conditions of whole line operation scheduling;
developing an artificial intelligence evolution calculation algorithm as an inner core of the intelligent execution engine on the basis of the objective function and the constraint condition, and performing whole-line off-line dynamic simulation operation by using the built intelligent execution engine;
and finally analyzing the simulation result of the production line, performing iterative optimization operation, and adjusting the equipment combination mode and the production line planning until an optimal result is obtained.
Compared with the prior art, the invention has the following obvious and prominent substantive characteristics and remarkable advantages:
1. the invention relates to an artificial intelligence optimization algorithm-based production equipment offline virtual combined simulation system, which comprises a scheme management system, an enterprise resource planning system, a field operation system and a digital twin system, wherein the method comprises a geometric model lightweight processing method and a customized whole line and unit design method; the invention can effectively solve the problems of complex optimization and insufficient solving capability;
2. the method is simple and effective, the virtual combination simulation effect of the system production equipment is excellent, and the method is suitable for popularization and application.
Drawings
FIG. 1 is an overall schematic diagram of a virtual portfolio simulation system of a production facility according to a preferred embodiment of the present invention.
FIG. 2 is a schematic diagram of the field acquisition technique of the preferred embodiment of the present invention
FIG. 3 is a schematic diagram of an artificial intelligence unit of a preferred embodiment of the invention.
FIG. 4 is a schematic diagram of a digital twin unit of a preferred embodiment of the present invention.
Fig. 5 is a schematic diagram of an information integration interface technology according to a preferred embodiment of the present invention.
Fig. 6 is a flowchart of geometric model weight reduction processing according to the preferred embodiment of the present invention.
FIG. 7 is a flow chart of the design steps for the customized whole line and cell of the preferred embodiment of the present invention.
Detailed Description
The technical solution of the present invention will be described in detail with reference to the following preferred embodiments, but the scope of the present invention is not limited to the embodiments.
The first embodiment is as follows:
referring to fig. 1, a production equipment virtual combination simulation system based on an artificial intelligence optimization algorithm includes a scheme management system 1, an enterprise resource planning system 3, a field operating system 4 and a digital twin system 2; the digital twin system 2 is connected with the scheme management system 1, the enterprise resource planning system 3 and the field operating system 4;
the scheme management system 1 is responsible for making a scheme and controlling and analyzing, plays a role in predicting and regulating a virtual combined simulation system of production equipment, has interaction between the scheme management system 1 and the digital twin system 2, is responsible for reasonably and effectively analyzing a planning scheme manufactured by the digital twin system 2, and finally provides an analysis result for the digital twin system 2 for adjustment;
the enterprise resource planning system 3 is responsible for providing actual manufacturing feedback and virtual simulation feedback to the engineering leader, so that the leader judges the difference and feasibility between the obtained virtual simulation result and the actual manufacturing, and then provides production planning feedback to the digital twin system 2 again to continuously solve the optimal solution;
the field operation system 4 is applied to a physical workshop, provides a real production environment model for the digital twin system 2, is responsible for the control and execution of the physical workshop, and applies the optimal mode obtained from the digital twin system 2 to the actual production life, thereby achieving the effect of optimizing the actual production; the digital twin system 2 is used as a core system and is responsible for communication with other three systems, information interchange and scheme adjustment; the digital twin system 2 can generate data provided by the field operating system 4 to construct a virtual environment, collect various state and motion signals of equipment in the processing process, and the signals are transmitted to the digital twin system 2 after being subjected to noise reduction and clear preprocessing operations.
The production equipment offline virtual combined simulation system based on the artificial intelligence optimization algorithm comprises a scheme management system, an enterprise resource planning system, a field operation system and a digital twin system, and can effectively solve the problems of complex optimization and insufficient solving capability.
Example two:
this embodiment is substantially the same as the first embodiment, and is characterized in that:
in the embodiment, the digital twin system 2 comprises a field acquisition system 2-1, a signal interface system, a geometric model lightweight processing and customized whole line and unit design system; the on-site acquisition system 2-1 comprises three basic component units, namely an artificial intelligence unit 2-1-1, a physical unit 2-1-2 and a digital twin virtual unit 2-1-3;
the artificial intelligence unit 2-1-1 is responsible for planning and coordinating a series of operation processes of workshop environment simulation, equipment interaction communication, data recording and prediction for reasonably assigning tasks; the machine learning big data model comprises a machine learning big data model module 2-1-1-1 and a dispatcher 2-1-1-2, wherein the machine learning big data model module 2-1-1-1 can use received information to analyze and calculate, update an internal algorithm and output action decision information to be carried out to the dispatcher 2-1-1-2; the scheduler 2-1-1-2 is responsible for converting information transmitted from a machine learning algorithm big data model into a virtual manufacturing strategy which can be identified by a virtual environment and transmitting the virtual manufacturing strategy to the virtual simulation environment, and the virtual production process carries out virtual strategy execution; the reinforced learning strategy optimization process of one-time virtual simulation is realized;
the physical unit 2-1-2 refers to a series of production devices of a sensor 2-1-2-1, a PLC 2-1-2-2 and an execution unit 2-1-2-3, and provides a practical reference template for simulation of the digital twin module; the sensors 2-1-2-1 in the physical unit 2-1-2 are arranged at various parts on the workshop manufacturing and processing equipment and are used for collecting various states and motion signals of the equipment in the processing process; after the sensor signal is subjected to noise reduction and clear preprocessing operation, information data and signal data of a physical entity of the equipment are transmitted into a digital twin virtual unit;
the digital twin virtual units 2-1-3 can provide a digital environment, the signal generation and feedback rates can be greatly improved through the fictitious digital environment, the real production process is not influenced, and more possibilities of optimizing strategies can be safely explored; the digital twin virtual unit 2-1-3 updates and utilizes industrial simulation software to establish a corresponding three-dimensional digital model in real time according to the received information of the physical unit 2-1-2; meanwhile, the simulation feedback signals generated by the digital twin virtual unit 2-1-3 unit can be sent to the physical unit 2-1-2 in real time to realize real-time unification of the virtual environment and the physical environment.
In this embodiment, the artificial intelligence unit 2-1-1 includes a computer, a virtual environment, a model database, platform data, and an intelligent scheduler; the computer is used as a core and is responsible for coordinating the operation and logic of the rest parts; simulating the virtual environment by the computer according to the real environment and the production equipment, and performing multiple iterative optimization calculation in the virtual environment by the computer; the model database is used for storing and providing model data of the production equipment; the intelligent scheduler is responsible for providing solutions to the production process and providing timely feedback to the computer.
In the embodiment, the digital twin virtual unit 2-1-3 comprises a client-oriented form, a tool, a function, a database and an integrated interface, and the client-oriented form comprises three forms, namely a network Web, a client/server C/S and a browser/server B/S, according to the field requirements; a production line planning simulation, a production process simulation, a manufacturing process digital twin, production data visualization, three-dimensional scene management and a three-dimensional rendering engine are used as digital twin unit tools, so that the tools have the functions of service component management, message management, session management, authorization management and service registration management; the digital twin virtual unit 2-1-3 exchanges and feeds back information with other modules through an ESB (enterprise service bus), an OPC (optical proximity correction) and a network server, and used databases comprise a real-time database, a business database, an attribute database and a geometric model database, wherein the databases provide a basis for the functions of the unit, and the functions of the digital twin unit store the data in the databases after operation; the expansion integration interfaces of the digital twin virtual units 2-1-3 include a secondary development interface, a business system integration interface, a real-time data integration interface and a model data exchange interface, which are all integration interfaces prepared for redevelopment, data exchange and information communication.
In this embodiment, the integrated interface refers to an enterprise information service bus, a network server and industrial communication, wherein the interface between the simulation analysis system and the digital twin system is the enterprise information service bus and the network server, and there are 2 interfaces between the actual workshop and the digital twin system: one is an enterprise information service bus between the enterprise resource planning management system and the product development system to the digital twin system, and the other is industrial communication between the manufacturing execution system and the field operating system to the digital twin system.
The production equipment offline virtual combination simulation system based on the artificial intelligence optimization algorithm comprises a scheme management system, an enterprise resource planning system, a field operation system and a digital twin system, can effectively solve the problems of complex optimization and insufficient solving capability, and is high in production equipment offline virtual combination simulation accuracy and good in universality.
Example three:
a production equipment virtual combined simulation method based on an artificial intelligence optimization algorithm utilizes the production equipment virtual combined simulation system based on the artificial intelligence optimization algorithm to implement the production equipment virtual combined simulation, and comprises a geometric model lightweight processing method and a design method of a customized whole line and unit, wherein the geometric model lightweight method comprises the following operation steps:
a. after a model is established through three-dimensional modeling software, exporting the model into an STL format file;
b. importing the format file into Unity 3D software for analysis, judging whether redundant point lines exist or not, and deleting redundant point lines if the redundant point lines exist; if no redundant point line exists, whether the model needs to be subjected to surface reduction treatment is judged;
c. then judging whether the model has redundant surfaces, if so, performing surface reduction operation by using a Polygon Cruncher tool, and performing shell extraction treatment on the complex model;
d. correcting the simplified UV chartlet according to the UV chartlet position before the surface patch is simplified so as to ensure the chartlet consistency before and after the surface patch is simplified;
e. and finally, exporting the model in an FBX format through a plug-in the Unity 3D software, wherein export setting parameters comprise geometric parameter setting, scale factor parameter setting, FBX file format setting, animation setting and light setting.
The method for designing the customized whole line and unit comprises the following operation steps:
firstly, in the initial design stage, factors such as target product characteristics, factory site space, process characteristics, expected productivity and the like are researched, and the overall layout design is carried out according to the equipment requirements and configuration of an automatic production line;
secondly, three-dimensional modeling is carried out on the special machine equipment and the intermediate equipment by using a three-dimensional modeling tool, whole-line layout is carried out in simulation software according to a planned layout drawing, whole-line static planning is completed, and action design and action planning are carried out on an equipment model led into the simulation software;
abstracting and digitizing the whole line or unit model, and establishing an optimized objective function and constraint conditions of whole line operation scheduling;
developing an artificial intelligence evolution calculation algorithm as an inner core of the intelligent execution engine on the basis of the objective function and the constraint condition, and performing whole-line off-line dynamic simulation operation by using the built intelligent execution engine;
and finally analyzing the simulation result of the production line, performing iterative optimization operation, and adjusting the equipment combination mode and the production line planning until an optimal result is obtained.
The method comprises a geometric model lightweight processing method and a customized whole line and unit design method, and can effectively solve the problems of complex optimization and insufficient solving capability.
Example four:
referring to fig. 1, a virtual combined simulation of production equipment based on an artificial intelligence optimization algorithm includes four systems, a plan management system, a digital twin system, a field operation system and an enterprise resource planning system. The relation between the two is responsible for the digital twin system, and the specific relation is as follows:
firstly, a digital twin system constructs a virtual simulation environment according to an environment template provided by a field operation system, performs primary virtual production simulation according to an existing plan, provides a production planning and execution scheme to a scheme management system after obtaining a result, performs control analysis after the scheme management system obtains the planning and scheme, optimizes the production planning and execution scheme according to the analysis result, and feeds back the optimized production planning and execution scheme to the digital twin system for virtual simulation. And after the steps are repeated for a plurality of times to obtain an approved scheme, providing an optimized production scheme for the field operation system to carry out one-time actual production, and then feeding back the production data obtained by the actual production to the digital twin system. The digital twin system provides actual manufacturing feedback and corresponding virtual simulation feedback to an enterprise resource planning system managed by a responsible person, and after the approval and signature of the responsible person, the feedback is provided for the digital twin system, so that the mass production is started.
Therefore, the digital twin system has the functions of establishing a virtual simulation environment, analyzing a virtual simulation result, providing actual and virtual feedback to the enterprise resource planning system, providing a production planning and execution scheme to the scheme management system and providing a production scheme to the field operating system.
The scheme management system performs control analysis according to the production planning and execution scheme provided by the digital twin system, and optimizes the effect of the virtual scheme.
The field operation system has the functions of providing an actual operation environment and showing an actual production and manufacturing effect.
FIG. 2 is a schematic diagram of a field acquisition technique in a digital twinning system, the technique including an artificial intelligence unit, a digital twinning virtual unit, and a physical unit. The artificial intelligence unit, also known as the agent, mainly comprises two large modules: a machine learning big data model base and an intelligent scheduler. The intelligent agent obtains sensor signals and feedback from the physical unit and the digital twin virtual unit through an OPC protocol, and two strategy optimization mechanisms are respectively formed by the digital twin unit and the physical unit: virtual environment simulation optimization and physical environment simulation optimization. Under a virtual simulation optimization mechanism, simulation sensor signals and simulation feedback signals generated by a digital twin virtual environment are transmitted to an intelligent agent dispatcher through an OPC bus, and the dispatcher converts the signals into state updating and feedback information required by a reinforcement learning algorithm and transmits the state updating and feedback information to a machine learning algorithm big data model. The machine learning big data model uses the received information to carry out analysis and operation, updates internal algorithm parameters and outputs action decision information to be carried out to the scheduler. The scheduler converts the information transmitted from the big data model of the machine learning algorithm into a virtual manufacturing strategy which can be identified by the virtual environment and transmits the virtual manufacturing strategy to the virtual simulation environment, and a virtual production process carries out virtual strategy execution, thus forming a reinforced learning strategy optimization process of virtual simulation. By continuously iterating the virtual simulation process, the reinforcement learning algorithm gradually optimizes the parameters of the machine learning big data model, and gradually obtains an optimized manufacturing strategy for actual production and use. The digital environment provided by the digital twin can greatly improve the signal generation and feedback rates, does not influence the real production process, and can safely explore more possibilities of an optimization strategy.
The digital twin unit obtains the automation instruction from the physical unit through an OPC protocol so as to simulate the production process and transmit the manufacturing scheme and the simulated sensor signal to the physical unit. The digital twin unit can update in real time and utilize industrial simulation software and the like to establish a corresponding three-dimensional digital model according to the received physical unit information, and meanwhile, a simulation feedback signal generated by the digital twin unit can be sent to the physical unit in real time to realize real-time unification of a virtual environment and a physical environment.
The physical unit can collect various states and motion signals of the equipment in the processing process through the sensors arranged at various parts on the workshop manufacturing and processing equipment, and the sensor signals transmit information data and signal data of the physical entity of the equipment into the digital twin virtual unit after preprocessing operations such as noise reduction, clearness and the like. The sensor signals and the feedback signals generated by the real production process in the physical environment are real and reliable, and in the actual training of the artificial intelligent big data model, higher weight parameters can be fed back by the training of the physical environment to mine and verify the potential of the established production strategy.
Generally speaking, the three main units realize data connection of the three modules through OPC, and data in a simulation process is used for driving a training process of a machine learning big data model in an artificial intelligence unit. And continuously optimizing and iterating the simulation result in the training process to obtain a gradually optimized mathematical model, and outputting the strategy to the strategy execution intelligent agent. The data in the workshop processing process are monitored in real time through the digital twin body and the sensor of the physical unit, so that the state monitoring of equipment and the prediction of unknown factors are realized, and in the prediction process, on one hand, the real-time monitored data can be tested and corrected according to historical accumulated data, and on the other hand, the historical data can be updated and expanded through the real-time monitored data. The physical unit dynamically tracks and reflects the latest state of the equipment entity through the digital twin body, generates corresponding decision information through analog simulation, and evaluates and optimizes the execution unit by utilizing the constantly set decision information.
FIG. 3 is a schematic diagram of an artificial intelligence unit, wherein a computer transmits information obtained through platform data and a model database to an established virtual simulation environment for multiple iterative simulations to obtain an optimal solution, and then controls an intelligent scheduler to provide a scheme for a production process, so that the production process is adjusted.
Fig. 4 is a schematic diagram of a digital twin unit, which mainly comprises five aspects of a client-oriented form, tools, functions, a database and an integrated interface. The client-oriented form of the digital twin unit comprises three forms, namely a network (Web), a client/server (C/S) and a browser/server (B/S), and can be reasonably selected according to actual conditions on site. The tool has the advantages of production line planning simulation, production process simulation, digital twin in the manufacturing process, production data visualization, three-dimensional scene management and a three-dimensional rendering engine, and the tools enable a digital twin unit to have the capacity of simulating a virtual environment and the effect of virtual simulation. The functions of the digital twin unit comprise service component management, message management, session management, authorization management and service registration management, and information exchange and feedback are carried out between the service component management, the message management, the session management, the authorization management and the service registration management and with other modules through ESB, OPC and a network server. The digital twin unit database comprises a real-time database, a business database, an attribute database and a geometric model database. These databases provide the basis for the functionality of the unit, which operates to store data in the databases. The expansion integration interface of the digital twin unit comprises a secondary development interface, a business system integration interface, a real-time data integration interface and a model data exchange interface, and the interfaces are integrated interfaces prepared for redevelopment, data exchange and information communication.
Fig. 5 is a technical schematic diagram of an integrated interface in a digital twin unit, where the integrated interface mainly refers to an interface between a simulation analysis system and the digital twin system and an interface between an actual work shop and the digital twin system. The first interface exchanges information with the network server through an enterprise service information bus, the enterprise resource planning management system and the product research and development system in the second interface exchange information with the digital twin system through the enterprise information service bus, and the manufacturing execution system and the field operation system in the actual workshop exchange information with the digital twin system through industrial communication.
Fig. 6 is a flowchart of a lightweight processing geometric model, which is exported as an STL format file after a model is created by three-dimensional modeling software, and the STL format file is imported into Unity 3D software for analysis to determine whether an excess point line exists, and if an excess point line exists, the excess point line is deleted; if no redundant dotted line exists, whether the model needs to be subjected to surface reduction processing is judged. Then, whether the model has redundant surfaces or not is judged, if yes, a Polygon reduction operation can be carried out by utilizing a Polygon Cruncher tool, and the shell extraction treatment can be carried out on the complex model.
And correcting the simplified UV chartlet according to the position of the UV chartlet before the surface patch is simplified so as to ensure the chartlet consistency before and after the surface patch is simplified. And finally, exporting the model in an FBX format through a plug-in the Unity 3D software, wherein export setting parameters comprise geometric parameter setting, scale factor parameter setting, FBX file format setting, animation setting, light setting and the like.
FIG. 7 is a flow chart of the design steps for customizing whole lines and cells. In the initial design stage, factors such as target product characteristics, factory site space, process characteristics, expected capacity and the like are researched, and the overall layout design is carried out according to the equipment requirements and configuration of the automatic production line. And then, carrying out three-dimensional modeling on the special equipment, the intermediate equipment and other equipment by using a three-dimensional modeling tool, carrying out whole-line layout in simulation software according to the planned layout drawing, completing whole-line static planning, and carrying out action design and action planning on the equipment model led into the simulation software. And then abstracting and digitizing the whole line or unit model, and establishing an optimized objective function and constraint conditions of whole line operation scheduling. And on the basis of the objective function and the constraint condition, researching and developing an artificial intelligence evolution calculation algorithm as an inner core of the intelligent execution engine, and performing whole-line off-line dynamic simulation operation by using the built intelligent execution engine. And finally, analyzing the simulation result of the production line, performing iterative optimization operation, and adjusting the equipment combination mode and the production line planning until the optimal result is obtained.
The embodiments of the present invention have been described with reference to the accompanying drawings, but the present invention is not limited to the embodiments, and various changes and modifications can be made according to the purpose of the invention, and any changes, modifications, substitutions, combinations or simplifications made according to the spirit and principle of the technical solution of the present invention shall be equivalent substitutions, as long as the purpose of the present invention is met, and the present invention shall fall within the protection scope of the present invention without departing from the technical principle and inventive concept of the present invention.
Claims (7)
1. A production equipment virtual combined simulation system based on an artificial intelligence optimization algorithm comprises a scheme management system (1), an enterprise resource planning system (3), a field operating system (4) and a digital twin system (2), and is characterized in that: the digital twin system (2) is connected with the scheme management system (1), the enterprise resource planning system (3) and the field operation system (4);
the scheme management system (1) is responsible for making a scheme and controlling analysis, plays a role in predicting and regulating the virtual combined simulation system of the production equipment, has interaction between the scheme management system (1) and the digital twin system (2), is responsible for reasonably and effectively analyzing the planning scheme made by the digital twin system (2), and finally provides an analysis result for the digital twin system (2) for adjustment;
the enterprise resource planning system (3) is responsible for providing actual manufacturing feedback and virtual simulation feedback to the engineering leader, so that the leader judges the difference and feasibility between the obtained virtual simulation result and the actual manufacturing, and then provides production planning feedback to the digital twin system (2) again to continuously solve the optimal solution;
the application occasion of the field operation system (4) is in a physical workshop, the field operation system provides a real production environment model for the digital twin system (2), and is responsible for the control and execution of the physical workshop, and the field operation system applies the optimal mode obtained from the digital twin system (2) to the actual production life, so that the effect of optimizing the actual production is achieved; the digital twin system (2) is used as a core system and is responsible for communication with other three systems, information interchange and scheme adjustment; the digital twin system (2) can generate data provided by the field operating system (4) to construct a virtual environment, collect various state and motion signals of equipment in the machining process, and the signals are transmitted to the digital twin system (2) after being subjected to noise reduction and clear preprocessing operations.
2. The artificial intelligence optimization algorithm-based production equipment virtual combination simulation system according to claim 1, wherein: the digital twin system (2) comprises a field acquisition system (2-1), a signal interface system, a geometric model lightweight processing and customized whole line and unit design system; the field acquisition system (2-1) is composed of three basic units, namely an artificial intelligence unit (2-1-1), a physical unit (2-1-2) and a digital twin virtual unit (2-1-3);
the artificial intelligence unit (2-1-1) is responsible for planning and coordinating a series of operation processes of workshop environment simulation, equipment interaction communication, data recording and prediction for reasonably assigning tasks; the machine learning big data model comprises a machine learning big data model module (2-1-1-1) and a scheduler (2-1-1-2), wherein the machine learning big data model module (2-1-1-1) can analyze and calculate by using received information, update an internal algorithm and output action decision information to be carried out to the scheduler (2-1-1-2); the scheduler (2-1-1-2) is responsible for converting information transmitted from the machine learning algorithm big data model into a virtual manufacturing strategy which can be identified by a virtual environment and transmitting the virtual manufacturing strategy to the virtual simulation environment, and the virtual production process carries out virtual strategy execution; the reinforced learning strategy optimization process of one-time virtual simulation is realized;
the physical unit (2-1-2) refers to a series of production devices including a sensor (2-1-2-1), a PLC (2-1-2-2) and an execution unit (2-1-2-3), and provides a practical reference template for simulation of the digital twin module; the sensors (2-1-2-1) in the physical unit (2-1-2) are arranged at various parts on the workshop manufacturing and processing equipment and are used for collecting various states and motion signals of the equipment in the processing process; after the sensor signal is subjected to noise reduction and clear preprocessing operation, information data and signal data of a physical entity of the equipment are transmitted into a digital twin virtual unit;
the digital twin virtual unit (2-1-3) can provide a digital environment, the signal generation and feedback rates can be greatly improved through the fictitious digital environment, the real production process is not influenced, and more possibilities of an optimization strategy can be safely explored; the digital twin virtual unit (2-1-3) updates and utilizes industrial simulation software to establish a corresponding three-dimensional digital model in real time according to the received information of the physical unit (2-1-2); meanwhile, simulation feedback signals generated by the digital twin virtual unit (2-1-3) unit can be sent to the physical unit (2-1-2) in real time to realize real-time unification of the virtual environment and the physical environment.
3. The artificial intelligence optimization algorithm-based production equipment virtual combination simulation system according to claim 2, wherein: the artificial intelligence unit (2-1-1) comprises a computer, a virtual environment, a model database, platform data and an intelligent scheduler; the computer is used as a core and is responsible for coordinating the operation and logic of the rest parts; simulating the virtual environment by the computer according to the real environment and the production equipment, and performing multiple iterative optimization calculation in the virtual environment by the computer; the model database is used for storing and providing model data of the production equipment; the intelligent scheduler is responsible for providing solutions to the production process and providing timely feedback to the computer.
4. The artificial intelligence optimization algorithm-based production equipment virtual combination simulation system according to claim 2, wherein: the digital twin virtual unit (2-1-3) comprises a client-oriented form, a tool, a function, a database and an integrated interface, and the client-oriented form comprises three forms, namely a network Web, a client/server C/S and a browser/server B/S, according to the field requirements; a production line planning simulation, a production process simulation, a manufacturing process digital twin, production data visualization, three-dimensional scene management and a three-dimensional rendering engine are used as digital twin unit tools, so that the tools have the functions of service component management, message management, session management, authorization management and service registration management; the digital twin virtual unit (2-1-3) exchanges and feeds back information with other modules through an ESB (enterprise service bus), an OPC (optical proximity correction) and a network server, and used databases comprise a real-time database, a business database, an attribute database and a geometric model database, wherein the databases provide a basis for the function of the unit, and the data are stored in the databases after the function of the digital twin unit is operated; the expansion integration interfaces of the digital twin virtual units (2-1-3) comprise a secondary development interface, a business system integration interface, a real-time data integration interface and a model data exchange interface, and the interfaces are integrated interfaces prepared for redevelopment, data exchange and information communication.
5. The artificial intelligence optimization algorithm-based production equipment virtual combination simulation system according to claim 4, wherein: the integrated interface refers to an enterprise information service bus, a network server and industrial communication, wherein the interface between the simulation analysis system and the digital twin system is the enterprise information service bus and the network server, and 2 interfaces exist between the actual workshop and the digital twin system: one is an enterprise information service bus between the enterprise resource planning management system and the product development system to the digital twin system, and the other is industrial communication between the manufacturing execution system and the field operating system to the digital twin system.
6. A production equipment virtual combined simulation method based on an artificial intelligence optimization algorithm, which is used for implementing the production equipment virtual combined simulation by using the production equipment virtual combined simulation system based on the artificial intelligence optimization algorithm of claim 1, and comprises a geometric model lightweight processing method and a design method of a customized whole line and unit, wherein the geometric model lightweight method comprises the following operation steps:
a. after a model is established through three-dimensional modeling software, exporting the model into an STL format file;
b. importing the format file into Unity 3D software for analysis, judging whether redundant point lines exist or not, and deleting redundant point lines if the redundant point lines exist; if no redundant point line exists, whether the model needs to be subjected to surface reduction treatment is judged;
c. then judging whether the model has redundant surfaces, if so, performing surface reduction operation by using a Polygon Cruncher tool, and performing shell extraction treatment on the complex model;
d. correcting the simplified UV chartlet according to the UV chartlet position before the surface patch is simplified so as to ensure the chartlet consistency before and after the surface patch is simplified;
e. and finally, exporting the model in an FBX format through a plug-in the Unity 3D software, wherein export setting parameters comprise geometric parameter setting, scale factor parameter setting, FBX file format setting, animation setting and light setting.
7. The artificial intelligence optimization algorithm-based production equipment virtual combination simulation method according to claim 6, wherein: the method for designing the customized whole line and unit comprises the following operation steps:
firstly, in the initial design stage, factors such as target product characteristics, factory site space, process characteristics, expected productivity and the like are researched, and the overall layout design is carried out according to the equipment requirements and configuration of an automatic production line;
secondly, three-dimensional modeling is carried out on the special machine equipment and the intermediate equipment by using a three-dimensional modeling tool, whole-line layout is carried out in simulation software according to a planned layout drawing, whole-line static planning is completed, and action design and action planning are carried out on an equipment model led into the simulation software;
abstracting and digitizing the whole line or unit model, and establishing an optimized objective function and constraint conditions of whole line operation scheduling;
developing an artificial intelligence evolution calculation algorithm as an inner core of the intelligent execution engine on the basis of the objective function and the constraint condition, and performing whole-line off-line dynamic simulation operation by using the built intelligent execution engine;
and finally analyzing the simulation result of the production line, performing iterative optimization operation, and adjusting the equipment combination mode and the production line planning until an optimal result is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111107683.7A CN113741216A (en) | 2021-09-22 | 2021-09-22 | Production equipment virtual combined simulation system and method based on artificial intelligence optimization algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111107683.7A CN113741216A (en) | 2021-09-22 | 2021-09-22 | Production equipment virtual combined simulation system and method based on artificial intelligence optimization algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113741216A true CN113741216A (en) | 2021-12-03 |
Family
ID=78740262
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111107683.7A Pending CN113741216A (en) | 2021-09-22 | 2021-09-22 | Production equipment virtual combined simulation system and method based on artificial intelligence optimization algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113741216A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114578712A (en) * | 2022-03-08 | 2022-06-03 | 北京航空航天大学 | Multifunctional underwater autonomous vehicle cluster simulation system |
| CN114827265A (en) * | 2022-03-17 | 2022-07-29 | 元能星泰(天津)数字科技有限公司 | Cost reduction and speed increase method for large amount of instantaneous information flow of digital twin simulation algorithm |
| CN116757452A (en) * | 2023-08-18 | 2023-09-15 | 山东日辉电缆集团有限公司 | Intelligent scheduling management system for cable production and processing |
| EP4312091A1 (en) * | 2022-07-29 | 2024-01-31 | Siemens Aktiengesellschaft | Computer-implemented method and system for generating simulation models for a digital twin of a process of a production plant for a product |
-
2021
- 2021-09-22 CN CN202111107683.7A patent/CN113741216A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114578712A (en) * | 2022-03-08 | 2022-06-03 | 北京航空航天大学 | Multifunctional underwater autonomous vehicle cluster simulation system |
| CN114578712B (en) * | 2022-03-08 | 2023-09-26 | 北京航空航天大学 | Multifunctional underwater autonomous vehicle cluster simulation system |
| CN114827265A (en) * | 2022-03-17 | 2022-07-29 | 元能星泰(天津)数字科技有限公司 | Cost reduction and speed increase method for large amount of instantaneous information flow of digital twin simulation algorithm |
| CN114827265B (en) * | 2022-03-17 | 2024-01-23 | 元能星泰(天津)数字科技有限公司 | Cost-reducing and speed-increasing method for massive instantaneous information flow of digital twin simulation algorithm |
| EP4312091A1 (en) * | 2022-07-29 | 2024-01-31 | Siemens Aktiengesellschaft | Computer-implemented method and system for generating simulation models for a digital twin of a process of a production plant for a product |
| WO2024022887A1 (en) * | 2022-07-29 | 2024-02-01 | Siemens Aktiengesellschaft | Computer-implemented method and system for generating simulation models for a digital twin of a process of a production installation for a product |
| CN116757452A (en) * | 2023-08-18 | 2023-09-15 | 山东日辉电缆集团有限公司 | Intelligent scheduling management system for cable production and processing |
| CN116757452B (en) * | 2023-08-18 | 2023-12-05 | 山东日辉电缆集团有限公司 | Intelligent scheduling management system for cable production and processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113741216A (en) | Production equipment virtual combined simulation system and method based on artificial intelligence optimization algorithm | |
| CN107832497B (en) | A kind of intelligence workshop fast custom design method and system | |
| Ma et al. | A digital twin-driven production management system for production workshop | |
| CN114296408A (en) | Artificial intelligence optimization algorithm model system and algorithm for production workshop in digital twin environment | |
| Zhang et al. | Digital twin-driven cyber-physical production system towards smart shop-floor | |
| CN107807539B (en) | A kind of glass post-processing production line distributed integeration method and its system | |
| Zhou et al. | Upgrading pathways of intelligent manufacturing in China: Transitioning across technological paradigms | |
| CN113887016A (en) | Ship digital workshop simulation method and system based on digital twinning | |
| Madubuike et al. | A review of digital twin applications in construction. | |
| Liu et al. | Probing an intelligent predictive maintenance approach with deep learning and augmented reality for machine tools in IoT-enabled manufacturing | |
| Wang | An overview of function block enabled adaptive process planning for machining | |
| Liu et al. | Digital twin-enabled machining process modeling | |
| CN109857395A (en) | The integrated application method of smart camshaft grinding process software database system based on digital control system in open type | |
| CN112132359A (en) | Digital twin satellite pulsating assembly workshop design method and system | |
| US20210065006A1 (en) | Construction sequencing optimization | |
| CN109408857A (en) | The intelligent parameter drive module design method of spiral bevel gear shape Collaborative Manufacturing | |
| Wang et al. | Complex product manufacturing and operation and maintenance integration based on digital twin | |
| CN103500375A (en) | EM-Plant-based MES (Manufacturing Execution Systems) dispatching control method | |
| CN112508489A (en) | Top-level planning design method for complex equipment manufacturing | |
| CN113822585A (en) | Intelligent smelting factory informatization management system | |
| CN116523187A (en) | Engineering progress monitoring method and system based on BIM | |
| Li et al. | A multi-granularity NC program optimization approach for energy efficient machining | |
| Lan et al. | Simulation analysis of production scheduling algorithm for intelligent manufacturing cell based on artificial intelligence technology | |
| Liu et al. | Digital twin model-driven capacity evaluation and scheduling optimization for ship welding production line | |
| CN113128041A (en) | Digital factory MES system development method based on digital twin virtual technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |