Production line simulation method based on digital twinning and related equipment

Abstract

The invention relates to a digital twinning-based production line simulation method and related equipment, and the method comprises the following steps: carrying out the historical operation data mining of a target production line, and obtaining the historical time sequence operation data; performing dynamic logic simulation construction on the target production line based on the mapping space digital topological structure and the historical time sequence operation data to obtain a production line operation logic model; analyzing whether an abnormal position exists in the production line operation logic model or not through a preset anomaly detection algorithm, and if yes, labeling the abnormal position to obtain an abnormal labeling position; wherein the abnormal labeling position is a logic conflict point position and a non-optimal configuration area; and performing flow reconstruction on the production line operation logic model based on the abnormal labeling position to obtain a target virtual three-dimensional model, thereby solving the technical problems that the mapping relation between the physical entity and the digital model of the production line is not accurate enough and the actual operation state of the production line is difficult to reflect truly.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twins, and in particular to a production line simulation method based on digital twins and related equipment. Background Art

[0002] Digital transformation has become an inevitable trend in the development of the manufacturing industry. Traditional production lines often rely on manual experience in planning, design and operation, lacking scientific analysis methods and quantitative evaluation methods, which leads to problems such as low efficiency and unstable operation of production lines. Digital twin technology, as a key enabling technology for realizing intelligent manufacturing, provides new ideas and methods for solving these problems.

[0003] However, most companies are still facing many challenges when applying digital twin technology. First, the mapping relationship between the physical entity of the production line and the digital model is not accurate enough to truly reflect the actual operating status of the production line; second, the existing production line simulation methods often ignore the value of historical data and cannot make full use of existing operating experience to make optimization decisions; third, there is a lack of effective anomaly detection and optimization mechanisms, making it difficult to timely discover and solve potential problems in the production process.

[0004] In this context, it is urgent to develop a production line simulation method that can comprehensively utilize physical layout information, IoT data, and historical operation data. This method should be able to achieve accurate digital mapping of the production line, build an accurate operation logic model in a data-driven way, and automatically identify and optimize abnormal locations in the production process, thereby providing companies with a more intelligent and efficient production line optimization solution. This can not only improve production efficiency, but also reduce operating costs and enhance the market competitiveness of enterprises. Summary of the invention

[0005] The main purpose of the present invention is to provide a production line simulation method and related equipment based on digital twins, which solves the technical problem that the mapping relationship between the physical entity of the production line and the digital model is not accurate enough and it is difficult to truly reflect the actual operating status of the production line.

[0006] To achieve the above object, the present invention provides a production line simulation method based on digital twin, comprising the following steps: Acquire a physical layout blueprint and an Internet of Things topology map of a target production line, and digitally map the target production line based on the physical layout blueprint and the Internet of Things topology map to obtain a digital topological structure of a mapping space; Performing historical operation data mining on the target production line to obtain historical time series operation data; Based on the mapping space digital topology and the historical time series operation data, a dynamic logic simulation is constructed for the target production line to obtain a production line operation logic model; By using a preset anomaly detection algorithm, the operation logic model of the production line is analyzed to determine whether there is an abnormal position. If so, the abnormal position is marked to obtain an abnormal marked position; wherein the abnormal marked position is a logical conflict point position and a non-optimal configuration area; The process of the production line operation logic model is reconstructed based on the abnormal marked positions to obtain a target virtual three-dimensional model.

[0007] Furthermore, the target production line includes a plurality of devices, each of which has a connection line relationship between them. The target production line is digitally mapped based on the physical layout blueprint and the Internet of Things topology map to obtain a mapping space digital topology structure, including: Performing graph structure analysis on the physical layout blueprint to obtain an abstract graph element set; wherein the abstract graph element set includes device nodes, connection line segments and key area identifiers, and the connection line segments represent network link relationships; Based on the Internet of Things topology diagram, semantically parse and annotate the network link relationship between each of the devices to obtain semantically annotated link information; Fusing the abstract primitive set and the semantically annotated link information to obtain a plurality of fused topological relationship graphs; Performing spatial coordinate transformation on the fused topological relationship graphs of different viewing angles to obtain normalized spatial coordinate relationship graphs; Based on the normalized spatial coordinate relationship diagram, the target production line is divided into hierarchical structures to obtain a hierarchical structure diagram; Constructing a logical relationship on the hierarchical structure diagram to obtain a hierarchical logical relationship structure diagram; The hierarchical logical relationship structure diagram is digitally mapped to obtain a mapping space digital topological structure.

[0008] Furthermore, the physical layout blueprint is subjected to a graph structure analysis to obtain an abstract graph element set, including: Performing topological abstraction extraction on the spatial layout relationship in the physical layout blueprint to obtain an initial topological relationship diagram; Performing semantic annotation analysis on the initial topological relationship graph to obtain functional semantic annotation data; Performing weighted processing on the initial topological relationship graph based on the functional semantic annotation data to obtain a weighted topological relationship graph; Based on the weighted topological relationship graph, the physical layout blueprint is subjected to sub-graph division analysis to obtain a regional sub-graph set; Mining and analyzing the element layout rules of the regional sub-graph set to obtain a layout pattern set; Based on the layout pattern set, the physical layout blueprint is subjected to abstract extraction of graphic elements to obtain an abstract graphic element set.

[0009] Furthermore, the historical operation data of the target production line is mined to obtain historical time series operation data, including: Performing historical data mining on the target production line to obtain historical mining data; Marking the historical mining data for data traceability to obtain marked mining data; Conduct credibility analysis on the tag mining data to obtain credibility analysis results; Eliminate the marked mining data with a credibility lower than a preset credibility in the credibility analysis result to obtain the eliminated marked mining data; Performing time series coding on the removed marked mining data to obtain time series coded data; The periodic spectrum analysis is performed on the time series coded data by a preset periodic spectrum analysis algorithm to obtain historical time series operation data; wherein the periodic spectrum analysis is used to identify periodic patterns and long-term trends in production line operation.

[0010] Furthermore, the target production line is dynamically simulated and constructed based on the mapping space digital topology structure and the historical time series operation data to obtain a production line operation logic model, including: Decomposing the mapping space digital topological structure into topological relationships to obtain abstract topological logical relationships; Performing operation process trajectory mining based on the historical time series operation data to obtain operation process logic data; Using a preset logic isomorphism technology, the abstract topology logic relationship is logically mapped with the operation process logic data to obtain topology-operation logic mapping data; Based on the topology-operation logic mapping data, link function analysis is performed on the digital topology structure of the mapping space to obtain a link logic function relationship; Based on the historical time sequence operation data, interactive logic constraints are constructed on the link logical function relationship to obtain logic constraint operation data; Based on the logic constraint operation data, abstract topology logic relationship, operation process logic data and link logic function relationship, a dynamic logic simulation is constructed for the target production line to obtain a production line operation logic model.

[0011] Furthermore, the analysis of whether there is an abnormal position in the production line operation logic model by a preset abnormality detection algorithm includes: Extracting operation logic relationships from the production line operation logic model to obtain an operation logic relationship diagram; By using a depth-first search algorithm, a graph theory analysis is performed on the operation logic relationship diagram to obtain a structure representation of the logic relationship diagram; Performing spectral clustering analysis on the device nodes in the logical relationship graph structure representation to obtain clustering results; wherein the clustering results include functional categories, topological similarity measures and group interaction characteristics of the device nodes; Performing association analysis on the clustering results based on the historical time series operation data to obtain clustering-time series association information; Based on the clustering-timing association information, a Bayesian network is constructed for the device nodes in the operation logic relationship diagram to obtain a Bayesian network structure; Performing variational inference on the Bayesian network structure through a preset mean field variational inference algorithm to obtain an approximate distribution of conditional probabilities; By using a preset anomaly detection algorithm, anomaly detection is performed on the production line operation logic model based on the approximate distribution of the conditional probability to analyze whether there is an abnormal position in the production line operation logic model.

[0012] Furthermore, the process of reconstructing the operation logic model of the production line based on the abnormal marked position to obtain a target virtual three-dimensional model includes: Perform causal analysis on the logical conflict point positions in the abnormal marked positions to obtain the logical conflict causal relationship chain; Based on the logic conflict causal relationship chain, a topological optimization analysis is performed on the non-optimal configuration area in the abnormal marked position to obtain an optimized area topological relationship; Based on the optimized regional topological relationship, the production process of the production line operation logic model is disassembled to obtain a disassembled process unit; wherein the disassembled process unit is a function description unit and a process number unit; Based on the disassembly process unit, the resource allocation logic of the production line operation logic model is optimized to obtain an optimized resource allocation logic; Rewriting the device interaction logic in the production line operation logic model by Petri net based on the optimized resource allocation logic to obtain the rewritten device interaction logic; The production line operation logic model is reconstructed based on the rewritten device interaction logic to obtain a target virtual three-dimensional model.

[0013] The present invention also provides a production line simulation device based on digital twin, comprising: An acquisition module is used to acquire a physical layout blueprint and an Internet of Things topology map of a target production line, and digitally map the target production line based on the physical layout blueprint and the Internet of Things topology map to obtain a digital topological structure of a mapping space; A mining module, used for mining historical operation data of the target production line to obtain historical time series operation data; A construction module, used to perform dynamic logic simulation construction on the target production line based on the mapping space digital topological structure and the historical time series operation data to obtain a production line operation logic model; A marking module is used to analyze whether there is an abnormal position in the production line operation logic model through a preset abnormality detection algorithm, and if so, mark the abnormal position to obtain an abnormal marked position; wherein the abnormal marked position is a logical conflict point position and a non-optimal configuration area; A reconstruction module is used to reconstruct the process of the production line operation logic model based on the abnormal marked position to obtain a target virtual three-dimensional model.

[0014] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0015] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0016] The production line simulation method based on digital twin provided by the present invention comprises the following steps: obtaining a physical layout blueprint and an Internet of Things topology map of a target production line, digitally mapping the target production line based on the physical layout blueprint and the Internet of Things topology map, and obtaining a mapping space digital topology structure; mining historical operation data of the target production line to obtain historical time series operation data; constructing a dynamic logic simulation of the target production line based on the mapping space digital topology structure and the historical time series operation data to obtain a production line operation logic model; analyzing whether there is an abnormal position in the production line operation logic model through a preset abnormality detection algorithm, and if so, marking the abnormal position to obtain an abnormal marked position; wherein the abnormal marked position is a logical conflict point position and a non-optimal configuration area; reconstructing the process of the production line operation logic model based on the abnormal marked position to obtain a target virtual three-dimensional model, and solving the technical problem that the mapping relationship between the physical entity and the digital model of the production line is not accurate enough and it is difficult to truly reflect the actual operation status of the production line through the above-mentioned technical means, making full use of the actual operation experience accumulated by the production line, so that the simulation process no longer relies solely on the theoretical model, but can be based on real data to perform more accurate dynamic logic construction, thereby improving the beneficial effects of the authenticity and reliability of the simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the steps of a production line simulation method based on digital twins in one embodiment of the present invention; Figure 2It is a structural block diagram of a production line simulation device based on digital twin in one embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0018] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0020] like Figure 1 As shown, Figure 1 It is a schematic diagram of the steps of a production line simulation method based on digital twin in one embodiment of the present invention; In one embodiment of the present invention, a production line simulation method based on digital twin is provided, comprising the following steps: Step S1, obtaining a physical layout blueprint and an Internet of Things topology map of a target production line, and digitally mapping the target production line based on the physical layout blueprint and the Internet of Things topology map to obtain a digital topological structure of a mapping space.

[0021] Specifically, when implementing this step, it is necessary to first obtain the physical layout blueprint of the target production line, which includes the plane layout diagram of the production line, the equipment layout diagram, and related process flow charts and other engineering documents. At the same time, it is also necessary to obtain the Internet of Things topology diagram, which reflects the connection relationship and communication mode of various Internet of Things devices (such as sensors, controllers, actuators, etc.) in the production line. For example, in an intelligent manufacturing production line, the physical layout blueprint may contain the specific location information of equipment such as robot workstations, conveyor belts, and machining centers, while the Internet of Things topology diagram describes the network connection structure of equipment such as temperature sensors, pressure sensors, and RFID readers. Subsequently, based on these two types of drawing information, digital mapping technology is used to convert the physical elements in the physical space into virtual objects in the digital space. This process requires the establishment of a one-to-one correspondence between physical entities and digital objects, and the retention of the physical position relationship and logical connection relationship between them. In the mapping process, not only the static spatial layout information should be considered, but also the dynamic connection relationship of the Internet of Things devices should be mapped into the digital space, and finally a complete digital topological structure of the mapping space should be formed. This mapping structure can truly reflect the spatial position relationship and information interaction relationship of each component in the production line, and provide basic support for subsequent dynamic logic simulation. For example, when a workpiece flows on a production line, the position changes of the workpiece, as well as the related changes in sensor data and the transmission process of control instructions can be accurately tracked through this mapping space digital topology.

[0022] Step S2, performing historical operation data mining on the target production line to obtain historical time series operation data.

[0023] Specifically, in the process of mining the historical operation data of the target production line, it is first necessary to extract all historical operation records from the database system of the production line. These records usually include multi-dimensional information such as equipment operation status, production parameters, quality data, and energy consumption data. Through data cleaning and preprocessing, outliers and redundant data are removed, and the original data is converted into a standardized format. Then, the time series data analysis method is used to perform correlation analysis and feature extraction on these historical data in the time dimension, and finally obtain the historical time series operation data reflecting the operation law of the production line. For example, in an intelligent manufacturing production line, by analyzing the production records of the past three months, time series information such as the motion trajectory data of the robot workstation, the speed change data of the conveyor belt, and the process parameter data of the machining center can be extracted. These historical time series operation data not only contain timestamp information, but also record the specific parameter values ​​and status information corresponding to each time point, forming a complete time series data set. By deeply mining these data, valuable information such as the periodic law of equipment operation, the change trend of process parameters, and the key influencing factors of quality fluctuations can be found. This information provides an important data basis for the subsequent construction of an accurate production line operation logic model. For example, by analyzing historical time-series operation data, we can find out the optimal operating parameter combination of equipment under different working conditions, as well as the optimal connection time between each process. This information is of great guiding significance for optimizing the production process.

[0024] Step S3, constructing a dynamic logic simulation of the target production line based on the mapping space digital topological structure and the historical time series operation data to obtain a production line operation logic model.

[0025] Specifically, when constructing a dynamic logic simulation, it is first necessary to use the aforementioned mapping space digital topology as the basic framework. This framework defines the spatial position relationship of each device and component in the production line. Then, the timing features and operation rules extracted from the historical timing operation data are mapped to this framework, and the dynamic operation logic of the production line is constructed in a data-driven manner. In the specific implementation process, it is necessary to establish a set of rule engines to convert the equipment operation rules, process parameter change rules and various constraints found in the historical timing operation data into executable logic rules. These rules will guide the behavior of the production line operation logic model. For example, in an intelligent manufacturing production line, a complete production line operation logic model can be constructed based on the spatial relationship between the robot workstation, conveyor belt and machining center in the mapping space digital topology structure, combined with the workpiece flow timing, equipment state switching timing and process parameter adjustment timing recorded in the historical timing operation data. This model can accurately reflect various dynamic characteristics in the production process, such as the transmission logic of workpieces between different workstations, the coordination logic between equipment, and the dynamic adjustment logic of process parameters. The production line operation logic model constructed in this way not only contains static spatial relationship information, but also incorporates dynamic time sequence change rules, which can truly simulate the actual operation status of the production line. For example, when a complex production process needs to be simulated, the model can accurately predict the processing time of the workpiece at each station, the state changes of the equipment, and the possible bottleneck locations, providing a reliable basis for subsequent optimization.

[0026] Step S4, using a preset anomaly detection algorithm, analyze whether there are abnormal positions in the production line operation logic model. If so, mark the abnormal positions to obtain abnormal marked positions; wherein the abnormal marked positions are logic conflict point positions and non-optimal configuration areas.

[0027] Specifically, in this step, we first need to design a complete set of anomaly detection algorithms, which can comprehensively analyze the production line operation logic model and identify potential abnormal locations by setting specific detection rules and thresholds. The anomaly detection algorithm mainly focuses on two types of problems: one is the location of logical conflict points, and the other is the non-optimal configuration area. In the actual execution process, the anomaly detection algorithm will automatically scan and analyze the production line operation logic model. When an abnormal situation is found, the system will automatically mark the corresponding position to form a list of abnormal marked positions. For example, in an intelligent manufacturing production line, the location of the logical conflict point may be manifested as a conflict in the timing of the action between the robot workstation and the conveyor belt, or a conflict in resource competition between multiple workstations; while the non-optimal configuration area may be reflected in the unreasonable configuration of the process parameters of the machining center, resulting in excessive energy consumption or product quality fluctuations. Through the analysis of the anomaly detection algorithm, the system can accurately locate these problem areas and make intuitive visual annotations in the digital model. Specifically, when the algorithm detects that the equipment utilization rate of a certain workstation is significantly lower than the average level of the production line, or finds that material backlogs frequently occur in a certain conveyor belt section, these positions will be marked as abnormal positions. The information of these abnormal marked positions provides a clear direction for improvement for subsequent process optimization. For example, when interference between actions is detected between two adjacent workstations, the system will mark this area as a logical conflict point, indicating the need to adjust the workstation layout or optimize the operation sequence. When it is found that the energy efficiency of a certain machining center is far below the ideal state, the system will mark the area as a non-optimal configuration area and recommend optimization of its process parameters.

[0028] Step S5, reconstructing the process of the production line operation logic model based on the abnormal marked positions to obtain a target virtual three-dimensional model.

[0029] Specifically, when reconstructing the process, the system will first systematically optimize and adjust the production line operation logic model based on the abnormal annotated positions identified above. This process requires corresponding optimization strategies for different types of abnormal positions, including solving the problem of logical conflict point locations and optimizing the parameters of non-optimal configuration areas. After completing the optimization and adjustment, the system converts the reconstructed logical relationship into a visual target virtual three-dimensional model. For example, in an intelligent manufacturing production line, if the abnormal annotated position shows that there is action interference between the robot workstation and the conveyor belt, the system will automatically adjust their action timing or spatial layout during process reconstruction, and intuitively display the optimized equipment layout and motion trajectory in the target virtual three-dimensional model; for the non-optimal configuration area of ​​the machining center, the system will reset the process parameter combination based on the best practices in the historical operation data, and display the optimized parameter configuration and expected results in the three-dimensional model. This target virtual three-dimensional model not only contains the optimized spatial layout information, but also incorporates the reconstructed operation logic, so that users can intuitively understand and verify the feasibility of the optimization plan. Specifically, when a process in the production line is marked as an abnormal position, the system will analyze the relationship between the upstream and downstream of the process, and reconstruct the process by adjusting the process sequence, optimizing the equipment layout or improving the process parameters. The reconstructed plan is then presented in a three-dimensional visual form, including detailed information such as the spatial position of the equipment, the flow path of the material, and the configuration interface of the process parameters, thereby providing decision makers with an intuitive reference for the optimization plan.

[0030] In a specific embodiment, the target production line includes a plurality of devices, each of which has a connection line relationship between them. The target production line is digitally mapped based on the physical layout blueprint and the Internet of Things topology map to obtain a mapping space digital topology structure, including: Performing graph structure analysis on the physical layout blueprint to obtain an abstract graph element set; wherein the abstract graph element set includes device nodes, connection line segments and key area identifiers, and the connection line segments represent network link relationships; Based on the Internet of Things topology diagram, semantically parse and annotate the network link relationship between each of the devices to obtain semantically annotated link information; Fusing the abstract primitive set and the semantically annotated link information to obtain a plurality of fused topological relationship graphs; Performing spatial coordinate transformation on the fused topological relationship graphs of different viewing angles to obtain normalized spatial coordinate relationship graphs; Based on the normalized spatial coordinate relationship diagram, the target production line is divided into hierarchical structures to obtain a hierarchical structure diagram; Constructing a logical relationship on the hierarchical structure diagram to obtain a hierarchical logical relationship structure diagram; The hierarchical logical relationship structure diagram is digitally mapped to obtain a mapping space digital topological structure.

[0031] Specifically, in the process of digital mapping, it is necessary to first systematically analyze and process the connection line relationship between multiple devices in the target production line. In the specific implementation, the physical layout blueprint is first analyzed in the graph structure. Through image processing and feature recognition technology, the various elements in the blueprint are abstracted into a standardized set of primitives. This abstract set of primitives includes device nodes (such as robot workstations, conveyor belts, machining centers, etc.), connection segments (indicating the physical connection relationship between devices) and key area identifiers (such as workstations, buffers, etc.), where the connection segments are particularly used to represent network link relationships, reflecting the data communication and control signal transmission paths between devices. Then, the system will conduct in-depth semantic analysis and annotation of the network link relationship between devices based on the IoT topology map. For example, in an intelligent manufacturing production line, when analyzing the connection relationship between a robot workstation and an adjacent conveyor belt, it is necessary not only to annotate their physical connection methods, but also to parse the semantic information such as the data exchange protocol, control signal type, and communication bandwidth between them to form detailed semantically annotated link information. This information accurately describes the communication and control logic between devices. Then, the system fuses the abstract primitive set and semantically annotated link information, and organically combines physical space information and network topology information through data association and feature matching technologies to generate multiple fused topology relationship diagrams. These diagrams show the overall structural relationship of the production line from different angles, including equipment layout view, network connection view, signal flow view, etc. In order to unify the spatial expression of different perspectives, the system needs to perform spatial coordinate conversion on these fused topology relationship diagrams, and obtain a normalized spatial coordinate relationship diagram by standardizing the coordinate system and scale. This process ensures that all spatial information is expressed under the same reference system, which is convenient for subsequent analysis and processing. Based on the normalized spatial coordinate relationship diagram, the system continues to divide the target production line into hierarchical structures. This division takes into account multiple dimensions such as the functional attributes, spatial position relationship, and control level of the equipment, forming a clear hierarchical structure diagram. For example, the production line can be divided into different levels such as equipment layer, control layer, and management layer, and each level contains corresponding functional units and interface definitions. After forming the hierarchical structure diagram, the system needs to logically construct the association relationship between each level, and establish a complete hierarchical logical relationship structure diagram by analyzing the control dependency, data flow, and process flow between equipment. This structure diagram not only reflects the hierarchical relationship in the vertical direction, but also includes the business association in the horizontal direction. Finally, the system converts the hierarchical logical relationship structure diagram into a digital mapping space digital topology structure. This process requires converting all spatial relationships, network connections, logical dependencies and other information into a standardized digital model. For example, in an intelligent manufacturing production line, a complete mapping space digital topology structure may include detailed information such as the three-dimensional coordinate information of the robot workstation, the communication interface definition with the adjacent conveyor belt, the data exchange protocol with the upper control system, and the collaborative logic with other workstations.This digital mapping process realizes the complete transformation from physical space to digital space, laying the foundation for subsequent dynamic simulation and optimization analysis. The whole process forms a complete mapping system from the underlying physical layout to the high-level logical relationship, ensuring the high consistency between the digital twin model and the actual production line.

[0032] In a specific embodiment, the graph structure analysis of the physical layout blueprint is performed to obtain an abstract graph element set, including: Performing topological abstraction extraction on the spatial layout relationship in the physical layout blueprint to obtain an initial topological relationship diagram; Performing semantic annotation analysis on the initial topological relationship graph to obtain functional semantic annotation data; Performing weighted processing on the initial topological relationship graph based on the functional semantic annotation data to obtain a weighted topological relationship graph; Based on the weighted topological relationship graph, the physical layout blueprint is subjected to sub-graph division analysis to obtain a regional sub-graph set; Mining and analyzing the element layout rules of the regional sub-graph set to obtain a layout pattern set; Based on the layout pattern set, the physical layout blueprint is subjected to abstract extraction of graphic elements to obtain an abstract graphic element set.

[0033] Specifically, in the process of graph structure analysis of the physical layout blueprint, it is first necessary to perform topological abstraction extraction on the spatial layout relationship in the physical layout blueprint. This process converts the spatial information such as the device position and connection relationship in the blueprint into a standardized topological structure through image processing and pattern recognition technology to form an initial topological relationship graph. For example, in an intelligent manufacturing production line, the system will identify the relative position relationship of equipment such as robot workstations, conveyor belts, and machining centers, as well as the physical connection method between them, and convert this information into the form of nodes and edges. Subsequently, the system performs semantic annotation analysis on the initial topological relationship graph, and adds semantic information to each element in the topological graph by deeply understanding the functional attributes and operating characteristics of the equipment to obtain functional semantic annotation data. For example, for a robot workstation, the system will annotate its functional characteristics such as working range, operation type, and load capacity; for a conveyor belt, it will annotate its attribute information such as transmission direction, speed range, and load capacity. Based on these functional semantic annotation data, the system further performs weighted processing on the initial topological relationship graph, and assigns different weight values ​​to the nodes and edges in the topological graph according to factors such as the importance of the equipment, connection strength, and interaction frequency, thereby obtaining a weighted topological relationship graph. This weighted processing can better reflect the relative importance and degree of association of each component in the production line. After obtaining the weighted topological relationship graph, the system performs sub-graph division analysis on the physical layout blueprint based on the graph clustering algorithm and regional division rules, and divides the equipment combinations with similar functions and adjacent spaces into different regional sub-graphs to form a regional sub-graph set. For example, a complete processing procedure-related equipment can be divided into a sub-graph, including loading equipment, processing equipment, testing equipment, and unloading equipment. Then, the system mines and analyzes the element layout rules of the regional sub-graph set, and discovers the typical layout patterns and spatial arrangement rules of equipment in different regions through data mining algorithms to obtain a layout pattern set. These patterns may include different types such as linear layout, ring layout, and star layout. Each pattern corresponds to specific production functions and efficiency requirements. Finally, based on the identified layout pattern set, the system performs abstract extraction of the physical layout blueprint, simplifies the complex physical layout into a standardized abstract primitive representation, and forms the final abstract primitive set. This abstract primitive set contains basic elements such as equipment nodes, connecting segments, and key area identifiers. Each element carries corresponding attribute information and association relationships. For example, in an intelligent manufacturing production line, a complete set of abstract primitives may include node primitives representing robot workstations (including their working range and functional attributes), connecting line primitives representing conveyor belts (including their transmission characteristics and direction information), and regional identification primitives representing key working areas (such as buffer areas, handover areas, etc.). These abstract primitives describe the spatial layout characteristics of the production line in a standardized way, laying the foundation for subsequent digital mapping.The whole process forms a complete conversion chain from specific physical layout to abstract graphic representation, ensuring the accurate extraction and effective expression of spatial layout information.

[0034] In a specific embodiment, the historical operation data mining of the target production line to obtain the historical time series operation data includes: Performing historical data mining on the target production line to obtain historical mining data; Marking the historical mining data for data traceability to obtain marked mining data; Conduct credibility analysis on the tag mining data to obtain credibility analysis results; Eliminate the marked mining data with a credibility lower than a preset credibility in the credibility analysis result to obtain the eliminated marked mining data; Performing time series coding on the removed marked mining data to obtain time series coded data; The periodic spectrum analysis is performed on the time series coded data by a preset periodic spectrum analysis algorithm to obtain historical time series operation data; wherein the periodic spectrum analysis is used to identify periodic patterns and long-term trends in production line operation.

[0035] Specifically, in the process of mining historical operation data of the target production line, it is first necessary to collect historical data from various data sources of the production line, including equipment operation records, process parameter records, quality inspection data, energy consumption data and other multi-dimensional information, and systematically extract and organize these raw data through data mining technology to obtain historical mining data. For example, in an intelligent manufacturing production line, the system will collect historical records such as motion trajectory data of robot workstations, speed change data of conveyor belts, and process parameter data of machining centers. Subsequently, the system will mark the acquired historical mining data for data traceability. This process requires adding source information, acquisition time, data type and other identifiers to each piece of data to ensure the traceability of the data and form marked mining data. For example, for the motion data of a robot workstation, the system will mark the data acquisition sensor number, acquisition timestamp, data format and other information to facilitate subsequent analysis and verification. Next, the system will conduct a credibility analysis on the marked mining data, and quantitatively evaluate the reliability of the data by setting multiple evaluation dimensions (such as data integrity, acquisition stability, signal quality, etc.) to obtain the credibility analysis results. For example, the system will check whether the sampling frequency of sensor data is stable, whether there are abnormal fluctuations in the signal, whether there are obvious missing data, and give corresponding credibility scores. Based on the credibility analysis results, the system removes the labeled mining data that is lower than the preset credibility threshold. This process can filter out data samples with poor quality and ensure the data quality of subsequent analysis, thereby obtaining the labeled mining data after removal. For example, when the equipment operation data of a certain period is incompletely collected due to network failure, this part of the data will be automatically removed by the system. Then, the system performs time series encoding on the removed labeled mining data, standardizes the encoding and organization of different types of data in chronological order, and forms structured time series encoding data. This process needs to consider factors such as the time granularity of the data, the sampling frequency, and the data correlation to ensure that the time series characteristics of the data are fully preserved. Finally, the system uses the preset periodic spectrum analysis algorithm to conduct an in-depth periodic analysis of the time series encoding data, identify various periodic patterns and long-term change trends in the operation of the production line, and obtain the final historical time series operation data. For example, in the intelligent manufacturing production line, the periodic spectrum analysis can discover regular characteristics such as equipment maintenance cycle, product switching cycle, and energy consumption fluctuation cycle. These characteristics are of great value for optimizing production plans and predicting equipment status. Specifically, the algorithm may find that a robot workstation needs to be calibrated every 4 hours, or that a machining center has a significant drop in efficiency after running continuously for 8 hours. At the same time, through long-term trend analysis, the system can also identify important information such as the gradual degradation trend of equipment performance and the drift trend of process parameters.These historical time-series operation data provide reliable data support for the subsequent construction of an accurate production line operation logic model, enabling the model to better reflect the actual operation characteristics and changing patterns of the production line.

[0036] In a specific embodiment, the target production line is dynamically simulated and constructed based on the mapping space digital topology structure and the historical time series operation data to obtain a production line operation logic model, including: Decomposing the mapping space digital topological structure into a topological relationship to obtain an abstract topological logical relationship; Performing operation process trajectory mining based on the historical time series operation data to obtain operation process logic data; Using a preset logic isomorphism technology, the abstract topology logic relationship is logically mapped with the operation process logic data to obtain topology-operation logic mapping data; Based on the topology-operation logic mapping data, link function analysis is performed on the mapping space digital topology structure to obtain link logic function relationship; Based on the historical time sequence operation data, interactive logic constraints are constructed on the link logical function relationship to obtain logic constraint operation data; Based on the logic constraint operation data, abstract topology logic relationship, operation process logic data and link logic function relationship, a dynamic logic simulation is constructed for the target production line to obtain a production line operation logic model.

[0037] Specifically, in the process of constructing dynamic logic simulation, it is first necessary to decompose the topological relationship of the digital topological structure of the mapping space, and decompose the complex topological structure into basic logical relationship units through graph theory analysis methods to obtain abstract topological logical relationships. For example, in the intelligent manufacturing production line, the system will convert the spatial connection relationship between equipment such as robot workstations, conveyor belts, and machining centers into abstract logical relationship expressions, including information such as the connection sequence, control dependency, and data flow between equipment. Subsequently, the system mines the operation process trajectory based on historical time series operation data, and extracts information such as equipment operation sequence, process flow path, and control instruction sequence from historical data through data mining and pattern recognition technology to form operation process logic data. For example, the system will analyze the operation trajectory information such as the action sequence of the robot workstation under different working conditions, the material transmission path of the conveyor belt, and the process parameter adjustment sequence of the machining center. Then, the system uses the preset logical isomorphism technology to match and map the abstract topological logical relationship with the operation process logic data, and forms the topological-operation logic mapping data by establishing a corresponding relationship between the two. This process ensures the consistency between the spatial structure and the operation process. For example, when analyzing the collaborative operation between the robot workstation and the conveyor belt, the system will establish a mapping relationship between their spatial position relationship and the actual material transfer operation. Then, based on the topology-operation logic mapping data, the system performs link function analysis on the digital topology structure of the mapping space, identifies the functional characteristics and performance indicators of different links, and obtains the link logic function relationship. These relationships describe the functional dependencies and performance constraints between devices. For example, the system analyzes the functional characteristics such as the transmission capacity and response time of the material transmission link between the conveyor belt and the machining center. Next, the system constructs interactive logic constraints on the link logic function relationship based on the historical timing operation data, and forms logic constraint operation data by analyzing the interaction rules and operation constraints between devices. These constraint data define the boundary conditions and restriction rules for the operation of the equipment. For example, the system will formulate rules such as safety distance constraints and action timing constraints between the robot workstation and adjacent equipment based on historical data. Finally, the system uses logic constraint operation data, abstract topology logic relations, operation process logic data and link logic function relations to build a complete production line operation logic model. This model can accurately reflect the dynamic operation characteristics of the production line. For example, in an intelligent manufacturing production line, when a complete production process needs to be simulated, the model can arrange the equipment location based on the spatial topology, determine the process sequence according to the operation process logic, ensure equipment coordination through link function relationships, and ensure operation safety based on logical constraints. This fusion of multi-dimensional information enables the model to truly reflect the operation rules of the production line and provide a reliable basis for subsequent optimization analysis. Specifically, the model can accurately predict key indicators such as production efficiency, equipment utilization, energy consumption, etc. under different working conditions, and can identify potential efficiency bottlenecks and optimization space.

[0038] In a specific embodiment, analyzing whether there is an abnormal position in the production line operation logic model by a preset abnormality detection algorithm includes: Extracting operation logic relationships from the production line operation logic model to obtain an operation logic relationship diagram; By using a depth-first search algorithm, a graph theory analysis is performed on the operation logic relationship diagram to obtain a structure representation of the logic relationship diagram; Performing spectral clustering analysis on the device nodes in the logical relationship graph structure representation to obtain clustering results; wherein the clustering results include functional categories, topological similarity measures and group interaction characteristics of the device nodes; Performing association analysis on the clustering results based on the historical time series operation data to obtain clustering-time series association information; Based on the clustering-timing association information, a Bayesian network is constructed for the device nodes in the operation logic relationship diagram to obtain a Bayesian network structure; Performing variational inference on the Bayesian network structure through a preset mean field variational inference algorithm to obtain an approximate distribution of conditional probabilities; By using a preset anomaly detection algorithm, anomaly detection is performed on the production line operation logic model based on the approximate distribution of the conditional probability to analyze whether there is an abnormal position in the production line operation logic model.

[0039] Specifically, in the process of anomaly detection, it is first necessary to extract the operation logic relationship of the production line operation logic model, and transform these relationships into a standardized operation logic relationship diagram by analyzing the logical dependencies and interaction relationships between the various components in the model. For example, in an intelligent manufacturing production line, the system will extract the control dependency, material transfer relationship, and information interaction relationship between equipment such as robot workstations, conveyor belts, and machining centers to form a complete logical relationship network. Subsequently, the system conducts a systematic graph analysis of the operation logic relationship diagram through a depth-first search algorithm. This process will traverse all nodes and edges in the graph, identify features such as critical paths, loop structures, and branch conditions, and obtain a standardized logical relationship diagram structure representation. For example, the algorithm can discover the key process path from raw material input to finished product output, as well as the parallel and serial relationships between different processes. Next, the system performs spectral clustering analysis on the device nodes in the logical relationship diagram structure representation, and divides devices with similar functions into different categories by calculating the similarity matrix and eigenvectors between nodes to obtain clustering results. This result includes the functional category of the equipment nodes (such as processing equipment, transmission equipment, etc.), topological similarity measurement (reflecting the degree of structural similarity between equipment), and group interaction characteristics (describing the collaborative mode between equipment groups). For example, the system may classify multiple processing centers into the same functional category and identify the topological similarity and interaction characteristics between them. Then, the system performs association analysis on the clustering results based on the historical time series operation data, and discovers the association law between the equipment category and the operation status through the time series analysis method to form clustering-time series association information. For example, the system analyzes the operation characteristics of different categories of equipment in each time period and identifies the time series dependency between them. Based on this clustering-time series association information, the system constructs a Bayesian network for the equipment nodes in the operation logic relationship diagram, and forms a complete Bayesian network structure by establishing conditional probability relationships between nodes. This network can describe the probabilistic dependency between equipment states, such as how the operation status of a processing center affects the working efficiency of the downstream conveyor belt. Next, the system performs variational inference on the Bayesian network structure through the preset mean field variational inference algorithm. This process approximates the conditional probability distribution of each node in the network through iterative calculation to obtain the approximate distribution of conditional probabilities. These probability distributions reflect the uncertainty and changing patterns of the equipment's operating status. Finally, the system uses a preset anomaly detection algorithm to perform anomaly detection on the production line operation logic model based on the approximate distribution of conditional probabilities. The algorithm calculates the operating status probability of each device node. When the state probability of a node deviates significantly from the normal range, it will be marked as a potential abnormal location. For example, in an intelligent manufacturing production line, if the algorithm finds that the probability of a certain robot workstation's action sequence is significantly lower than normal, or that the operating status of a conveyor belt is significantly inconsistent with the status of adjacent equipment, these locations will be identified as abnormal locations.Through this probabilistic inference-based anomaly detection method, the system can accurately detect logical conflict points and non-optimal configuration areas in the production line, providing a clear direction for subsequent optimization.

[0040] In a specific embodiment, the process of reconstructing the production line operation logic model based on the abnormal marked position to obtain a target virtual three-dimensional model includes: Perform causal analysis on the logical conflict point positions in the abnormal marked positions to obtain the logical conflict causal relationship chain; Based on the logic conflict causal relationship chain, a topological optimization analysis is performed on the non-optimal configuration area in the abnormal marked position to obtain an optimized area topological relationship; Based on the optimized regional topological relationship, the production process of the production line operation logic model is disassembled to obtain a disassembled process unit; wherein the disassembled process unit is a function description unit and a process number unit; Based on the disassembly process unit, the resource allocation logic of the production line operation logic model is optimized to obtain an optimized resource allocation logic; Rewriting the device interaction logic in the production line operation logic model by Petri net based on the optimized resource allocation logic to obtain the rewritten device interaction logic; The production line operation logic model is reconstructed based on the rewritten device interaction logic to obtain a target virtual three-dimensional model.

[0041] Specifically, in the process of process reconstruction, it is necessary to first perform causal analysis on the logical conflict point position in the abnormal marked position, and build a complete logical conflict causal relationship chain by analyzing the causes and impact paths of the logical conflict. For example, in an intelligent manufacturing production line, if the system finds that there is an action conflict between the robot workstation and the conveyor belt, it will analyze whether the conflict is caused by unreasonable equipment layout, inconsistent control timing, or improper parameter configuration, and track how the conflict affects the operation of downstream processes. Subsequently, the system performs topological optimization analysis on the non-optimal configuration area in the abnormal marked position based on the established logical conflict causal relationship chain, and obtains the topological relationship of the optimized area by evaluating the feasibility and effect of different optimization schemes. For example, the system will analyze the effects of different optimization schemes such as adjusting equipment layout, optimizing process parameters, or improving control strategies for the non-optimal energy consumption configuration of the machining center, and select the most suitable optimization scheme. Then, based on the topological relationship of the optimized area, the system disassembles the production process of the production line operation logic model, decomposes the complex production process into basic functional units and process units, and obtains the disassembled process unit. These disassembled process units include functional description units (describing the specific functions and performance indicators of each unit) and process number units (identifying the order and hierarchy of the units in the entire production process). For example, a complete processing procedure may be disassembled into basic functional units such as loading unit, processing unit, inspection unit and unloading unit. Then, the system optimizes the resource allocation logic of the production line operation logic model based on the disassembled process units, and replans the resource allocation scheme of each functional unit through the optimization algorithm to obtain the optimized resource allocation logic. This process takes into account resource constraints in multiple dimensions such as equipment capacity, material requirements, and manpower allocation to ensure efficient use of resources. For example, the system will optimize the task allocation of the robot workstation, adjust the running speed of the conveyor belt, and rearrange the processing sequence of the machining center. Next, the system rewrites the device interaction logic in the production line operation logic model based on the optimized resource allocation logic. By constructing a new Petri net model to describe the optimized device interaction relationship, the rewritten device interaction logic is obtained. This rewriting process ensures that the collaborative operation between devices is more efficient and reliable. For example, through Petri net rewriting, the system can optimize the interaction timing between the robot workstation and the conveyor belt to eliminate possible deadlocks and conflicts. Finally, the system reconstructs the process of the production line operation logic model based on the rewritten equipment interaction logic, and integrates the optimized components into a complete target virtual 3D model. This model not only includes the optimized spatial layout, but also incorporates the improved operation logic and control strategy. For example, in an intelligent manufacturing production line, the target virtual 3D model will display the optimized equipment layout, improved material flow path, optimized control logic, and expected performance indicators, providing decision makers with an intuitive reference for optimization solutions.Through this systematic process reconstruction, we can effectively resolve logical conflicts and efficiency issues in the production line and improve overall production efficiency.

[0042] The above describes the production line simulation method based on digital twin in the embodiment of the present invention. The following describes the production line simulation device based on digital twin in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, a production line simulation device based on digital twins includes: An acquisition module 21 is used to acquire a physical layout blueprint and an Internet of Things topology map of a target production line, and digitally map the target production line based on the physical layout blueprint and the Internet of Things topology map to obtain a digital topological structure of a mapping space; A mining module 22, used for mining historical operation data of the target production line to obtain historical time series operation data; A construction module 23 is used to perform dynamic logic simulation construction on the target production line based on the mapping space digital topological structure and the historical time series operation data to obtain a production line operation logic model; The marking module 24 is used to analyze whether there is an abnormal position in the production line operation logic model through a preset abnormality detection algorithm, and if so, mark the abnormal position to obtain an abnormal marked position; wherein the abnormal marked position is a logical conflict point position and a non-optimal configuration area; The reconstruction module 25 is used to reconstruct the process of the production line operation logic model based on the abnormal marked position to obtain a target virtual three-dimensional model.

[0043] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the above method embodiment, which will not be repeated here.

[0044] Reference Figure 3 The present invention also provides a computer device in an embodiment, wherein the internal structure of the computer device can be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0045] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0046] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0047] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0048] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0049] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A production line simulation method based on digital twin, characterized in that: The following steps are involved: Acquire a physical layout blueprint and an Internet of Things topology map of a target production line, and digitally map the target production line based on the physical layout blueprint and the Internet of Things topology map to obtain a digital topological structure of a mapping space; Performing historical operation data mining on the target production line to obtain historical time series operation data; Based on the mapping space digital topology and the historical time series operation data, a dynamic logic simulation is constructed for the target production line to obtain a production line operation logic model; By using a preset anomaly detection algorithm, the operation logic model of the production line is analyzed to determine whether there is an abnormal position. If so, the abnormal position is marked to obtain an abnormal marked position; wherein the abnormal marked position is a logical conflict point position and a non-optimal configuration area; The process of the production line operation logic model is reconstructed based on the abnormal marked positions to obtain a target virtual three-dimensional model.

2. The production line simulation method based on digital twin according to claim 1, characterized in that: The target production line includes a plurality of devices, each of which has a connection line relationship between them. The target production line is digitally mapped based on the physical layout blueprint and the Internet of Things topology map to obtain a mapping space digital topology structure, including: Performing graph structure analysis on the physical layout blueprint to obtain an abstract graph element set; wherein the abstract graph element set includes device nodes, connection line segments and key area identifiers, and the connection line segments represent network link relationships; Based on the Internet of Things topology diagram, semantically parse and annotate the network link relationship between each of the devices to obtain semantically annotated link information; Fusing the abstract primitive set and the semantically annotated link information to obtain a plurality of fused topological relationship graphs; Performing spatial coordinate transformation on the fused topological relationship graphs of different viewing angles to obtain normalized spatial coordinate relationship graphs; Based on the normalized spatial coordinate relationship diagram, the target production line is divided into hierarchical structures to obtain a hierarchical structure diagram; Constructing a logical relationship on the hierarchical structure diagram to obtain a hierarchical logical relationship structure diagram; The hierarchical logical relationship structure diagram is digitally mapped to obtain a mapping space digital topological structure.

3. The production line simulation method based on digital twin according to claim 2 is characterized in that: The graph structure analysis is performed on the physical layout blueprint to obtain an abstract graph element set, including: Performing topological abstraction extraction on the spatial layout relationship in the physical layout blueprint to obtain an initial topological relationship diagram; Performing semantic annotation analysis on the initial topological relationship graph to obtain functional semantic annotation data; Performing weighted processing on the initial topological relationship graph based on the functional semantic annotation data to obtain a weighted topological relationship graph; Based on the weighted topological relationship graph, the physical layout blueprint is subjected to sub-graph division analysis to obtain a regional sub-graph set; Mining and analyzing the element layout rules of the regional sub-graph set to obtain a layout pattern set; Based on the layout pattern set, the physical layout blueprint is subjected to abstract extraction of graphic elements to obtain an abstract graphic element set.

4. The production line simulation method based on digital twin according to claim 1, characterized in that: The historical operation data mining of the target production line is performed to obtain historical time series operation data, including: Performing historical data mining on the target production line to obtain historical mining data; Marking the historical mining data for data traceability to obtain marked mining data; Conduct credibility analysis on the tag mining data to obtain credibility analysis results; Eliminate the marked mining data with a credibility lower than a preset credibility in the credibility analysis result to obtain the eliminated marked mining data; Performing time series coding on the removed marked mining data to obtain time series coded data; The periodic spectrum analysis is performed on the time series coded data by a preset periodic spectrum analysis algorithm to obtain historical time series operation data; wherein the periodic spectrum analysis is used to identify periodic patterns and long-term trends in production line operation.

5. The production line simulation method based on digital twin according to claim 1, characterized in that: The dynamic logic simulation construction of the target production line based on the mapping space digital topological structure and the historical time series operation data to obtain the production line operation logic model includes: Decomposing the mapping space digital topological structure into topological relationships to obtain abstract topological logical relationships; Performing operation process trajectory mining based on the historical time series operation data to obtain operation process logic data; Using a preset logic isomorphism technology, the abstract topology logic relationship is logically mapped with the operation process logic data to obtain topology-operation logic mapping data; Based on the topology-operation logic mapping data, link function analysis is performed on the mapping space digital topology structure to obtain link logic function relationship; Based on the historical time sequence operation data, interactive logic constraints are constructed on the link logical function relationship to obtain logic constraint operation data; Based on the logic constraint operation data, abstract topology logic relationship, operation process logic data and link logic function relationship, a dynamic logic simulation is constructed for the target production line to obtain a production line operation logic model.

6. The production line simulation method based on digital twin according to claim 2, characterized in that: The analyzing whether there is an abnormal position in the production line operation logic model by using a preset abnormality detection algorithm includes: Extracting the operation logic relationship of the production line operation logic model to obtain an operation logic relationship diagram; By using a depth-first search algorithm, a graph theory analysis is performed on the operation logic relationship diagram to obtain a structure representation of the logic relationship diagram; Performing spectral clustering analysis on the device nodes in the logical relationship graph structure representation to obtain clustering results; wherein the clustering results include functional categories, topological similarity measures and group interaction characteristics of the device nodes; Performing association analysis on the clustering results based on the historical time series operation data to obtain clustering-time series association information; Based on the clustering-timing association information, a Bayesian network is constructed for the device nodes in the operation logic relationship diagram to obtain a Bayesian network structure; Performing variational inference on the Bayesian network structure through a preset mean field variational inference algorithm to obtain an approximate distribution of conditional probabilities; By using a preset anomaly detection algorithm, anomaly detection is performed on the production line operation logic model based on the approximate distribution of the conditional probability to analyze whether there is an abnormal position in the production line operation logic model.

7. The production line simulation method based on digital twin according to claim 1, characterized in that: The process of reconstructing the operation logic model of the production line based on the abnormal marked position to obtain a target virtual three-dimensional model includes: Perform causal analysis on the logical conflict point positions in the abnormal marked positions to obtain the logical conflict causal relationship chain; Based on the logic conflict causal relationship chain, a topological optimization analysis is performed on the non-optimal configuration area in the abnormal marked position to obtain an optimized area topological relationship; Based on the optimized regional topological relationship, the production process of the production line operation logic model is disassembled to obtain a disassembled process unit; wherein the disassembled process unit is a function description unit and a process number unit; Based on the disassembly process unit, the resource allocation logic of the production line operation logic model is optimized to obtain an optimized resource allocation logic; Rewriting the device interaction logic in the production line operation logic model by Petri net based on the optimized resource allocation logic to obtain the rewritten device interaction logic; The production line operation logic model is reconstructed based on the rewritten device interaction logic to obtain a target virtual three-dimensional model.

8. A production line simulation device based on digital twin, characterized in that: include: An acquisition module is used to acquire a physical layout blueprint and an Internet of Things topology map of a target production line, and digitally map the target production line based on the physical layout blueprint and the Internet of Things topology map to obtain a digital topological structure of a mapping space; A mining module, used for mining historical operation data of the target production line to obtain historical time series operation data; A construction module, used to perform dynamic logic simulation construction on the target production line based on the mapping space digital topological structure and the historical time series operation data to obtain a production line operation logic model; A marking module is used to analyze whether there is an abnormal position in the production line operation logic model through a preset abnormality detection algorithm, and if so, mark the abnormal position to obtain an abnormal marked position; wherein the abnormal marked position is a logical conflict point position and a non-optimal configuration area; A reconstruction module is used to reconstruct the process of the production line operation logic model based on the abnormal marked position to obtain a target virtual three-dimensional model.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images