Digital twinborn model construction method and system

Abstract

The invention relates to the field of digital twinning, and discloses a digital twinning model construction method and system, and the method comprises the steps: determining a model construction target and range, collecting and processing various data generated by a physical entity, carrying out the feature analysis of the processed data, and extracting the features for constructing a digital twinning model from the data. The method comprises the following steps: constructing a three-dimensional model of a physical entity by utilizing three-dimensional modeling software according to project requirements, mapping characteristics of the physical entity into a mathematical model, constructing a digital twinborn model, performing simulation analysis and evaluation on the constructed digital twinborn model by utilizing a simulation tool, and performing difference analysis according to a simulation evaluation result and actual requirements, thereby obtaining a digital twinborn model. The digital twinborn model is adjusted and deployed, the deviation between the model and an actual system is reduced, the difference and deviation between the virtual model and the actual system can be accurately identified, and the optimization and adjustment of the model can be accurately positioned.

Description

Technical Field

[0001] The present invention relates to the field of digital twin technology, and more specifically to a digital twin model construction method and system. Background Art

[0002] Digital twin technology has been widely used in manufacturing, smart cities, transportation, and medical fields. With the rapid development of technologies such as the Internet of Things, big data, and cloud computing, the implementation of digital twin technology has become more feasible and efficient. The construction system of the digital twin model uses technical means such as the Internet of Things (IoT), sensors, and RFID (radio frequency identification) to achieve real-time and accurate data collection of the physical world, and uses multidisciplinary technologies such as computer graphics, artificial intelligence, and system dynamics to digitally restore physical objects in time and space into virtual models;

[0003] However, the above process still has the following disadvantages:

[0004] First, the existing technology is based on traditional demand analysis methods, which may not be able to fully and deeply understand the real needs of complex systems, resulting in deviations between the model and the actual system, and lack of sufficient extraction of key features in the data, resulting in the omission of important information during the model construction process, affecting the performance of the model;

[0005] Second, the existing technology lacks the analysis of the differences between the simulation analysis results and the actual needs, and may not be able to accurately identify the differences and deviations between the virtual model and the actual system, resulting in the inability to accurately locate the optimization and improvement measures. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a digital twin model construction method and system to solve the problems existing in the above-mentioned background technology.

[0007] The present invention provides the following technical solution: a method for constructing a digital twin model, comprising:

[0008] S1: Determine the goal and scope of model construction, collect user requirements for the digital twin model, set the business goals of the model in terms of performance, thereby creating the physical entity of the digital twin, and transmit the determined model construction goals and scope to S2;

[0009] S2: accepts user request data for model construction, uses sensors, monitoring equipment and the Internet to collect various data generated by physical entities, including environmental parameters, equipment status information parameters, and production process parameters, and processes the collected data, and transmits the processed data to S3;

[0010] S3: Perform feature analysis on the processed data, extract features for building a digital twin model from the data, and transmit the extracted features for building the digital twin model to S4;

[0011] S4: Based on project requirements, use 3D modeling software to build a 3D model of the physical entity, and map the characteristics of the physical entity into a mathematical model to build a digital twin model;

[0012] S5: Use simulation tools to simulate and analyze the constructed digital twin model to obtain simulation evaluation coefficients, evaluate the performance of the physical entity through the simulation evaluation coefficients, and transmit the simulation evaluation coefficients to S6;

[0013] S6: used to calculate the actual demand assessment coefficient, perform difference analysis based on the simulation assessment coefficient and the actual demand assessment coefficient, calculate the difference factor, optimize and adjust the digital twin model based on the difference factor, and transmit the optimized and adjusted digital twin model to S7 for deployment;

[0014] S7: Deploy the adjusted digital twin model to the target platform, integrate the digital twin model with the system, and realize data sharing and interaction.

[0015] Preferably, S2 collects users' specific requirements for building a digital twin model through questionnaires, meetings, and work visits, sets performance goals for the model based on user requirements, including real-time, accuracy, and stability, and determines the entities, processes, environmental factors, interactions between entities and processes, interactions between entities and environmental factors, and interactions between processes and environmental factors of the digital twin model. The collected requirements are then organized into a document, including functional requirements, non-functional requirements, and constraints. The requirements document is reviewed by users, and user confirmation of the requirements document is obtained. The goals and scope of model construction are prioritized based on user requirements and resource conditions.

[0016] Preferably, the S2 installs sensors and monitoring equipment to collect the required data types, including environmental parameters, equipment status information, and production process parameters, through the user's request data for model construction, and transmits the collected data to the computer terminal through a wireless network connection. The sensors are installed on physical entities to collect environmental parameters and equipment status information such as temperature, humidity, pressure, and vibration. The monitoring equipment is configured with monitoring cameras and RFID readers to collect visual information and location data, and then the collected data is cleaned, converted, and stored.

[0017] Preferably, the feature analysis of S3 is performed by defining the network architecture of the encoder and decoder, including the number and size of hidden layers, and first inputting the original data x ito the encoder, which passes x through a series of hidden layers. i Convert it to a lower-dimensional feature representation h, which is then received by the decoder and converted into the reconstructed original data Calculate the reconstruction error n represents the number of samples in the training data set, x i represents the i-th sample in the original data set, Represents the i-th sample reconstructed by the automatic encoder. By setting a reconstruction error threshold M, the reconstruction error L is compared with the reconstruction error threshold M to evaluate whether the feature h extracted by the encoder is valid. When the reconstruction error L≤reconstruction error threshold M, the feature h is considered to be valid, and the valid feature h is screened out and transmitted to S4. When the reconstruction error L>reconstruction error threshold M, the feature h is considered to be invalid, and the invalid feature h is screened out.

[0018] Preferably, the specific steps of constructing the data twin model of S4 are:

[0019] Step S411: Select appropriate 3D modeling software according to project requirements, including AutoCAD, 3DMax, and SolidWorks;

[0020] Step S412: drawing a three-dimensional model, drawing a corresponding three-dimensional model according to the shape, size and characteristics of the physical entity, and optimizing and adjusting the three-dimensional model, including adjusting the geometric shape, texture and lighting of the model;

[0021] Step S413: Establish a data mapping relationship, input the pre-processed environmental parameters, equipment status information parameters, and production process parameters into the created digital twin model, define the correspondence between the physical entity layout and the elements in the digital twin model by creating a mapping function, and set the actual situation of the physical entity and the requirements of the digital twin model as the mapping relationship between the data, including matching the position data in the physical entity with the coordinate points in the model, and the visual information with the visual information fields in the model, and then compare the model with the physical entity layout to ensure that the position, size, and proportion of all elements in the model are consistent with the physical entity layout;

[0022] Step S414: define physical attributes, environmental attributes, and device attributes;

[0023] Step S415: Establish corresponding mathematical equations and models according to the working principle of the physical entity;

[0024] Step S416: Integrate the mathematical model with the three-dimensional geometric model to form a digital twin model that simulates the behavior of the entity;

[0025] Step S417: Divide the feature h into a training set, a validation set, and a test set, use the training set data to input into the digital twin model to train the model, use the validation set data to adjust the trained model parameters, and then use the test set data to evaluate the performance indicators of the model.

[0026] Preferably, S5 performs simulation analysis on the digital twin model by simulating the operating state and behavior of the physical entity, and calculates the simulation evaluation coefficient as

[0027] S j represents the simulation value of the jth model output, O j represents the jth actual measured physical entity performance value, t p represents the optimal time required to complete the simulation, t a Indicates the time actually required to complete the simulation, D s Indicates the number of successful simulation runs, D t represents the total number of simulation runs, N represents the number of measured physical entity performance values, and α1, α2, α3, and α4 are weight coefficients.

[0028] Preferably, S6 analyzes the needs and goals of the actual application scenario and calculates the actual demand assessment coefficient as U k represents the user's satisfaction score for the kth requirement, U max represents the maximum satisfaction score, K represents the total number of requirements, S c represents the measured value of the quality index of the cth actual physical entity, S sc represents the standard value of the cth actual physical entity quality indicator, C represents the total number of quality indicators, B represents the total benefit brought by the model, H represents the total cost of the model, and Y v represents the adaptability score of the model under the vth environmental condition, Y max represents the maximum score of adaptability, V represents the total number of environmental conditions, β1, β2, β3, β4 are weight coefficients;

[0029] The specific calculation formula of the difference factor is: R represents the simulation evaluation coefficient, and Q represents the actual demand evaluation coefficient;

[0030] By comparing the difference factor F with the preset difference threshold θ, if F≤θ, it is considered that the difference between the simulation result and the actual demand is within an acceptable range, and the constructed digital twin model is directly deployed to the target platform. If F>θ, it is considered that there is a large difference, and the digital twin model needs to be optimized and adjusted, and the changes in the adjusted difference factor are monitored in real time until the value of the difference factor is within the preset difference threshold, and then the digital twin model is deployed.

[0031] Preferably, the S7 converts the digital twin model into a format that matches running on the platform, and transmits it to the platform through a communication interface for deployment and integration.

[0032] To achieve the above object, the present invention provides the following technical solution: a digital twin model construction system, implementing the above digital twin model construction method, comprising:

[0033] Demand determination module: Determines the goal and scope of model construction, collects user requirements for the digital twin model, sets the business goals of the model in terms of performance, thereby creating the physical entity of the digital twin, and transmits the determined model construction goals and scope to the data collection module;

[0034] Data collection module: receives user request data for model construction, uses sensors, monitoring equipment and the Internet to collect various data generated by physical entities, including environmental parameters, equipment status information parameters, production process parameters, and processes the collected data;

[0035] Feature analysis module: performs feature analysis on the processed data and extracts features from the data to build the digital twin model;

[0036] Model building module: Based on project requirements, use 3D modeling software to build a 3D model of the physical entity, map the characteristics of the physical entity into a mathematical model, and build a digital twin model;

[0037] Simulation analysis module: Use simulation tools to simulate and analyze the constructed digital twin model to obtain the simulation evaluation coefficient, and evaluate the performance of the physical entity through the simulation evaluation coefficient;

[0038] Difference analysis module: used to calculate the actual demand assessment coefficient, perform difference analysis based on the simulation assessment coefficient and the actual demand assessment coefficient, calculate the difference factor, and optimize and adjust the digital twin model based on the difference factor;

[0039] Model deployment module: deploys the adjusted digital twin model to the target platform, integrates the digital twin model with the system, and realizes data sharing and interaction.

[0040] Technical effects and advantages of the present invention:

[0041] The present invention collects user requirements for digital twin models and sets business objectives for the model in terms of performance, thereby creating a digital twin physical entity. After receiving user request data for model construction, the present invention uses sensors, monitoring equipment, and the Internet to collect various data generated by the physical entity, processes the collected data, performs feature analysis on the processed data, and extracts features for building a digital twin model from the data. According to project requirements, a three-dimensional model of the physical entity is constructed using three-dimensional modeling software, and the features of the physical entity are mapped to a mathematical model to construct a digital twin model. The constructed digital twin model is then simulated, analyzed, and evaluated using simulation tools. Evaluation, perform difference analysis based on simulation evaluation results and actual needs, optimize and adjust the digital twin model, and deploy the adjusted digital twin model to the target platform, integrate the digital twin model with the system, realize data sharing and interaction, and use intelligent demand analysis methods to fully and deeply understand the real needs of complex systems, reduce the deviation between the model and the actual system, conduct in-depth feature mining and analysis on the collected data, reveal the laws and trends behind the data, and analyze the differences between the simulation analysis results and the actual needs, which is conducive to accurately identifying the differences and deviations between the virtual model and the actual system, and accurately positioning the optimization and adjustment of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a diagram of the method steps of the present invention.

[0043] Figure 2 It is a system structure block diagram of the present invention. DETAILED DESCRIPTION

[0044] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. In addition, the forms of the various structures recorded in the following embodiments are merely illustrative. The digital twin model construction method and system involved in the present invention are not limited to the various structures recorded in the following embodiments. All other implementations obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0045] like Figure 1 This embodiment provides a method for constructing a digital twin model, including:

[0046] S1: Determine the goals and scope of model construction, collect user requirements for the digital twin model, set the business goals of the model in terms of performance, thereby creating the physical entity of the digital twin, and transmit the determined model construction goals and scope to S2.

[0047] In this embodiment, S2 collects users' specific requirements for building a digital twin model through questionnaires, meetings, and work visits, sets performance goals for the model based on user requirements, including real-time, accuracy, and stability, and determines the entities, processes, environmental factors, interactions between entities and processes, interactions between entities and environmental factors, and interactions between processes and environmental factors of the digital twin model. The collected requirements are then organized into a document, including functional requirements, non-functional requirements, and constraints. The requirements document is reviewed by users, and user confirmation of the requirements document is obtained. The goals and scope of model construction are prioritized based on user requirements and resource conditions.

[0048] It should be noted that the specific operation process of prioritizing model building goals and scope includes:

[0049] Step 1: Collect user needs through research, interviews, questionnaires, etc., and record the needs in detail;

[0050] Step 2: Analyze the collected requirements, evaluate and score each goal and scope according to the evaluation criteria, and classify the requirements into "must have", "should have", "can have" and "will not have this time". Create a matrix to match the requirements with the evaluation criteria, let stakeholders score the requirements, and determine the priority based on the scores obtained;

[0051] Step 3: Based on the evaluation results, rank the goals and scopes. The ranking includes:

[0052] (1) Set high-value, low-difficulty requirements as the highest priority;

[0053] (2) Setting high-value and high-difficulty requirements as requiring further assessment of resource conditions and risks;

[0054] (3) Put low-value, low-difficulty requirements in a secondary position;

[0055] (4) Low-value, high-difficulty requirements are set as likely to be postponed or cancelled.

[0056] S2: Receives user request data for model construction, uses sensors, monitoring equipment, and the Internet to collect various data generated by physical entities, including environmental parameters, equipment status information parameters, and production process parameters, processes the collected data, and transmits the processed data to S3.

[0057] In this embodiment, the S2 installs sensors and monitoring equipment to collect the required data types, including environmental parameters, equipment status information, and production process parameters, through the user's request data for model construction, and transmits the collected data to the computer terminal through a wireless network connection. The sensor is installed on a physical entity to collect environmental parameters and equipment status information such as temperature, humidity, pressure, and vibration. The monitoring equipment is configured with monitoring cameras and RFID readers to collect visual information and location data, and then the collected data is cleaned, converted, and stored.

[0058] S3: Perform feature analysis on the processed data, extract features for building a digital twin model from the data, and transmit the extracted features for building a digital twin model to S4.

[0059] In this embodiment, the feature analysis of S3 is performed by defining the network architecture of the encoder and decoder, including the number and size of hidden layers. First, the original data x is input. i to the encoder, which passes x through a series of hidden layers. i Convert it to a lower-dimensional feature representation h, which is then received by the decoder and converted into the reconstructed original data Calculate the reconstruction error n represents the number of samples in the training data set, x i represents the i-th sample in the original data set, Represents the i-th sample reconstructed by the automatic encoder. By setting a reconstruction error threshold M, the reconstruction error L is compared with the reconstruction error threshold M to evaluate whether the feature h extracted by the encoder is valid. When the reconstruction error L≤reconstruction error threshold M, the feature h is considered to be valid, and the valid feature h is screened out and transmitted to S4. When the reconstruction error L>reconstruction error threshold M, the feature h is considered to be invalid, and the invalid feature h is screened out.

[0060] S4: According to project requirements, use 3D modeling software to build a 3D model of the physical entity, and map the characteristics of the physical entity into a mathematical model to build a digital twin model.

[0061] In this embodiment, the specific steps of constructing the data twin model of S4 are as follows:

[0062] Step S411: Select appropriate 3D modeling software according to project requirements, including AutoCAD, 3DMax, and SolidWorks;

[0063] Step S412: drawing a three-dimensional model, drawing a corresponding three-dimensional model according to the shape, size and characteristics of the physical entity, and optimizing and adjusting the three-dimensional model, including adjusting the geometric shape, texture and lighting of the model;

[0064] Step S413: Establish a data mapping relationship, input the pre-processed environmental parameters, equipment status information parameters, and production process parameters into the created digital twin model, define the correspondence between the physical entity layout and the elements in the digital twin model by creating a mapping function, and set the actual situation of the physical entity and the requirements of the digital twin model as the mapping relationship between the data, including matching the position data in the physical entity with the coordinate points in the model, and the visual information with the visual information fields in the model, and then compare the model with the physical entity layout to ensure that the position, size, and proportion of all elements in the model are consistent with the physical entity layout;

[0065] Step S414: define physical attributes, environmental attributes, and device attributes;

[0066] Step S415: Establish corresponding mathematical equations and models according to the working principle of the physical entity;

[0067] Step S416: Integrate the mathematical model with the three-dimensional geometric model to form a digital twin model that simulates the behavior of the entity;

[0068] Step S417: Divide the feature h into a training set, a validation set, and a test set, use the training set data to input into the digital twin model to train the model, use the validation set data to adjust the trained model parameters, and then use the test set data to evaluate the performance indicators of the model.

[0069] S5: Use simulation tools to perform simulation analysis on the constructed digital twin model to obtain simulation evaluation coefficients, evaluate the performance of the physical entity through the simulation evaluation coefficients, and transmit the simulation evaluation coefficients to S6.

[0070] In this embodiment, S5 simulates the operation status and behavior of the physical entity to perform simulation analysis on the digital twin model and calculates the simulation evaluation coefficient as

[0071] S j represents the simulation value of the jth model output, O j represents the jth actual measured physical entity performance value, t p represents the optimal time required to complete the simulation, t a Indicates the time actually required to complete the simulation, D s Indicates the number of successful simulation runs, D trepresents the total number of simulation runs, N represents the number of measured physical entity performance values, and α1, α2, α3, and α4 are weight coefficients.

[0072] S6: used to calculate the actual demand assessment coefficient, perform difference analysis based on the simulation assessment coefficient and the actual demand assessment coefficient, calculate the difference factor, optimize and adjust the digital twin model based on the difference factor, and transfer the optimized and adjusted digital twin model to S7 for deployment.

[0073] In this embodiment, S6 analyzes the needs and goals of the actual application scenario and calculates the actual demand assessment coefficient as U k represents the user's satisfaction score for the kth requirement, U max represents the maximum satisfaction score, K represents the total number of requirements, S c represents the measured value of the quality index of the cth actual physical entity, S sc represents the standard value of the cth actual physical entity quality indicator, C represents the total number of quality indicators, B represents the total benefit brought by the model, H represents the total cost of the model, and Y v represents the adaptability score of the model under the vth environmental condition, Y max represents the maximum score of adaptability, V represents the total number of environmental conditions, β1, β2, β3, β4 are weight coefficients;

[0074] The specific calculation formula of the difference factor is: R represents the simulation evaluation coefficient, and Q represents the actual demand evaluation coefficient;

[0075] By comparing the difference factor F with the preset difference threshold θ, if F≤θ, it is considered that the difference between the simulation result and the actual demand is within an acceptable range, and the constructed digital twin model is directly deployed to the target platform. If F>θ, it is considered that there is a large difference, and the digital twin model needs to be optimized and adjusted, and the changes in the adjusted difference factor are monitored in real time until the value of the difference factor is within the preset difference threshold, and then the digital twin model is deployed.

[0076] S7: Deploy the adjusted digital twin model to the target platform, integrate the digital twin model with the system, and realize data sharing and interaction.

[0077] In this embodiment, the S7 converts the digital twin model into a format that matches running on the platform, and transmits it to the platform through a communication interface for deployment and integration.

[0078] like Figure 2The embodiment shown provides an implementation system corresponding to a digital twin model construction method, including a demand determination module, a data collection module, a feature analysis module, a model construction module, a simulation analysis module, a difference analysis module and a model deployment module. The demand determination module is connected to the data collection module, the data collection module is connected to the feature analysis module, the feature analysis module is connected to the model construction module, the model construction module is connected to the simulation analysis module, the simulation analysis module is connected to the difference analysis module, and the difference analysis module is connected to the model deployment module.

[0079] The demand determination module determines the goal and scope of model construction, collects user requirements for the digital twin model, sets the business goals of the model in terms of performance, thereby creating the physical entity of the digital twin, and transmits the determined model construction goal and scope to the data collection module;

[0080] The data collection module receives the user's request data for model construction, uses sensors, monitoring equipment and the Internet to collect various data generated by physical entities, including environmental parameters, equipment status information parameters, production process parameters, and processes the collected data;

[0081] The feature analysis module performs feature analysis on the processed data and extracts features for building a digital twin model from the data;

[0082] The model building module uses 3D modeling software to build a 3D model of the physical entity according to project requirements, and maps the characteristics of the physical entity into a mathematical model to build a digital twin model;

[0083] The simulation analysis module uses a simulation tool to perform simulation analysis on the constructed digital twin model to obtain a simulation evaluation coefficient, and evaluates the performance of the physical entity through the simulation evaluation coefficient;

[0084] The difference analysis module is used to calculate the actual demand assessment coefficient, perform difference analysis based on the simulation assessment coefficient and the actual demand assessment coefficient, calculate the difference factor, and optimize and adjust the digital twin model based on the difference factor;

[0085] The model deployment module deploys the adjusted digital twin model to the target platform, integrates the digital twin model with the system, and realizes data sharing and interaction.

[0086] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0087] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for constructing a digital twin model, characterized in that: include: S1: Determine the goal and scope of model construction, collect user requirements for the digital twin model, set the business goals of the model in terms of performance, thereby creating the physical entity of the digital twin, and transmit the determined model construction goals and scope to S2; S2: accepts user request data for model construction, uses sensors, monitoring equipment and the Internet to collect various data generated by physical entities, including environmental parameters, equipment status information parameters, and production process parameters, and processes the collected data, and transmits the processed data to S3; S3: Perform feature analysis on the processed data, extract features for building a digital twin model from the data, and transmit the extracted features for building the digital twin model to S4; S4: Based on project requirements, use 3D modeling software to build a 3D model of the physical entity, and map the characteristics of the physical entity into a mathematical model to build a digital twin model; S5: Use simulation tools to simulate and analyze the constructed digital twin model to obtain simulation evaluation coefficients, evaluate the performance of the physical entity through the simulation evaluation coefficients, and transmit the simulation evaluation coefficients to S6; S6: used to calculate the actual demand assessment coefficient, perform difference analysis based on the simulation assessment coefficient and the actual demand assessment coefficient, calculate the difference factor, optimize and adjust the digital twin model based on the difference factor, and transmit the optimized and adjusted digital twin model to S7 for deployment; S7: Deploy the adjusted digital twin model to the target platform, integrate the digital twin model with the system, and realize data sharing and interaction.

2. A digital twin model construction system, implementing a digital twin model construction method as claimed in claim 1, characterized in that: include: Demand determination module: Determines the goal and scope of model construction, collects user requirements for the digital twin model, sets the business goals of the model in terms of performance, thereby creating the physical entity of the digital twin, and transmits the determined model construction goals and scope to the data collection module; Data collection module: receives user request data for model construction, uses sensors, monitoring equipment and the Internet to collect various data generated by physical entities, including environmental parameters, equipment status information parameters, production process parameters, and processes the collected data; Feature analysis module: performs feature analysis on the processed data and extracts features from the data to build the digital twin model; Model building module: Based on project requirements, use 3D modeling software to build a 3D model of the physical entity, map the characteristics of the physical entity into a mathematical model, and build a digital twin model; Simulation analysis module: Use simulation tools to simulate and analyze the constructed digital twin model to obtain the simulation evaluation coefficient, and evaluate the performance of the physical entity through the simulation evaluation coefficient; Difference analysis module: used to calculate the actual demand assessment coefficient, perform difference analysis based on the simulation assessment coefficient and the actual demand assessment coefficient, calculate the difference factor, and optimize and adjust the digital twin model based on the difference factor; Model deployment module: deploys the adjusted digital twin model to the target platform, integrates the digital twin model with the system, and realizes data sharing and interaction.

Images