Model design simulation method and system based on artificial intelligence and computer medium

Abstract

The invention discloses an artificial intelligence-based model design simulation method, a model design simulation system and a computer medium. The artificial intelligence-based model design simulation method comprises the following steps of: obtaining a target simulation model and a simulation template containing simulation environment information; defining physical attributes of the target simulation model; defining a target simulation condition by using the simulation template; and generating a prediction simulation result of the target simulation model in a specified simulation environment and condition through the artificial intelligence model. According to the method, the simulation states under different models and conditions are quickly predicted through the artificial intelligence technology, the product design and verification period is effectively shortened, the design cost is reduced, and the engineering design efficiency and the simulation reliability are improved.

Description

Technical Field

[0001] This application is aimed at the field of industrial design, and in particular at an artificial intelligence-based model design simulation method, a model design simulation system, and a computer medium. Background Art

[0002] In the current industrial design field, in-house engineers are often required to create customized templates tailored to specific scenarios and design requirements. These templates guide engineers in modeling and simulation setup, ensuring standardized and standardized simulations. To further improve efficiency, existing simulation tools are often redeveloped based on standard specifications to create simplified, relatively fixed-function wizard-style simulation templates. This allows engineers to quickly execute specific simulation types and quickly obtain corresponding simulation results and analysis reports.

[0003] However, this traditional design simulation method also has some drawbacks. First, the templates have limited functionality and are relatively fixed. Redevelopment and optimization are often required for different business scenarios, and ordinary engineers often lack the skills to develop and optimize templates, resulting in high labor costs. Furthermore, even after the templates are developed, engineers are still required to perform simulations based on the actual design models. This requires a large amount of computation and a long simulation cycle. Finally, after obtaining simulation results, engineers must re-optimize the design and re-enter the simulation process for unsatisfactory results, further increasing the design cycle and design costs. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide an industrial design software design simulation method based on artificial intelligence and database.

[0005] To achieve the above objectives, the present invention discloses a model design simulation method based on artificial intelligence, comprising:

[0006] Acquire a target simulation model and a simulation template, wherein the simulation template includes simulation environment information;

[0007] Define the physical properties of the target simulation model;

[0008] Based on the simulation template, define target simulation conditions, where the target simulation conditions include but are not limited to one or more of simulation conditions, boundary conditions, connection definitions, and fixed definitions;

[0009] The state changes of the target simulation model after the target simulation conditions are performed under the simulation environment information are generated based on the artificial intelligence model, and the state changes are output as the predicted simulation results.

[0010] Preferably, obtaining the target simulation model includes:

[0011] Obtaining customized initial design parameters, and generating a target simulation model based on the customized initial design parameters;

[0012] Alternatively, a template library is called to obtain a simulation model in the template library as a target simulation model;

[0013] The template library includes but is not limited to one or more of design templates, simulation models, design data, simulation data, and simulation conclusions.

[0014] Preferably, defining the physical properties of the target simulation model includes:

[0015] Calibrate the physical properties of the target simulation model, the calibration content includes physical properties;

[0016] Alternatively, the target simulation model is input into the artificial intelligence model, and the physical properties of the target simulation model are identified by the artificial intelligence model;

[0017] The physical properties include, but are not limited to, one or more of the target simulation model or the component names, part sizes, part shapes, part types, part materials, connection relationships between parts, fastener information, and the fixing methods between parts.

[0018] Preferably, generating a state change of a target simulation model based on an artificial intelligence model after performing target simulation conditions under simulation environment information, and outputting the state change as a predicted simulation result includes:

[0019] generating an artificial intelligence model based on a database, wherein the database includes but is not limited to one or more of historical design parameters, historical simulation templates, and historical simulation results based on the historical simulation templates;

[0020] At least one prediction function is established in the artificial intelligence model, wherein the prediction function is constructed by cleaning the data in the database and extracting simulation features.

[0021] Preferably, the model design simulation method further includes:

[0022] Input the target simulation model into the simulation template to obtain the actual simulation results;

[0023] Upload the target simulation model, simulation template, and actual simulation results to the artificial intelligence model and output them as new data;

[0024] The artificial intelligence model is trained on the newly added data to modify the prediction function and correct the prediction simulation results.

[0025] Preferably, the model design simulation method further includes:

[0026] Based on the predicted simulation results, the AI ​​model generates optimization suggestions for the target simulation model.

[0027] The present invention also discloses a model design simulation method system based on artificial intelligence, comprising:

[0028] An acquisition module is used to acquire a target simulation model and a simulation module, wherein the simulation module includes simulation environment information;

[0029] A definition module defines the physical properties of the target simulation model; and based on the simulation template, defines the target simulation conditions, wherein the target simulation conditions include but are not limited to one or more of simulation conditions, boundary conditions, connection definitions, and fixed definitions;

[0030] The artificial intelligence model generates the state change of the target simulation model after the target simulation conditions are performed under the simulation environment information, and outputs the state change as the predicted simulation result.

[0031] Preferably, the acquisition module includes:

[0032] A template building unit, storing or forming a simulation template;

[0033] Design unit, adjusts the parameter types of existing models in the simulation template;

[0034] The simulation unit generates actual simulation results for the target simulation model according to the simulation module;

[0035] Artificial intelligence models include:

[0036] The prediction unit is based on the simulation template and calls the database to generate the prediction simulation results.

[0037] Preferably, it further comprises a database;

[0038] The database is a local database and / or a cloud database; the artificial intelligence model communicates with the database and is deployed locally and / or in the cloud.

[0039] The present invention also discloses a computer storage medium, which stores a computer program. The computer program is used to execute the above-mentioned model design simulation method.

[0040] In summary, this application includes at least one of the following beneficial technical effects:

[0041] 1. The present invention can efficiently and quickly predict the state changes of the target simulation model under different simulation conditions and simulation environment information based on the artificial intelligence model, significantly improving the efficiency of engineering design simulation and saving manpower, material resources and time costs in the simulation verification process.

[0042] 2. A template library is provided. By calling existing data from the template library or quickly generating target simulation models based on customized initial design parameters, the system achieves diversified and highly adaptable model generation, facilitating rapid response to design change requirements. Furthermore, the system utilizes artificial intelligence models to automatically identify and calibrate the physical properties of the target simulation model, replacing the tedious manual property definition process. This significantly improves the accuracy of simulation modeling, reduces the probability of error, and optimizes the reliability of design simulation work.

[0043] 3. Utilizing databases: On the one hand, the historical data stored in the database can provide data support for AI prediction and simulation results. On the other hand, actual operation data can be collected to continuously improve the accuracy of AI prediction functions. By continuously increasing database capacity, the AI's prediction capabilities and accuracy can be enhanced, significantly improving design and simulation efficiency.

[0044] 4. Furthermore, based on the database, the AI ​​model can autonomously provide targeted optimization suggestions based on simulation prediction results. This provides users with design ideas, reducing the time spent on manual experimentation and trial and error. This effectively optimizes product design solutions, significantly accelerating product design decision-making and the optimization and upgrade process. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A logical diagram of the simulation method for the model design in this application;

[0046] Figure 2 This is a logical diagram of another implementation of the model design simulation method in this application. DETAILED DESCRIPTION

[0047] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0048] It should be noted that, unless otherwise specified, all technical and scientific terms used in this application have the same meaning as commonly understood by ordinary technicians in the technical field to which this application belongs.

[0049] See also Figure 1 , Figure 1 Schematic diagram of the logic of the simulation method for the model design in this application.

[0050] like Figure 1 As shown, the present invention discloses a model design simulation method based on artificial intelligence, comprising:

[0051] Acquire a target simulation model and a simulation template, wherein the simulation template includes simulation environment information;

[0052] Define the physical properties of the target simulation model;

[0053] Based on the simulation template, define target simulation conditions, where the target simulation conditions include but are not limited to one or more of simulation conditions, boundary conditions, connection definitions, and fixed definitions;

[0054] The state changes of the target simulation model after the target simulation conditions are performed under the simulation environment information are generated based on the artificial intelligence model, and the state changes are output as the predicted simulation results.

[0055] Here, the specific design process of industrial software needs to be described for easier understanding. Specifically, the industrial software can be CATIA, but it is not limited to this. It can also be other industrial software with simulation capabilities, such as UG, CREO, etc. In industrial software, design can be divided into model construction and model definition. In the model construction stage, the user designs a preliminary model structure, part type, and connection method based on the specific needs, functional specifications, and performance targets of the product, thereby obtaining a target simulation model. Then, the specific physical properties of the target simulation model need to be defined. After obtaining a complete target simulation model, a specific simulation analysis can be performed on the target simulation model. According to the specific simulation type requirements (such as structural dynamics analysis, fluid dynamics analysis, and / or electromagnetic analysis, etc.) or existing modeling standards or simulation specifications or different application scenarios, a simulation template is set. The simulation template includes simulation environment information, and the target simulation conditions are defined based on the simulation template. The target simulation conditions include but are not limited to one or more of simulation working conditions, boundary conditions, connection definitions, and fixed definitions, such as which positions should be fixed, which positions should be pressured or accelerated, etc., and the corresponding physical boundary conditions are applied in combination with the actual scenario. Then, simulation calculations are carried out through the simulation solver to obtain the dynamic and static system behavior data of the model under specified conditions.

[0056] However, this design simulation approach also has some drawbacks. The first is that the templates are limited in functionality and relatively fixed. Redevelopment and optimization are often necessary for different business scenarios, and ordinary engineers often lack the skills to develop and optimize templates, resulting in high labor costs. Furthermore, even after the templates are developed, engineers are still required to run simulation calculations based on the actual design model. This requires a large amount of computation and results in a long simulation cycle. Finally, after obtaining simulation results, engineers must re-optimize the design and re-enter the simulation process for any unsatisfactory results, further increasing the design cycle and design costs.

[0057] Unlike the prior art, this application can use the artificial intelligence model to directly predict the simulation results after setting the target simulation model and simulation template. When the prediction result is poor, there is no need to enter the actual simulation program, and the prediction result can also provide a guide for design optimization. And it is understood by those skilled in the art that compared to actually running the simulation calculation program, the artificial intelligence model can generate predicted simulation results in a shorter time (such as minutes or even seconds).

[0058] Therefore, the design simulation method provided in this application can efficiently and quickly predict the state changes of the target simulation model under different simulation conditions and simulation environment information based on the artificial intelligence model, significantly improving the efficiency of engineering design simulation and saving manpower, material and time costs in the simulation verification process.

[0059] It will be appreciated by those skilled in the art that the specific type of artificial intelligence model used herein is not limited. For example, it may be one or a combination of ChatGPT, deepseekV1, and deepseekR3. This application does not impose any limitation on this.

[0060] The above is a detailed description of the basic concept of this application. The following will describe the specific possible implementation methods of each step in conjunction with the accompanying drawings.

[0061] First, those skilled in the art will appreciate that there is no limitation to the specific method of obtaining the target simulation model.

[0062] In one possible implementation, obtaining the target simulation model includes obtaining customized initial design parameters and generating the target simulation model based on the customized initial design parameters. This may be understood as manually constructing the target simulation model based on the user's customization. The parameters may include, but are not limited to, specific model size and shape.

[0063] In another possible implementation, a template library is called and a simulation model from the template library is used as the target simulation model. Specifically, the user directly calls the template library and uses the simulation model obtained from the template library as the target simulation model, further reducing the difficulty of starting the design. The template library includes, but is not limited to, one or more of design templates, simulation models, design data, simulation data, and simulation results.

[0064] Secondly, there is no restriction on the way of defining the physical properties of the target simulation model.

[0065] In one possible implementation, the physical properties of the target simulation model are calibrated, including the physical properties. Alternatively, the target simulation model is input into the artificial intelligence model, and the artificial intelligence model identifies the physical properties of the target simulation model. The physical properties include, but are not limited to, one or more of the following: part name, part size, part shape, part type, part material, inter-part connection relationship, fastener information, and inter-part fixing method of the target simulation model or its components.

[0066] Specifically, after the target simulation model is preliminarily constructed, it is also necessary to perform specific calibration on the physical properties of the simulation model to obtain a complete target simulation model that can be used for simulation. Therefore, this application provides two modes. One is for the user to manually define the physical properties. The other is to calibrate through an artificial intelligence model. In addition, the physical properties can be part information of the target simulation model, such as information on each part in the entire vehicle, or part information that constitutes the target simulation model components, such as information on each part of the door in the entire vehicle. For example, taking the analysis of the door as an example, the artificial intelligence model can automatically perform grid division and part classification on the target simulation model to achieve one-click operation for the user, further simplifying the operation.

[0067] The above are the design steps in the model design and simulation method provided by this application. In the above steps, a complete target simulation model that can be used for simulation is obtained. The following will explain how to simulate the target simulation model.

[0068] Those skilled in the art will understand that after obtaining the target simulation model, a simulation template needs to be constructed. The simulation template needs to define conditions including but not limited to working condition and boundary condition definition, connection definition, hinge definition, welding information definition, analysis plan, etc.

[0069] As mentioned above, the template library also includes simulation templates. Therefore, this application also provides two methods for using simulation templates. In one possible implementation, users can define simulation templates themselves. In another possible implementation, users can directly call existing simulation templates from the template library, thereby minimizing the need for manual work.

[0070] After obtaining the target simulation model and simulation template, you can enter the simulation process.

[0071] In one possible implementation, the simulation process can predict simulation results for the output of the artificial intelligence model.

[0072] The state change of the target simulation model generated based on the artificial intelligence model after the target simulation conditions are carried out under the simulation environment information, and the state change is output as a prediction simulation result including:

[0073] generating an artificial intelligence model based on a database, wherein the database includes but is not limited to one or more of historical design parameters, historical simulation templates, and historical simulation results based on the historical simulation templates;

[0074] At least one prediction function is established in the artificial intelligence model, wherein the prediction function is constructed by cleaning the data in the database and extracting simulation features.

[0075] Specifically, the present application also provides a database that includes, but is not limited to, one or more of historical design parameters, historical simulation templates, and historical simulation results based on these historical simulation templates. This provides data support for the artificial intelligence model to predict simulation results, enabling it to make predictions for similar scenarios and simulation conclusions with relatively accurate prediction results. In addition, at least one prediction function is established within the artificial intelligence model. The prediction function is constructed by cleaning the data in the database and extracting simulation features.

[0076] The above are predictive results obtained using artificial intelligence models. We can also use simulation solvers to carry out simulation calculations to obtain actual simulation results.

[0077] See also Figure 2 , Figure 2 This is a logical diagram of another implementation of the model design simulation method in this application.

[0078] Therefore, it will be understood by those skilled in the art that Figure 2 As shown, in another possible implementation, the model design simulation method further includes:

[0079] Input the target simulation model into the simulation template to obtain the actual simulation results;

[0080] Upload the target simulation model, simulation template, and actual simulation results to the artificial intelligence model and output them as new data;

[0081] The artificial intelligence model is trained on the newly added data to modify the prediction function and correct the prediction simulation results.

[0082] Specifically, in certain scenarios (for example, when the artificial intelligence model predicts that the target simulation model meets the design requirements), the simulation solver can actually apply the simulation template to the target simulation model for simulation calculations, and obtain the actual calculated simulation results. The target simulation model, simulation template, and actual simulation results can then be uploaded to the artificial intelligence model and output as new data, which can be further used to train the artificial intelligence model based on the new data. The prediction function and simulation results can be corrected through deep learning algorithms and other methods, thereby achieving the recycling of each simulation result. By continuously collecting data such as actual running design parameters, simulation templates, and simulation structures, the accuracy of the prediction function in the artificial intelligence model can be continuously improved, and the database data can be continuously enriched, thereby improving the prediction ability and accuracy of the artificial intelligence model, thereby greatly improving the design and simulation efficiency.

[0083] The above is a way to improve the prediction ability of the artificial intelligence model of this application for results. Furthermore, based on the artificial intelligence model, the design efficiency can be further improved.

[0084] For example, in one possible implementation, the model design simulation method further includes:

[0085] Based on the predicted simulation results, the AI ​​model generates optimization suggestions for the target simulation model. Specifically, based on a database, the AI ​​model can autonomously provide targeted optimization suggestions based on the simulation prediction results. This provides users with design ideas, reducing the time required for manual experimentation and trial and error. This effectively optimizes product design solutions, significantly accelerating product design decision-making and the optimization and upgrade process.

[0086] The above is a specific implementation of the model design simulation method provided in this application.

[0087] This application also discloses a model design simulation method system based on artificial intelligence, including:

[0088] An acquisition module is used to acquire a target simulation model and a simulation module, wherein the simulation module includes simulation environment information;

[0089] A definition module defines the physical properties of the target simulation model; and based on the simulation template, defines the target simulation conditions, wherein the target simulation conditions include but are not limited to one or more of simulation conditions, boundary conditions, connection definitions, and fixed definitions;

[0090] The artificial intelligence model generates the state change of the target simulation model after the target simulation conditions are performed under the simulation environment information, and outputs the state change as the predicted simulation result.

[0091] Specifically, the present application also provides an artificial intelligence-based model design simulation method system, including operation modules corresponding to the aforementioned method. Through these modules, rapid and accurate prediction of simulation results can be achieved.

[0092] Furthermore, in a possible implementation, the acquisition module includes:

[0093] A template building unit, storing or forming a simulation template;

[0094] Design unit, adjusts the parameter types of existing models in the simulation template;

[0095] The simulation unit generates actual simulation results for the target simulation model according to the simulation module;

[0096] Artificial intelligence models include:

[0097] Prediction unit,The prediction unit is based on the simulation template and calls the database to generate prediction simulation results.

[0098] This allows users to easily import and adjust target simulation models and simulation templates in a variety of ways, and to predict simulation results using the prediction unit and database.

[0099] It will be understood by those skilled in the art that the specific implementation of the database and artificial intelligence model in the aforementioned model design simulation system is not limited. In one possible implementation, the database is a local database and / or a cloud database. Specifically, when data security requirements are high, the database can be deployed locally to improve its security. When the database is deployed in the cloud, more templates and more data can be obtained through the network, while providing more template options and improving the design accuracy of the artificial intelligence model with big data. Those skilled in the art can design it themselves as needed, and this application does not constitute any limitation here.

[0100] Similarly, the artificial intelligence model is connected to the database. In one possible implementation, the artificial intelligence model is deployed locally and / or in the cloud. Those skilled in the art can design it as needed, and this application does not constitute any limitation thereto.

[0101] The present invention also discloses a computer storage medium, which stores a computer program. The computer program is used to execute the above-mentioned model design simulation method.

[0102] The above is a detailed description of the basic concept and various possible solutions of this application. An exemplary embodiment will be given below to facilitate a better understanding of the solution.

[0103] The following is an example of car door analysis:

[0104] First, the acquisition module is called to obtain the target simulation model of the vehicle door and design the door model. Specifically, the acquisition module can obtain the model based on user instructions, local files (XML files), or by selecting a design template from a template library. For example, the acquired content may include the dimensions of the vehicle door's outer panel, the structure of the inner panel, the beam layout, the thickness and shape of the reinforcement frame, and the welding method.

[0105] Next, the definition module is called to define the physical properties of the vehicle door. Specifically, the definition module can define the vehicle door based on user instructions or automatic identification by the AI ​​model. For example, the AI ​​model automatically scans the 3D geometry of the vehicle door and identifies part types (such as the outer door panel, side impact beam, inner door panel, hinge, and lock assembly), materials, connection methods, and specifications based on built-in algorithms. It then automatically completes the door model meshing and physical property calibration.

[0106] Next, the simulation module is called to generate a simulation template for this simulation based on user instructions or simulation templates in the template library. For example, the simulation template includes hinge constraint definitions, door lock point load definitions, gravity or other standard load conditions, boundary constraints (such as rigid fixation at the upper and lower hinge positions), and load application locations and intensities (such as acceleration loads, collision loads, and operational loads).

[0107] Finally, the artificial intelligence model uses the aforementioned door model and simulation template, and calls the historical design parameters, historical simulation templates, and historical simulation results of similar scenarios in the database. It uses the prediction function to quickly calculate the simulation results that may be produced under the door design conditions, and quickly provides prediction simulation reports such as stiffness, deformation data, and stress distribution cloud maps. Based on the prediction results, it provides possible optimization suggestions, such as recommended areas and methods for local reinforcement, optimization suggestions for key connection methods and materials, weight reduction optimization locations and measures, etc. The report is automatically generated and output in the form of a local file (XML file). The report contains door structure data and physical property information, simulation working conditions and boundary condition descriptions, artificial intelligence model prediction results, structural optimization solutions recommended by the artificial intelligence model, and predicted component stress and displacement cloud map data.

[0108] Or, optionally, the system inputs the target simulation model and template data into a standard industrial simulation solver (such as Ansys, Nastran, etc.), performs actual simulation calculations, and the software generates accurate simulation results (displacement, stiffness, stress and strain, etc.), and uploads the door model, simulation template and actual simulation results to the database and artificial intelligence model, and outputs them as new data, so as to further train the artificial intelligence model with the new data.

[0109] The embodiments of the methods and systems described above are merely illustrative and not exhaustive. For example, the division of the units is merely a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some communication interface, device or unit, which may be electrical, mechanical or other forms. All other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

Claims

1. A model design simulation method based on artificial intelligence, characterized in that: The model design simulation method comprises: Acquire a target simulation model and a simulation template, wherein the simulation template includes simulation environment information; defining physical properties of the target simulation model; Based on the simulation template, defining target simulation conditions, wherein the target simulation conditions include but are not limited to one or more of simulation conditions, boundary conditions, connection definitions, and fixed definitions; The state change of the target simulation model after the target simulation conditions are performed under the simulation environment information is generated based on the artificial intelligence model, and the state change is output as a predicted simulation result.

2. The model design simulation method according to claim 1, wherein: The acquiring target simulation model comprises: Obtaining customized initial design parameters, and generating a target simulation model based on the customized initial design parameters; Alternatively, calling a template library and obtaining a simulation model in the template library as the target simulation model; The template library includes but is not limited to one or more of design templates, simulation models, design data, simulation data, and simulation conclusions.

3. The model design simulation method according to claim 1, wherein: Defining the physical properties of the target simulation model includes: Calibrate the physical properties of the target simulation model, wherein the calibration content includes the physical properties; Alternatively, the target simulation model is input into the artificial intelligence model, and the physical properties of the target simulation model are identified by the artificial intelligence model; Among them, the physical properties include but are not limited to one or more of the target simulation model or the part name, part size, part shape, part type, part material, connection relationship between parts, fastener information and the fixing method between parts.

4. The model design simulation method according to claim 1, wherein: The generating, based on the artificial intelligence model, a state change of the target simulation model after performing target simulation conditions under the simulation environment information, and outputting the state change as a prediction simulation result comprises: generating an artificial intelligence model based on a database, wherein the database includes but is not limited to one or more of historical design parameters, historical simulation templates, and historical simulation results based on the historical simulation templates; At least one prediction function is established in the artificial intelligence model, wherein the prediction function is constructed by cleaning the data in the database and extracting simulation features.

5. The model design simulation method according to claim 1 or 4, characterized in that: The model design simulation method also includes: Inputting the target simulation model into the simulation template to obtain actual simulation results; Uploading the target simulation model, the simulation template, and the actual simulation results to the artificial intelligence model and outputting them as new data; The artificial intelligence model is trained on the newly added data to modify the prediction function and correct the prediction simulation results.

6. The model design simulation method according to claim 1, wherein: The model design simulation method also includes: Based on the predicted simulation results, the artificial intelligence model generates optimization suggestions for the target simulation model.

7. A model design simulation method system based on artificial intelligence, characterized in that: The model design simulation system includes: An acquisition module is used to acquire a target simulation model and a simulation module, wherein the simulation module includes simulation environment information; A definition module defines the physical properties of the target simulation model; and defines target simulation conditions based on the simulation template, wherein the target simulation conditions include but are not limited to one or more of simulation conditions, boundary conditions, connection definitions, and fixed definitions; The artificial intelligence model generates a state change of the target simulation model after the target simulation conditions are performed under the simulation environment information, and outputs the state change as a predicted simulation result.

8. The model design simulation system according to claim 7, characterized in that: The acquisition module includes: A template building unit, storing or forming a simulation template; A design unit, adjusting the types of parameters of the existing model in the simulation template; A simulation unit, which generates actual simulation results for the target simulation model according to the simulation module; The artificial intelligence model includes: A prediction unit, wherein the prediction unit is based on the simulation template and calls a database to generate the prediction simulation result.

9. The model design simulation system according to claim 8, characterized in that: Also included is a database; The database is a local database and / or a cloud database; the artificial intelligence model is communicated with the database and deployed locally and / or in the cloud.

10. A computer storage medium storing a computer program, wherein the computer program is used to execute the model design simulation method according to any one of claims 1 to 6.

Images