Development and application method and system of low-code-based industrial simulation software

Abstract

The invention relates to the technical field of software, in particular to a low-code-based industrial simulation software development and application method and system.The method comprises the steps that firstly, component packaging is conducted on multiple subjects, and an industrial simulation component library is generated; then, based on a user dragging instruction, assembling a target component in an industrial simulation component library to determine a function node corresponding to the industrial simulation software; generating and operating industrial simulation software on the basis of the assembly result and configuration parameters which are input or preset by a user and correspond to the components; training an AI agent model based on historical simulation data of a plurality of industrial simulation software; and receiving and analyzing the natural language instruction of the user, and generating result data corresponding to the natural language instruction of the user based on an analysis result and the AI proxy model. Therefore, the industrial simulation software can be simply and efficiently generated, the development difficulty and cost of the industrial simulation software are reduced, and the computing power cost of actual simulation calculation is greatly reduced through the use of the AI proxy model.

Description

Technical Field

[0001] The present invention relates to the field of software technology, and in particular to a development and application method and system for low-code-based industrial simulation software. Background Art

[0002] Industrial simulation software is a core technical tool in fields such as intelligent manufacturing and digital twins. It is used to simulate the operating status of physical equipment or systems, thereby optimizing design, predicting performance, and reducing R&D costs. With the advancement of Industry 4.0, companies are facing a surge in demand for real-time, accurate, and interdisciplinary simulation. However, these demands are facing the following bottlenecks:

[0003] Traditional development models require building simulation systems from the ground up, relying on specialized programming skills. This leads to high development costs and lengthy development cycles. Current general-purpose low-code platforms lack specialized components like industrial physics modeling and equipment digital twins, making it difficult to meet specialized simulation needs. There's a significant disconnect between industrial domain knowledge and IT technology, requiring engineers to master cross-disciplinary skills, making talent acquisition difficult and the technical threshold high. The existing industrial software ecosystem is fragmented, with incompatible data formats and complex user interfaces between different tools, resulting in inefficient cross-platform collaboration. Furthermore, traditional simulations generally rely solely on local computing, making it difficult to flexibly schedule cloud resources and resulting in low resource utilization. Therefore, how to efficiently develop and apply industrial simulation software has become a pressing issue. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a development and application method and system for low-code industrial simulation software to overcome the current problems of difficulty, low efficiency and high cost in the development of industrial simulation software.

[0005] To achieve the above objectives, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present application provides a method for developing and applying low-code-based industrial simulation software, comprising:

[0007] Carry out component packaging for multiple disciplines and generate an industrial simulation component library;

[0008] Based on the user's drag and drop instructions, target components in the industrial simulation component library are assembled to determine the functional nodes corresponding to the industrial simulation software;

[0009] generating and running the industrial simulation software based on the assembly results and user-input or preset configuration parameters corresponding to the components;

[0010] Training an AI agent model based on historical simulation data of multiple industrial simulation software;

[0011] Receive and parse user natural language instructions, and generate result data corresponding to the user natural language instructions based on the parsing results and the AI ​​agent model.

[0012] Furthermore, in some embodiments of the present application, the multi-disciplines include multi-body dynamics, thermal analysis, fluid mechanics, structural mechanics, and electromagnetics; the components include a form input component, an input file component, a pre-processing component, a calculation operation component, an application software component, an output file display component, a post-processing component, and an output display component;

[0013] Among them, the form input component is used for users to input forms; the input file component is used for users to input files; the pre-processing component is used to process the data input by users; the calculation operation component is used to include professional application software or physical equations corresponding to industrial simulation software; the application software component is used to apply the data of industrial simulation software; the output file display component is used to display the files input by industrial simulation software; the post-processing component is used to post-process the data output by industrial simulation software; and the output display component is used to display the data output by industrial simulation software.

[0014] Furthermore, in some embodiments of the present application, the following is further included:

[0015] Build semantic associations between components;

[0016] When the user drags the target component, based on the semantic association relationship, the component corresponding to the target component is recommended to the user; or when the user drags the target component, based on the semantic association relationship, conflict detection is performed on the multiple target components dragged by the user and a reminder is issued.

[0017] Furthermore, in some embodiments of the present application, it also includes: inserting a monitoring module into the code of the industrial simulation software, collecting data when the industrial simulation software is running, and storing it in a time series database.

[0018] Furthermore, in some embodiments of the present application, generating and running the industrial simulation software based on the assembly result and user-input or preset configuration parameters corresponding to the component includes:

[0019] Determine the real-time computing power status of cloud servers and edge nodes;

[0020] The simulation task is decomposed into multiple subtasks, and the subtasks are distributed to cloud servers and edge nodes.

[0021] Furthermore, in some embodiments of the present application, the training of the AI ​​agent model based on the historical simulation data of the plurality of industrial simulation software includes:

[0022] Based on the historical simulation data of the plurality of industrial simulation software, a plurality of AI agent models are trained respectively, and an agent model library including the plurality of AI agent models is constructed.

[0023] Furthermore, in some embodiments of the present application, the following is further included:

[0024] Receive and parse user natural language instructions to obtain parsing results;

[0025] Based on the analysis results, a simulation case configuration file is generated through a preset rule template engine;

[0026] A simulation model corresponding to the generated industrial simulation software is matched to the simulation example configuration file, and result data corresponding to the user's natural language instructions is generated through the simulation model and the simulation example configuration file.

[0027] In a second aspect, the present application provides a development and application system for industrial simulation software based on low code, characterized by including:

[0028] Building modules for packaging components for multiple disciplines and generating industrial simulation component libraries;

[0029] A generation module is configured to assemble target components in the industrial simulation component library based on user drag and drop instructions to determine functional nodes corresponding to industrial simulation software; and generate and run the industrial simulation software based on the assembly results and user-input or preset configuration parameters corresponding to the components;

[0030] A training module, configured to train an AI agent model based on historical simulation data of a plurality of said industrial simulation software;

[0031] The application module is used to receive and parse user natural language instructions, and generate result data corresponding to the user's natural language instructions based on the parsing results and the AI ​​agent model.

[0032] The present invention relates to the field of software technology, and specifically to a development and application method and system for industrial simulation software based on low code, the method comprising: first, performing component encapsulation for multiple disciplines to generate an industrial simulation component library; then, based on user drag and drop instructions, assembling target components in the industrial simulation component library to determine the functional nodes corresponding to the industrial simulation software; and generating and running the industrial simulation software based on the assembly results and user input or preset configuration parameters corresponding to the components; then training an AI agent model based on historical simulation data of multiple industrial simulation software; and receiving and parsing user natural language instructions, and generating result data corresponding to the user's natural language instructions based on the parsing results and the AI ​​agent model. In this way, not only can the industrial simulation software be generated conveniently and efficiently, reducing the difficulty and cost of industrial simulation software development, but also the computing power cost of actual simulation calculations can be greatly reduced through the use of the AI ​​agent model. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 It is a flowchart of the development and application method of low-code-based industrial simulation software provided by an embodiment of the present invention.

[0035] Figure 2 It is a schematic diagram of the principles of industrial simulation software development and application in the low-code-based industrial simulation software development and application method provided in an embodiment of the present invention.

[0036] Figure 3 It is a structural diagram of the development and application system of low-code-based industrial simulation software provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0037] To make the objectives, technical solutions, and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other implementations obtained by those of ordinary skill in the art without inventive effort are within the scope of protection of the present invention.

[0038] Figure 1 This is a flowchart of the development and application method of low-code industrial simulation software provided by an embodiment of the present invention. Please refer to Figure 1 , this embodiment may include the following steps:

[0039] S101. Carry out component packaging for multiple disciplines and generate an industrial simulation component library.

[0040] Specifically, based on modular packaging technology, at the basic physical field level, physical equations, professional application software and related content are packaged for multiple disciplines to generate components corresponding to disciplines such as multi-body dynamics, thermal analysis, fluid mechanics, structural mechanics and electromagnetics for subsequent use.

[0041] It should be noted that for each of the above subjects, the generated components can include form input components, input file components, pre-processing components, calculation running components, application software components, output file display components, post-processing components and output display components.

[0042] Among them, the form input component is used for users to input forms; the input file component is used for users to input files; the pre-processing component is used to process the data input by users; the calculation operation component is used to include professional application software or physical equations corresponding to the industrial simulation software; the application software component is used to apply the data of the industrial simulation software; the output file display component is used to display the files input by the industrial simulation software; the post-processing component is used to post-process the data output by the industrial simulation software; and the output display component is used to display the data output by the industrial simulation software.

[0043] As mentioned above, the computing operation component is used to include professional application software or physical equations corresponding to the industrial simulation software. In other embodiments of the present application, a device model with adjustable parameters can also be generated based on the device instance level. Users can generate a specific device model as a computing operation component by adjusting the parameters.

[0044] In actual applications, users can select the above components based on actual needs to choose appropriate professional application software or physical formulas, as well as appropriate input, pre-processing, application and display methods corresponding to the selected professional application software or physical formulas.

[0045] Furthermore, the present application can also pre-set parameterized templates for various professional application software at the domain component level, segmented by industry, as part of the corresponding input component. In actual applications, all or some parameters can be set based on actual conditions, allowing users to enter other parameters as needed, thereby reducing user operations while flexibly meeting various user needs.

[0046] On this basis, some embodiments of the present application also include: constructing semantic association relationships between components; when the user drags the target component, based on the semantic association relationship, recommending components corresponding to the target component to the user; or when the user drags the target component, based on the semantic association relationship, conflict detection is performed on multiple target components dragged by the user and a reminder is issued.

[0047] Specifically, based on the industrial ontology library, etc., semantic association relationships between various components of the above-mentioned disciplines can be constructed to achieve intelligent component recommendation and conflict detection.

[0048] S102: Based on the user's dragging instruction, assemble the target components in the industrial simulation component library to determine the functional nodes corresponding to the industrial simulation software.

[0049] Specifically, in this application, the simulation logic is broken down into reusable functional nodes such as parameter input, model solution, process output, and result output. In actual application, the user only needs to drag and drop components (the components dragged by the user are the target components selected by the user) to connect lines to define the data flow between nodes, thereby realizing graphical process orchestration for generation. It can be understood that in this process, the user can achieve node planning by dragging and dropping the target components.

[0050] S103 : Generate and run industrial simulation software based on the assembly result and user-input or preset configuration parameters corresponding to the components.

[0051] Specifically, based on the above steps, the specific operations to be performed by the industrial simulation software can be determined, such as how to process data (such as through a professional application software or physical equation) and flow, and parameter configuration can be performed. At this time, the releasable industrial simulation software, namely the simulation APP, is generated.

[0052] On this basis, data collection can be performed while the simulation app is running. In this application, this can be achieved through an embedded data probe. Specifically, a lightweight monitoring module can be inserted into the simulation code generated based on the above steps (the code of the developed and releasable simulation app) to collect key data in real time. At the same time, a time series database is used to store simulation input, output, and process data, and label them according to industry standards.

[0053] S104. Train an AI agent model based on historical simulation data from multiple industrial simulation software.

[0054] S105: Receive and parse the user's natural language instructions, and generate result data corresponding to the user's natural language instructions based on the parsing results and the AI ​​agent model.

[0055] Specifically, in the present application, the AI ​​agent model can be pre-trained using a large amount of historical simulation data from currently generated industrial simulation software, so that the AI ​​agent model has the ability to "memorize" the results of typical scenarios, so that the AI ​​agent model can be used to directly obtain the result data required by the user. In this way, the result data required by the user can be obtained directly without running the industrial simulation software, thereby realizing another calculation prediction technology.

[0056] Among them, a corresponding AI proxy model can be trained with data from an industrial simulation software to generate multiple AI proxy models trained by multiple industrial simulation software, thereby generating a proxy model library. Of course, in some embodiments, an AI proxy model can also be generated by industrial simulation software with similar field scenarios. In addition, in the above introduction, it is assumed that an industrial simulation software corresponds to an actual scenario. In actual applications, an industrial simulation software can also correspond to multiple actual scenarios. At this time, the AI ​​proxy model can also be trained based on historical simulation data distinguished by scenarios to obtain the corresponding AI proxy model.

[0057] Based on this, users can directly input natural language instructions, and the system automatically analyzes the user's needs and then dynamically selects the optimal AI proxy model based on the user's needs. In actual applications, the AI ​​proxy models in the proxy model library may not necessarily meet customer needs. Therefore, when the user's needs match the pre-training scenario, that is, when a suitable AI proxy model can be matched in the proxy model library, the AI ​​proxy model can be directly called to output the result data, thus skipping the numerical simulation calculation step. When the user's needs exceed the pre-training range, that is, no suitable AI proxy model can be matched in the proxy model library, then the result data is obtained through actual simulation calculation, and the AI ​​proxy model library is synchronously updated based on the data from this process.

[0058] The development and application method of low-code industrial simulation software provided in this application first performs component encapsulation for multiple disciplines to generate an industrial simulation component library; then, based on user drag and drop instructions, the target components in the industrial simulation component library are assembled to determine the functional nodes corresponding to the industrial simulation software; and based on the assembly results and the user input or preset configuration parameters corresponding to the components, the industrial simulation software is generated and run; then, an AI agent model is trained based on the historical simulation data of multiple industrial simulation software; and the user's natural language instructions are received and parsed, and based on the parsing results and the AI ​​agent model, result data corresponding to the user's natural language instructions is generated. In this way, not only can industrial simulation software be generated conveniently and efficiently, reducing the difficulty and cost of industrial simulation software development, but also the computing power cost of actual simulation calculations can be greatly reduced through the use of AI agent models.

[0059] Furthermore, in some embodiments of the present application, the above process mentioned generating and running industrial simulation software based on the assembly results and user input or preset configuration parameters corresponding to the components includes: determining the real-time computing power status of the cloud server and edge nodes; decomposing the simulation task into multiple subtasks, and distributing the multiple subtasks to the cloud server and edge nodes.

[0060] In practical applications, this can be achieved based on distributed task scheduling technology. Specifically, a dynamic load balancing algorithm is designed to break down simulation tasks into subtasks and execute them in a distributed manner based on the real-time computing power status of cloud servers and edge nodes. For example, complex subtasks are assigned to cloud servers, while simple subtasks are assigned to edge nodes. The lightweight edge runtime allows for the deployment of microservices frameworks on edge devices while supporting low-latency responses. Furthermore, cloud servers and edge nodes can interact with third-party design tools via RESTful APIs.

[0061] Furthermore, the present application also includes receiving and parsing user natural language instructions to obtain parsing results; based on the parsing results, generating a simulation example configuration file through a preset rule template engine; matching the simulation example configuration file with a simulation model corresponding to the generated industrial simulation software, and obtaining result data corresponding to the user's natural language instructions through the simulation model and the simulation example configuration file.

[0062] In the above process, the process of generating industrial simulation software based on components actually generates a simulation model corresponding to the industrial simulation software (including data through physical equations and professional application software). In this application, when the user enters a natural language instruction, a configuration file can be generated based on the natural language instruction, and the corresponding simulation model can be associated and matched, thereby generating result data corresponding to the user's natural language instruction.

[0063] Specifically, in this application, a natural language interaction engine can be pre-set for semantic parsing. In actual applications, based on a multimodal model, the user's natural language instructions are first parsed into structured query conditions, and then implicit parameters are completed based on the industrial knowledge graph.

[0064] On this basis, requirements are mapped to case studies. Specifically, based on a pre-set rule template engine, the results are mapped into simulation case configuration files, automatically associating them with the corresponding simulation models. If the user's natural language instructions do not fully correspond to the requirements, the system can call on historical data averages or trigger follow-up interactions to implement fuzzy condition processing.

[0065] Furthermore, in some embodiments of the present application, it also includes driving the evolution of the AI ​​agent model based on the data twin engine architecture.

[0066] Specifically, the data twin engine architecture is first constructed. This includes unified storage of the simulation input parameters, intermediate process data, and output results involved in the above process to support efficient queries. Furthermore, at the computational level, real-time statistics on the distribution of simulation data are collected. Furthermore, twin data is annotated, including automatic association with the version of the industrial simulation software, the physical model (i.e., the actual running model corresponding to the simulation software), and the AI ​​agent model version to ensure data traceability.

[0067] On this basis, a closed-loop model evolution is achieved, including a closed-loop simulation data-model training loop. Specifically, this involves running industrial simulation software to generate data for data storage; scheduled tasks filter high-quality data for training AI proxy models; and, on the other hand, the data from the newly trained AI proxy model can be deployed to the industrial simulation software to improve the efficiency and accuracy of subsequent simulations. In practical applications, this process can use high-precision simulation results as the teacher model and the AI ​​proxy model as the student model, transferring key features through the attention mechanism.

[0068] Figure 2 This is a schematic diagram of the principle of industrial simulation software development and application in the low-code-based industrial simulation software development and application method provided by an embodiment of the present invention, such as Figure 2 As shown, in this application, component construction is first performed to generate an industrial policy component library including form input components, input file components, pre-processing components, calculation and operation components, application software components, output file display components, post-processing components, and output display components; on this basis, users can drag and drop the above components to define simulation input parameters, define simulation pre-processing, define simulation calculations, define simulation post-processing, define simulation result display, and finally generate a publishable industrial simulation APP (i.e., industrial simulation software), etc., and realize the definition and release of industrial simulation APP through visual process construction technology; in the calling stage, simulation APP management capabilities such as APP permission allocation, user APP list, APP instance, simulation automation, simulation result extraction, and simulation result visualization can be performed; and AI agent model training is performed based on the data generated by the above call, including AI pre-training, generation of agent model knowledge base, determination of multimodal model, semantic mapping, zero-computation prediction, agent model result output, etc., thereby constructing a system for building, publishing and calling industrial policy APP based on low code.

[0069] The development and application method of low-code industrial simulation software provided in this application is

[0070] First, through the visual modeling engine, it provides a drag-and-drop process design and dynamic parameter configuration interface, which can realize code-free operation and reduce learning costs; users can describe their needs in colloquial language, and use AI agents to automatically parse and generate simulation examples without any programming or modeling knowledge, realizing natural language interaction and significantly lowering the development threshold. Furthermore, by automating the entire process from demand input, visual modeling, code generation, and cloud-edge deployment, the development cycle is shortened; using simulation results to automatically feed back to the AI ​​agent model, the parameters are dynamically adjusted through reinforcement learning optimization design, reducing the workload of manual parameter adjustment, achieving full-link automation, and improving development efficiency. Furthermore, an industry-specific component library is built, covering pre-packaged models in various fields; at the same time, knowledge graph constraints are used to implement industry-standard parameter verification rules to avoid illegal parameter configuration, achieve deep industry adaptation, and break through the limitations of universality. Furthermore, based on a cloud-edge-end collaborative architecture, a dynamic load balancing algorithm splits computing tasks between the cloud for high-precision simulation and the edge for real-time prediction, improving resource utilization. Furthermore, through AI proxy model zero-computation prediction, it enables direct invocation of pre-trained model output results for high-frequency scenarios, reducing redundant calculations, enabling efficient utilization of computing resources, and reducing costs. Furthermore, based on the data twin engine, industrial simulation apps automatically collect data during runtime and store it in a standardized database, continuously optimizing the AI ​​proxy model through learning. Furthermore, through dynamic knowledge distillation, high-precision simulation results are migrated to a lightweight proxy model, ensuring that the edge model accuracy is consistent with the cloud, achieving a data-driven continuous evolution.

[0071] Based on the same inventive concept, the present invention also provides a development and application device based on low-code industrial simulation software, which is used to implement the above method embodiment. Figure 3 This is a structural diagram of a low-code-based industrial simulation software development and application system provided by an embodiment of the present invention. Figure 3 As shown, the system includes:

[0072] The construction module 11 is used to package components for multiple disciplines and generate an industrial simulation component library.

[0073] The generation module 12 is used to assemble the target components in the industrial simulation component library based on the user's drag and drop instructions to determine the functional nodes corresponding to the industrial simulation software; and generate and run the industrial simulation software based on the assembly results and the user input or preset configuration parameters corresponding to the components.

[0074] The training module 13 is used to train the AI ​​agent model based on historical simulation data of multiple industrial simulation software.

[0075] The application module 14 is used to receive and parse the user's natural language instructions, and generate result data corresponding to the user's natural language instructions based on the parsing results and the AI ​​agent model.

[0076] Regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0077] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0078] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" is at least two.

[0079] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0080] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0081] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing module, or each unit may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or in the form of software functional modules. If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium.

[0082] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0083] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0084] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for developing and applying industrial simulation software based on low code, characterized in that: include: Carry out component packaging for multiple disciplines and generate an industrial simulation component library; Based on the user's drag and drop instructions, target components in the industrial simulation component library are assembled to determine the functional nodes corresponding to the industrial simulation software; generating and running the industrial simulation software based on the assembly results and user-input or preset configuration parameters corresponding to the components; Training an AI agent model based on historical simulation data of multiple industrial simulation software; Receive and parse user natural language instructions, and generate result data corresponding to the user natural language instructions based on the parsing results and the AI ​​agent model.

2. The method according to claim 1, characterized in that The multi-disciplines include multi-body dynamics, thermal analysis, fluid mechanics, structural mechanics and electromagnetics; the components include form input component, input file component, pre-processing component, calculation operation component, application software component, output file display component, post-processing component and output display component; Among them, the form input component is used for users to input forms; the input file component is used for users to input files; the pre-processing component is used to process the data input by users; the calculation operation component is used to include professional application software or physical equations corresponding to industrial simulation software; the application software component is used to apply the data of industrial simulation software; the output file display component is used to display the files input by industrial simulation software; the post-processing component is used to post-process the data output by industrial simulation software; and the output display component is used to display the data output by industrial simulation software.

3. The method according to claim 1, characterized in that Also includes: Build semantic associations between components; When the user drags the target component, based on the semantic association relationship, the component corresponding to the target component is recommended to the user; or when the user drags the target component, based on the semantic association relationship, conflict detection is performed on the multiple target components dragged by the user and a reminder is issued.

4. The method according to claim 1, wherein Also includes: By inserting a monitoring module into the code of the industrial simulation software, data is collected when the industrial simulation software is running and stored in a time series database.

5. The method according to claim 1, characterized in that The generating and running of the industrial simulation software based on the assembly result and the configuration parameters corresponding to the component input by the user or preset by the user includes: Determine the real-time computing power status of cloud servers and edge nodes; The simulation task is decomposed into multiple subtasks, and the subtasks are distributed to cloud servers and edge nodes.

6. The method according to claim 1, characterized in that The AI ​​agent model is trained based on historical simulation data of a plurality of industrial simulation software, including: Based on the historical simulation data of the plurality of industrial simulation software, a plurality of AI agent models are trained respectively, and an agent model library including the plurality of AI agent models is constructed.

7. The method according to claim 6, characterized in that Also includes: Receive and parse user natural language instructions to obtain parsing results; Based on the analysis results, a simulation case configuration file is generated through a preset rule template engine; A simulation model corresponding to the generated industrial simulation software is matched to the simulation example configuration file, and result data corresponding to the user's natural language instructions is generated through the simulation model and the simulation example configuration file.

8. A development and application system for industrial simulation software based on low code, characterized in that: include: Building modules for packaging components for multiple disciplines and generating industrial simulation component libraries; A generation module is configured to assemble target components in the industrial simulation component library based on user drag and drop instructions to determine functional nodes corresponding to industrial simulation software; and generate and run the industrial simulation software based on the assembly results and user-input or preset configuration parameters corresponding to the components; A training module, configured to train an AI agent model based on historical simulation data of a plurality of said industrial simulation software; The application module is used to receive and parse user natural language instructions, and generate result data corresponding to the user's natural language instructions based on the parsing results and the AI ​​agent model.

Images