Photoelectric system digital prototype function model generation method and device based on artificial intelligence

Abstract

The invention discloses a photoelectric system digital prototype function model generation method and device based on artificial intelligence, and belongs to the technical field of artificial intelligence and photoelectric system digital prototypes, and the method comprises the steps: inputting training data of an artificial intelligence model, carrying out semantic analysis, marking and key parameter extraction operation, and forming a knowledge graph; establishing a logic association and constraint relationship among the parameters; training an artificial intelligence model by using the analyzed training data, optimizing a preset general rule base of the artificial intelligence model, and constructing a photoelectric system modeling rule base and a photoelectric system artificial intelligence model; calling the trained artificial intelligence model of the photoelectric system, extracting a model demand, calling the modeling rule base of the photoelectric system, and generating a function model of the photoelectric system; and visually displaying the photoelectric system function model, modifying and updating the photoelectric system function model, and dynamically updating the optimization strategy. Different function model modeling languages can be selected according to actual system requirements, and an artificial intelligence algorithm is adjusted.

Description

Technical Field

[0001] This invention belongs to the field of artificial intelligence and optoelectronic system digital prototype technology, specifically relating to a method and apparatus for generating functional models of optoelectronic system digital prototypes based on artificial intelligence. Background Technology

[0002] Digital prototype models in the optoelectronic field involve requirement models, functional models, and performance models, resulting in high model complexity. They encompass a wide range of disciplines, including optics, mechanics, electricity, control, thermodynamics, and fluid mechanics. Traditional digital prototype modeling for optoelectronic systems faces the following technical bottlenecks:

[0003] (1) Low modeling efficiency: Traditional modeling relies on manual parsing of requirement documents, requiring repeated coding or the use of professional modeling tools (such as SysML, Modelica, etc.).

[0004] (2) High knowledge barriers: Engineers need to master knowledge of multiple disciplines and modeling languages, and cross-team collaboration costs are high;

[0005] (3) Poor reusability: Historical project data has not formed a standardized knowledge base, making it difficult for new projects to directly reuse experience.

[0006] Current model generation methods have not solved the problems of multidisciplinary data fusion, unstructured knowledge extraction and dynamic modeling requirements. Therefore, there is an urgent need for a multidisciplinary collaborative modeling method that integrates artificial intelligence. Summary of the Invention

[0007] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0008] A method for generating a digital prototype functional model of an optoelectronic system based on artificial intelligence, comprising:

[0009] Step 1, Multi-source data processing: Semantic parsing, labeling, and key parameter extraction are performed on the training data of the artificial intelligence model to form a knowledge graph; Topological relationships between data are identified and redundant parameters are automatically removed, and data is completed to establish logical connections and constraints between parameters;

[0010] Step 2, Model Training; The artificial intelligence model is trained using the training data analyzed in Step 1. The general rule base preset by the artificial intelligence model is optimized to construct the modeling rule base and artificial intelligence model of the optoelectronic system.

[0011] Step 3, Functional Model Generation: The trained photoelectric system artificial intelligence model is called from the knowledge base module. First, the user input text is parsed and the model requirements are extracted. Then, the photoelectric system modeling rule base is called to generate the photoelectric system functional model.

[0012] Step 4, Model Validation and Iterative Optimization: Visualize the generated optoelectronic system functional model, verify the optoelectronic system functional model through simulation, modify and update the optoelectronic system functional model, and dynamically update the optimization strategy based on the modification results.

[0013] An artificial intelligence-based digital prototype functional model generation device for optoelectronic systems includes:

[0014] The multi-source data processing module performs semantic parsing, labeling, and key parameter extraction on the training data of the artificial intelligence model to form a knowledge graph; it identifies the topological relationships between data and automatically removes redundant parameters, completes the data, and establishes logical connections and constraints between parameters.

[0015] The model training module uses the training data parsed by the multi-source data processing module to train the artificial intelligence model and optimizes the pre-set general rule base of the artificial intelligence model, thereby constructing a modeling rule base for the optoelectronic system and an artificial intelligence model for the optoelectronic system.

[0016] The functional model generation module calls the trained optoelectronic system artificial intelligence model from the knowledge base module. It first parses the text input by the user and extracts the model requirements, and then calls the optoelectronic system modeling rule base to generate the optoelectronic system functional model.

[0017] The model verification and iterative optimization module performs model verification and iterative optimization; it visualizes the generated optoelectronic system functional model, verifies the optoelectronic system functional model through simulation, modifies and updates the optoelectronic system functional model, and dynamically updates the optimization strategy based on the modification results.

[0018] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method for generating a digital prototype functional model of an artificial intelligence-based optoelectronic system.

[0019] A non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method for generating a digital prototype functional model of an artificial intelligence-based optoelectronic system.

[0020] The present invention has the following beneficial effects:

[0021] (1) This invention utilizes a large language model when modeling functional models, which improves the efficiency of the transformation from concept to model;

[0022] (2) This invention uses artificial intelligence algorithms to generate modeling rules for functional models that match functional models, reducing the requirements for the professional knowledge of modelers and improving the accuracy of modeling;

[0023] (3) This invention uses artificial intelligence algorithms to summarize a large number of existing project documents, establish a knowledge graph, and improve the efficiency of utilizing existing knowledge;

[0024] (4) The present invention can select different functional modeling languages ​​and adjust artificial intelligence algorithms according to actual system requirements, and has flexibility and scalability. Attached Figure Description

[0025] Figure 1 This is a structural block diagram of the digital prototype functional model generation device for an artificial intelligence-based optoelectronic system according to the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0027] This invention utilizes artificial intelligence methods, such as deep learning, to train an AI model using mature design documents, models, and data. This trains the AI ​​model to form an AI model of the optoelectronic system, containing a complete knowledge base specific to the optoelectronic system. When a new project requires a functional model of the optoelectronic system, the AI ​​model is used to automatically generate the functional model by defining model requirements and boundary conditions. The model is displayed through a visual interface, and constraints can be manually modified or updated via a dialog interface, enabling intelligent model updates and iterations to quickly generate the required digital prototype functional model of the optoelectronic system. This invention solves the problems of low efficiency and high cost in traditional digital prototype modeling. This invention provides an AI-based method for generating functional models of digital prototypes for optoelectronic systems. Figure 1 As shown, its framework mainly includes a multi-source data processing module, a model training module, a functional model generation module, a model verification and iteration module, and a knowledge base module. Through these modules, a functional model of the optoelectronic system digital prototype is generated, and a knowledge base is established. The main functions of each module are as follows:

[0028] Multi-source data processing module: The multi-source data processing module includes two main functions: input data and data parsing and processing. It is responsible for performing semantic parsing, labeling, and extracting key parameters from the training data of the input artificial intelligence (AI) model to form a knowledge graph; identifying the topological relationships between data and automatically removing redundant parameters; supplementing data through methods such as generative adversarial networks (GANs); and establishing logical connections and constraints between parameters.

[0029] Model training module: The training data parsed by the multi-source data processing module is used to train the artificial intelligence (AI) model and optimize its preset general rule base, thereby constructing a modeling rule base for optoelectronic systems and an artificial intelligence model for optoelectronic systems.

[0030] Functional Model Generation Module: This module includes functions for requirement input and parsing, model generation, and model output. It calls the trained optoelectronic system AI model from the knowledge base module, first parsing the text input by the user through dialog boxes, document uploads, etc., and extracting the model requirements. Then, it calls the optoelectronic system modeling rule base to generate the optoelectronic system functional model.

[0031] Model Validation and Iterative Optimization Module: This module mainly includes two functions: simulation validation and strategy optimization. It visualizes the generated optoelectronic system functional model, verifies the optoelectronic system functional model through simulation, modifies and updates the optoelectronic system functional model, and dynamically updates the optimization strategy based on the modification results.

[0032] Knowledge Base Module: This module is responsible for storing data such as knowledge graphs, artificial intelligence (AI) models, a dedicated modeling rule base for optoelectronic systems, functional models of optoelectronic systems, and optimization strategies. It performs operations such as reading, storing, and updating related data.

[0033] The present invention will be further described below with reference to specific implementation steps. The implementation of the present invention mainly includes four steps: multi-source data processing, model training, functional model generation, and model verification and iterative optimization.

[0034] Step 1, Multi-source data processing; Input the training data of the artificial intelligence model (AI), use the Large Language Model (LLM) for semantic parsing and labeling, and extract key parameters and construct a knowledge graph based on the Graph Neural Network (GNN); Perform semantic parsing, labeling, key parameter extraction and other operations on the training data of the input artificial intelligence model to form a knowledge graph; Identify the topological relationships between data and automatically remove redundant parameters, supplement the data through methods such as the Generative Adversarial Network (GAN), and establish logical connections and constraints between parameters.

[0035] This step corresponds to the function of the multi-source data processing module. The input data consists of completed, mature project documents, such as design (scheme) documents, test reports, and simulation data. In this step, data from various documents is parsed, labeled, integrated, and correlated. It is necessary to ensure semantic consistency within the same document and between different documents to improve parsing accuracy. By using a large language model to assist in document semantic analysis—for example, automatically identifying related but semantically inconsistent entries and generating prompts for manual confirmation, modification, or manual labeling and correlation of data—the process can be improved.

[0036] In one embodiment, the multi-source data processing module acquired design documents from five completed and archived LiDAR projects as training data for the artificial intelligence model. These documents primarily included requirement specifications, design schemes, and test reports. Document types included PDF (Portable Document Format), tables, and images. They contained complete content covering the entire project lifecycle, including project requirement specifications, decomposition, functional design, parameter allocation, system composition, system parameter traceability, and system test data.

[0037] In this embodiment, the system requirement constraints extracted by the Large Language Model (LLM) in the multi-source data processing module include weight constraints and volume constraints, while the system functions include target ranging and target detection. Parameter indicators include system weight, wavelength, surface accuracy, and detection distance. Data from different documents is associated and labeled to form a data chain of requirements-functions-parameters-design-testing. Furthermore, multidisciplinary design parameters, such as optical component properties, mechanical structure parameters, and electrical performance parameters, are extracted from multidisciplinary design technical reports and test data. The complete project, after being organized, is stored in the knowledge base module in the form of a knowledge graph. For coupled designs, a knowledge graph of interdisciplinary relationships is constructed using methods such as Graph Neural Networks (GNNs). For example, the project involves multidisciplinary coupled designs of "optics-mechanics-thermal." The artificial intelligence model learns the inherent coupling relationships within these couplings and establishes a coupling weight parameter table. Specifically, when establishing the multi-science coupled design relationship of "optics-mechanics-thermal," the artificial intelligence model extracts the coupling design between temperature, structure, and optics through test data. Specifically, the temperature field causes changes in the optical support structure, which in turn causes changes in the optical surface shape, ultimately leading to changes in optical properties. When constructing the knowledge graph using a graph neural network (GNN), temperature attributes, mechanical structure parameters, and optical performance parameters are defined as nodes, and the physical coupling relationships between nodes (such as temperature changes causing structural deformation) are defined as edges. A message-passing-based GNN architecture, such as GraphSAGE (an inductive graph representation learning algorithm), is adopted. The model is trained using historical test data to learn the coupling weight parameter table. Specifically, the training data includes temperature field distribution and measured values ​​of optical surface shape changes. The GNN outputs coupling weights through multi-layer aggregation operations, such as a 1°C increase in temperature causing a 0.1mm deviation in optical surface shape.

[0038] Step 2, Model Training: Use the training data analyzed in Step 1 to train the artificial intelligence (AI) model and optimize its preset general rule base to build a rule base for modeling the optoelectronic system and an artificial intelligence model for the optoelectronic system.

[0039] This step is implemented by the model training module, performing deep training on the pre-trained model. The pre-trained model is a general artificial intelligence (AI) model containing a basic functional model framework. This step reads the knowledge graph stored in the knowledge base module in step 1, and performs deep training on the pre-trained model based on deep learning algorithms such as VGG (Visual Geometry Group) and YOLO (You Only Look Once, a general term for a class of single-shot object detection deep learning models), optimizing its preset general rule base, thereby constructing a rule base for modeling the optoelectronic system and an AI model for the optoelectronic system. For example, if the training data includes design drawings or mechanical sketches, the VGG algorithm is used to extract features from the images (such as the contour features of optical lenses), and the feature vectors are fused with text parameters in the knowledge graph. These parameters are converted into structured data and incorporated into the corresponding nodes of the knowledge graph as input to the deep learning model. During training, image feature vectors and text feature vectors are concatenated and jointly input into the fully connected layer to optimize the rule base, optimizing the basic and derived layers of the general rule base.

[0040] In one embodiment, a knowledge graph of five complete projects formed by the multi-source data processing module is used as training data for the model. From a pre-defined set of deep learning algorithms, an algorithm is selected to begin training the artificial intelligence model; in this embodiment, the VGG algorithm is used. During training, the general rule base is optimized, including optimizations of the basic rule layer and the derived rule layer, thereby constructing a rule base for modeling the optoelectronic system and storing it in the knowledge base module.

[0041] In this embodiment, at the basic rule layer optimization, rules related to electrical safety, mechanical equipment safety, data specifications, and working scenarios are optimized for the optoelectronic system. At the derived rule layer optimization, the artificial intelligence model optimizes the modeling rules for the optoelectronic system's functional model (e.g., the mapping relationship between system requirements and functional division, performance parameters), the allocation relationship between system functional division and module definitions, and modeling rules for module division, module interface definitions, and parameter definitions. Taking the allocation relationship between system functional division and module definitions as an example, the allocation relationship between optoelectronic system detection functions and detection modules, and the allocation relationship between tracking functions and rack modules are established through training, constructing an optoelectronic system modeling rule base, which then forms the optoelectronic system's artificial intelligence model and is stored in the knowledge base module.

[0042] Step 3, Functional Model Generation; This includes three steps: requirement input and parsing, model generation, and model output. The trained optoelectronic system artificial intelligence model is called from the knowledge base module. First, the text input by the user through dialog boxes, document uploads, etc., is parsed and the model requirements are extracted. Then, the optoelectronic system modeling rule base is called to generate the optoelectronic system functional model.

[0043] Specifically, the process begins with creating a new project and inputting model requirements. These requirements can be submitted in various formats, including uploaded documents and images, and the optoelectronic system AI model is then used to parse them. Next, the optoelectronic system functional model is generated. This generation can be implemented in two modes: fully automatic and step-by-step. The fully automatic generation mode is suitable for situations where the requirements are relatively vague. Based on a single input requirement, it directly generates the optoelectronic system functional model using the optoelectronic system modeling rule library, requiring no additional steps. When the requirements are vague, the fully automatic generation mode uses a Generative Adversarial Network (GAN) to fill in the missing information: the generator generates candidate parameters based on the input requirements, and the discriminator verifies the authenticity of the parameters; simultaneously, rule optimization uses reinforcement learning (such as Q-learning), adjusting weights based on historical model feedback. The analysis process of the optoelectronic system AI model can be viewed through model logs and message prompts. The step-by-step generation mode is implemented through a dialogue: the optoelectronic system modeling rule library is called, using a fixed model template. Requirements are input step-by-step through a dialog box, generating a partial model and providing feedback. Then, the parameters needed for the next step are input. After all dialogues are completed, the optoelectronic system modeling rule library is called to integrate the models and generate the optoelectronic system functional model. The final output is a functional model of the optoelectronic system, supporting standard modeling languages ​​such as SysML (System Modeling Language) and UML (Unified Modeling Language); it calls the artificial intelligence model of the optoelectronic system to generate a functional model file of the optoelectronic system. The file supports common formats such as XML (Extensible Markup Language) and can be imported into mainstream MBSE (Model-Based Systems Engineering) tools such as CATIAMagic (a model-based systems engineering solution software platform) and IBM Rhapsody (a model-based systems engineering modeling tool).

[0044] In one embodiment, the project type was first identified as a LiDAR project through a text-based dialogue, and the corresponding modeling rule base was matched from the knowledge base module. Then, a fully automated generation mode was used to write the design requirements, such as "detection range greater than 1km," "weight not exceeding 200kg," and "two-axis structure," into a document and upload it. An artificial intelligence model was used to parse the document and establish design constraints. Finally, SysML was selected as the design language, and the model was output in "MDZIP" format using CATIA Magic software, generating the functional model of the LiDAR project.

[0045] Step 4, Model Validation and Iterative Optimization; mainly includes simulation validation and strategy optimization; the generated optoelectronic system functional model is visualized, and the optoelectronic system functional model is validated through simulation, modified and updated, and the optimization strategy is dynamically updated based on the modification results.

[0046] The optoelectronic system functional model generated in step 3 can be viewed, edited, simulated, and exported. Modifications to the optoelectronic system functional model in this step provide feedback to the artificial intelligence model. The AI ​​model dynamically updates its optimization strategy based on the modifications. After model modification, a policy gradient algorithm is used to adjust the rule base weights based on simulation results. For example, if the user frequently adds 'standby state', the priority of the standby state in the state machine rules is increased, thereby achieving intelligent iterative optimization of the optimization strategy.

[0047] In one embodiment, the preview function of the model verification and iterative optimization module is used to view the functional model of the optoelectronic system. The generated functional model includes a requirement model, system architecture model, functional principle model, system parameter model, system interface model, etc. The model views include requirement diagrams, module definition diagrams, internal module diagrams, activity diagrams, and state machine diagrams. By viewing the module division, interface definition, function definition, requirement traceability, etc., for model views that need to be modified, editing can be done in the view or modification requests can be made to the artificial intelligence model through dialogue.

[0048] In this embodiment, the initial state machine diagram of the system workflow is defined as: Off - Power On - Self-Test - Working. A standby state is added to the self-test and working states via instructions, enabling dynamic modification of the optoelectronic system functional model and synchronous updating of modeling rules. The modified optoelectronic system functional model is dynamically updated in the preview interface, while the optimization strategies in the knowledge base module are also updated. Simulations are established to dynamically display the system's requirement tracing, indicator decomposition, function allocation, component architecture, data communication, function implementation process, and state parameter changes. In this embodiment, simulation of the system state machine diagram verifies the state changes and system activity triggering when the system receives signals such as "Power On" and "Self-Test". Upon receiving the "Power On" signal, the system executes the "Power On" process and then automatically transitions from the "Off" state to the "Power On" state, verifying the accuracy of the model's description of system state changes.

[0049] In addition, such as Figure 1 As shown, the digital prototype functional model in this invention, as part of the optoelectronic system digital prototype system, together with other models (such as the digital prototype geometric model, digital prototype performance model, etc.), constitutes a complete optoelectronic system digital prototype system.

[0050] In summary, this invention addresses the problems of low modeling efficiency, reliance on specialized modeling language knowledge, and lack of standardized references in existing technologies. It significantly lowers the modeling threshold, shortens the R&D cycle, and is suitable for rapid prototyping of functional models of complex optoelectronic systems. The artificial intelligence training methods, modeling software, and modeling languages ​​listed in this solution are merely examples illustrating the technical solution. Modeling, verification, and iteration of other functional modeling languages, such as UML (Unified Modeling Language), and other extended disciplines implemented by modeling software, are all considered within the scope of protection of this invention.

[0051] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0052] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0053] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0054] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0055] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

[0056] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

[0057] The above description is merely an embodiment of the present invention and does not limit the scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related system fields, are similarly included within the protection scope of the present invention.

[0058] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

Claims

1. An artificial intelligence-based optoelectronic system digital prototype function model generation method, characterized by, Comprise: Step 1, multi-source data processing; The training data of the input artificial intelligence model is parsed and labeled using a large language model, and key parameters are extracted based on a graph neural network to construct a knowledge graph; the training data of the input artificial intelligence model is parsed, labeled, and key parameters are extracted to form a knowledge graph; the topological relationship between the data is identified and redundant parameters are automatically removed, and the data is completed to establish the logical association and constraint relationship between the parameters; Step 2, model training; training the artificial intelligence model using the training data parsed in step 1, optimizing the general rule library preset for the artificial intelligence model, thereby constructing the photoelectric system modeling rule library and the photoelectric system artificial intelligence model; Step 3, function model generation; calling the trained photoelectric system artificial intelligence model from the knowledge base module, first parsing the user input text and extracting model requirements, then calling the photoelectric system modeling rule library to generate a photoelectric system function model; Step 4, model verification and iterative optimization; Visualize the generated photoelectric system function model and verify the photoelectric system function model through simulation, modify and update the photoelectric system function model, and dynamically update the optimization strategy according to the modification results. 2.The artificial intelligence-based optoelectronic system digital mockup function model generating method of claim 1, wherein, In step 1, the system requirement constraints extracted by the large language model; multi-disciplinary design parameters are extracted from multi-disciplinary design technical reports and test data, and the complete project is stored in the knowledge base module in the form of a knowledge graph after being sorted. 3.The AI-based optoelectronic system digital mockup function model generating method of claim 2, wherein, Step 2 includes: reading the knowledge graph stored in the knowledge base module in step 1, based on a deep learning algorithm, deeply training a pre-trained model, optimizing the general rule library preset for the artificial intelligence model, thereby constructing the photoelectric system modeling rule library and the photoelectric system artificial intelligence model.

4. The artificial intelligence based optoelectronic system digital mockup function model generating method of claim 1, wherein, In step 2, the deep learning algorithm is VGG or YOLO; Optimizing the general rule library preset for the artificial intelligence model includes optimizing the basic rule layer and the derived rule layer; in the optimization of the basic rule layer, the electrical safety, mechanical equipment safety, data specification and work scene rules of the photoelectric system are optimized; in the optimization of the derived rule layer, the artificial intelligence model optimizes the photoelectric system function model modeling rule, the allocation relationship of system function division and module definition, the modeling rule of module division, interface definition and parameter definition of the module. 5.The artificial intelligence based optoelectronic system digital mockup function model generating method according to claim 1, wherein, Step 3 includes: First, create a new project, input the model requirements, and call the photoelectric system artificial intelligence model to parse the input model requirements; then generate the photoelectric system function model in full-automatic generation mode or step-by-step generation mode; among them, when the demand is more fuzzy, the photoelectric system function model is generated in full-automatic generation mode: according to the one-time input demand, the photoelectric system modeling rule library is called to directly generate the photoelectric system function model, when the demand is fuzzy, the full-automatic generation mode uses the generative adversarial network to complete the missing information: the generator generates candidate parameters according to the input demand, and the discriminator verifies the authenticity of the parameters; at the same time, the rule optimization adopts reinforcement learning to adjust the weight according to the historical model feedback; The step-by-step generation mode is realized through a dialogue: the photoelectric system modeling rule base is called, a fixed model template is used, the requirements are input step by step through a dialogue box, part of the model is generated and fed back, then the required parameters of the next step are input, the dialogue is completed, the photoelectric system modeling rule base is called to integrate the model, and the photoelectric system function model is generated; finally, the photoelectric system function model is output.

6. The artificial intelligence based optoelectronic system digital mockup function model generating method of claim 1, wherein, Step 4 includes: the modification of the photoelectric system function model forms feedback to the artificial intelligence model; the artificial intelligence model dynamically updates the optimization strategy according to the modification result, and realizes intelligent iterative optimization of the optimization strategy.

7. The artificial intelligence based optoelectronic system digital mockup function model generating method of claim 1, wherein, Step 4 includes: the initial definition of the state machine diagram of the system workflow is: off-on-self-check-work; through instructions, the standby state is added in the self-check and work states, the dynamic modification of the photoelectric system function model is completed, the modeling rules are updated synchronously; the modified photoelectric system function model is dynamically updated in the preview interface, and the optimization strategy in the knowledge base module is updated.

8. An artificial intelligence-based optoelectronic system digital mockup function model generation apparatus, characterized by, It includes: A multi-source data processing module that inputs training data of an artificial intelligence model to perform semantic analysis, marking, and key parameter extraction operations to form a knowledge graph; identifies the topological relationship between data and automatically removes redundant parameters, and completes the data to establish logical association and constraint relationship between parameters; A model training module that trains the artificial intelligence model using the training data analyzed by the multi-source data processing module, optimizes the pre-set general rule base of the artificial intelligence model, thereby constructing the photoelectric system modeling rule base and the photoelectric system artificial intelligence model; A function model generation module that calls the trained photoelectric system artificial intelligence model from the knowledge base module, first analyzes the user input text and extracts model requirements, then calls the photoelectric system modeling rule base to generate the photoelectric system function model; A model verification and iterative optimization module that visually displays the generated photoelectric system function model, verifies the photoelectric system function model through simulation, modifies and updates the photoelectric system function model, and dynamically updates the optimization strategy according to the modification result.

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to realize the steps of the artificial intelligence-based photoelectric system digital prototype function model generation method according to any one of claims 1-7.

10. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the artificial intelligence-based photoelectric system digital prototype function model generation method according to any one of claims 1-7.

Images