Intelligent Design Interaction System Based on Large Language Models

Abstract

The present invention provides an intelligent design interaction system based on a large model, which relates to the technical field of electrical design. By utilizing historical lighting distribution design drawing data, the present invention constructs a scenario template database and a parameter allocation model, realizing the efficient automation of the design process, supporting users to input design requirements through an interactive manner, and customizing the generation of personalized distribution parameters to meet the specific requirements of different users and projects. Secondly, by deeply mining users' implicit requirements through a questionnaire survey and dynamically adjusting the matching threshold, the present invention can accurately capture users' preferences and make personalized adjustments to the scenario templates, thus highly fitting users' emotional and aesthetic demands while meeting the functional requirements. Moreover, the present invention can also screen out scenario templates that highly match users' requirements and achieve the optimal design parameters. In addition, the parameter allocation model obtained through machine learning model training outputs optimized distribution parameters, improving the quality and reliability of the design.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical design, and particularly to an intelligent design interaction system based on a large model. Background Art

[0002] In the field of lighting distribution design, with the continuous increase in the complexity and scale of construction projects, traditional design methods have been difficult to meet the current requirements. For example, traditional design methods require engineers to manually calculate and analyze a large amount of data. Although the design effect is outstanding, it also results in a long design cycle and cannot meet the rapidly changing market demands. At the same time, calculations are prone to errors, increasing the design cost and time cost. Due to the different experiences and levels of engineers, traditional design methods are difficult to ensure the consistency of design quality, are prone to errors, and bring difficulties to subsequent construction and maintenance. Moreover, with the continuous increase in the diverse and personalized requirements of construction projects, traditional design methods are difficult to meet the specific requirements of different users and projects. Engineers need to carry out customized designs for each project, which further increases the design difficulty and time cost.

[0003] Therefore, it is necessary to provide an intelligent design interaction system based on a large model to solve the above technical problems. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides an intelligent design interaction system based on a large model to solve the problems of low design efficiency, unstable design quality, and difficult-to-meet personalized requirements of traditional design methods.

[0005] The intelligent design interaction system based on a large model provided by the present invention includes:

[0006] A data processing module, configured to obtain historical lighting distribution design drawing data for preprocessing, and identify scene features in the lighting distribution design drawing, and construct a scene template database with the scene features;

[0007] A model construction module, configured to extract key feature parameters from historical lighting distribution design drawing data, use the key feature parameters as inputs, and use the distribution parameters of the lighting distribution design drawing as outputs to train a machine learning model to obtain a parameter allocation model;

[0008] An information screening module, configured to input design requirements through an interactive manner, extract scene information in the design requirements, and screen out all scene templates with a matching degree exceeding a matching threshold from the scene template database according to the scene information;

[0009] A dynamic adjustment module, configured to analyze the implicit preference emotions of users based on a questionnaire previously answered by the users to dynamically adjust the matching threshold, perform secondary screening and adjustment on the screened scene templates to obtain adjusted scene templates;

[0010] A data analysis module, which is used to extract key feature parameters through the input design requirements and input them into the parameter allocation model, and output the power distribution parameters of the current design;

[0011] An identification and generation module, which is used to supplement the power distribution parameters of the current design into the adjusted scene template, analyze its adaptability, and screen out the scene template with the highest adaptability as the lighting power distribution design drawing of the current design for display.

[0012] Preferably, the steps for preprocessing the obtained historical lighting power distribution design drawing data, identifying the scene features in the lighting power distribution design drawing, and constructing a scene template database with the scene features are as follows:

[0013] Collect historical lighting power distribution design drawing data from different projects and regions and preprocess it, where the preprocessing includes image clarification and format unification;

[0014] Use computer vision technology to identify the scene features in the historical lighting power distribution design drawing data, where the scene features include room type, lamp layout, and switch position;

[0015] Take the identified scene features and their corresponding complete historical lighting power distribution design drawings as entries and store them in the scene template database.

[0016] Preferably, the steps for extracting key feature parameters from the historical lighting power distribution design drawing data, using the key feature parameters as the input, and the power distribution parameters of the lighting power distribution design drawing as the output to train a machine learning model to obtain a parameter allocation model are as follows:

[0017] Extract key feature parameters from the historical lighting power distribution design drawing, specifically including lamp type, single lamp power, and lighting level;

[0018] Use the key feature parameters as input features and the power distribution parameters of the lighting power distribution design drawing as output parameters, and mark the mapping relationship between the input features and the output parameters, where the power distribution parameters include line layout, circuit breaker specifications, and cable specifications;

[0019] Adopt a machine learning model to train the machine learning model according to the marked mapping relationship between the input features and the output parameters to obtain a parameter allocation model.

[0020] Preferably, the steps for inputting design requirements through an interactive method, extracting the scene information in the design requirements, and screening out all scene templates in the scene template database whose matching degree exceeds the matching threshold according to the scene information are as follows:

[0021] The user inputs design requirements interactively, including room type, lighting style, energy efficiency requirements, etc. Among them, the interactive method includes text box input or integrating a speech recognition module to convert speech into text;

[0022] Extract the scene information in the design requirements and compare it with the entries in the scene template database to filter out all scene templates whose matching degree exceeds the preset matching threshold.

[0023] Preferably, based on the questionnaire replied by the user in advance, analyze the user's implicit preference emotions to dynamically adjust the matching threshold, and perform a secondary screening and adjustment on the filtered scene templates to obtain the adjusted scene templates. The specific steps are as follows:

[0024] Set up the questionnaire structure and send it to the user terminal. Among them, the questionnaire structure includes explicit requirement collection and implicit requirement collection;

[0025] Collect the reply results of the user to the questionnaire, and use a natural language processing model to extract the sentiment polarity and keywords from the reply results;

[0026] Based on the extracted sentiment polarity and keywords, adjust the matching threshold corresponding to the features of the sentiment polarity and keywords to obtain the adjusted matching threshold;

[0027] According to the adjusted matching threshold, perform a secondary screening on the filtered scene templates, and based on the features corresponding to the sentiment polarity and keywords, adjust the scene templates after the secondary screening.

[0028] Preferably, supplement the currently designed power distribution parameters to all the filtered scene templates, analyze their adaptability, and filter out the scene template with the highest adaptability as the lighting power distribution design drawing of the current design for display. The specific steps are as follows:

[0029] Supplement the generated power distribution parameters to the adjusted scene templates to generate multiple candidate templates;

[0030] Through the preset evaluation conditions, evaluate the adaptability of each candidate template after adding the new power distribution parameters respectively, and sort the multiple candidate templates according to the size of the adaptability to obtain the sorting result;

[0031] Based on the sorting result, filter out the scene template with the highest adaptability as the optimal template;

[0032] Use the optimal template as the lighting power distribution design drawing of the current design and display it to feedback to the user.

[0033] Compared with the related technologies, the intelligent design interaction system based on the large model provided by the present invention has the following beneficial effects:

[0034] By utilizing the historical lighting power distribution design drawing data, the present invention constructs a scene template database and a parameter allocation model, realizing efficient automation of the design process, greatly shortening the design cycle and improving the design efficiency. At the same time, it supports users to input design requirements interactively, customize and generate personalized power distribution parameters to meet the specific needs of different users and projects. Secondly, by deeply mining users' implicit needs through questionnaires and dynamically adjusting the matching threshold, it can accurately capture users' preferences and make personalized adjustments to the scene templates, so as to highly fit users' emotional and aesthetic demands while meeting the functional requirements. Moreover, the present invention can also screen out scene templates that highly match users' needs and achieve the optimal design parameters. In addition, the parameter allocation model obtained through machine learning model training outputs optimized power distribution parameters, improving the quality and reliability of the design. Brief Description of the Drawings

[0035] Figure 1 It is a system block diagram of an intelligent design interaction system based on a large model in an embodiment of the present invention. Detailed Embodiment

[0036] The present invention will be further described below in conjunction with the drawings and embodiments.

[0037] The intelligent design interaction system based on a large model specifically includes:

[0038] A data processing module for obtaining historical lighting power distribution design drawing data for preprocessing and identifying scene features in the lighting power distribution design drawing, and constructing a scene template database with the scene features.

[0039] A model construction module for extracting key feature parameters from the historical lighting power distribution design drawing data, training a machine learning model with the key feature parameters as input and the power distribution parameters of the lighting power distribution design drawing as output, and obtaining a parameter allocation model.

[0040] An information screening module for inputting design requirements interactively, extracting scene information from the design requirements, and screening out all scene templates with a matching degree exceeding the matching threshold from the scene template database according to the scene information.

[0041] A data analysis module for extracting key feature parameters from the input design requirements and inputting them into the parameter allocation model to output the power distribution parameters of the current design.

[0042] An identification and generation module for supplementing the power distribution parameters of the current design to all the screened scene templates, analyzing their adaptability, and screening out the scene template with the highest adaptability as the lighting power distribution design drawing of the current design for display.

[0043] A feedback adjustment module, which is used to generate a final lighting power distribution design drawing according to the displayed lighting power distribution design drawing. The user can adjust the parameters interactively and generate the final lighting power distribution design drawing after the user confirms.

[0044] In the specific implementation process, the specific steps of obtaining historical lighting power distribution design drawing data for preprocessing and identifying the scene features in the lighting power distribution design drawing, and constructing a scene template database with the scene features are as follows:

[0045] Collect historical lighting power distribution design drawing data from different projects and regions and preprocess it. Among them, the preprocessing includes image clarification processing and format unification.

[0046] Specifically, collect historical lighting power distribution design drawing data from different projects and regions. Among them, these historical lighting power distribution design drawing data can come from publicly available data channels such as architectural design companies, power companies or public databases. Then, use denoising and contrast enhancement in existing image processing technologies to improve the clarity of the lighting power distribution design drawing for subsequent scene feature recognition. And convert the collected design drawings into a unified file format (JPEG, PNG or PDF), and adjust the image size or resolution to ensure consistency.

[0047] Use computer vision technology to identify the scene features in the historical lighting power distribution design drawing data. Among them, the scene features include room type, lamp layout and switch position.

[0048] Specifically, use existing computer vision technology, such as traditional image processing algorithms, to identify the scene features in the design drawing. Specifically: identify the room type (such as office, meeting room, classroom, etc.) by analyzing the layout, furniture and decorative elements in the design drawing; identify the position, type and quantity of the lamps in the design drawing, and how they are distributed in the room; identify the position of the switches in the design drawing, including wall switches, floor sockets, etc.

[0049] Take the identified scene features and their corresponding complete historical lighting power distribution design drawing as entries and store them in the scene template database.

[0050] Specifically, organize the identified scene features and their corresponding complete historical lighting power distribution design drawing as entries, and use a database management system (MySQL, MongoDB) to store these entries to construct a scene template database.

[0051] In the specific implementation process, the specific steps of extracting key feature parameters from the historical lighting power distribution design drawing data, using the key feature parameters as input and the power distribution parameters of the lighting power distribution design drawing as output to train a machine learning model to obtain a parameter allocation model are as follows:

[0052] Extract key feature parameters from historical lighting power distribution design drawings, specifically including lamp types, single-lamp power, and lighting levels.

[0053] Specifically, to extract key feature parameters from historical lighting power distribution design drawings, it is as follows: Identify the lamp types used in the lighting power distribution design drawings, such as fluorescent lamps, LED lamps, halogen lamps, etc.; obtain the single-lamp power of each lamp by measurement or referring to the lamp specification table; according to the lighting layout and the number of lamps in the design drawing, and in accordance with existing lighting standards, identify the lighting level, such as standard lighting or energy-saving lighting.

[0054] Using the key feature parameters as input features and the power distribution parameters of the lighting power distribution design drawing as output parameters, and mark the mapping relationship between the input features and the output parameters. Among them, the power distribution parameters include line layout, circuit breaker specifications, and cable specifications.

[0055] Specifically, input features: lamp type, single-lamp power, lighting level; output parameters: line layout, circuit breaker specifications, cable specifications; for each historical lighting power distribution design drawing, according to its actual power distribution parameters, mark the corresponding mapping relationship between the input features and the output parameters, and ensure the accuracy and integrity of the marked data for subsequent training of the machine learning model.

[0056] Exemplarily, for an office lighting power distribution design drawing, its corresponding power distribution parameters include: the line layout is a branch circuit, the circuit breaker specification is 16A, and the cable specification is 2.5mm². Mark these power distribution parameters with the extracted key feature parameters to form a dataset for training the machine learning model.

[0057] Adopt a machine learning model to train the machine learning model according to the marked mapping relationship between the input features and the output parameters to obtain a parameter allocation model.

[0058] Specifically, select the neural network in the machine learning model, use the marked mapping relationship between the input features and the output parameters as the training dataset to train the machine learning model so that it can learn the mapping relationship between the input features and the output parameters. After training, obtain the parameter allocation model.

[0059] In the specific implementation process, the specific steps to input the design requirements through an interactive method, extract the scenario information in the design requirements, and screen out all scenario templates with a matching degree exceeding the matching threshold from the scenario template database are as follows:

[0060] The user inputs the design requirements through an interactive method, including room type, lighting style, energy efficiency requirements, etc. Among them, the interactive method includes text box input or integrating a voice recognition module to convert voice into text.

[0061] Specifically, the user connects to the system through a web interface on a terminal device (computer, tablet, or mobile phone) and inputs text in a text box to describe the design requirements, or the voice recognition module recognizes the text to form the input design requirements. Among them, the design requirements include room type (such as office, meeting room, classroom, etc.), lighting style (such as modern, simple, retro, etc.), energy efficiency requirements (such as energy saving, high efficiency, etc.), and other relevant parameters.

[0062] Extract the scene information from the design requirements and compare it with the entries in the scene template database to filter out all scene templates whose matching degree exceeds the preset matching threshold.

[0063] Specifically, the system extracts scene information from the design requirements input by the user, such as room type, lighting style, and energy efficiency requirements. Then, each scene template is pre-tagged in advance, and the extracted scene information is compared with the entries in the scene template database. The label coincidence degree is calculated as the matching degree, where the matching degree = the number of coincident labels / the total number of template labels. After calculating the matching degrees of all scene templates, filter out all scene templates whose matching degree exceeds the preset matching threshold.

[0064] In the specific implementation process, based on the questionnaire previously replied by the user, analyze the user's implicit preference emotions to dynamically adjust the matching threshold, and perform a secondary screening and adjustment on the filtered scene templates to obtain the adjusted scene templates. The specific steps are as follows:

[0065] Set the questionnaire structure and send it to the user terminal. Among them, the questionnaire structure includes explicit requirement collection and implicit requirement collection.

[0066] Specifically, set the questionnaire structure. Exemplarily, set the explicit requirement collection: such as what is the lighting brightness level expected by the user, whether intelligent control function is required, etc.; set the implicit requirement collection: such as in an ideal working environment, how should the lighting effect affect the user's mood or work efficiency, whether it is hoped that the lighting design can reflect a certain specific style or atmosphere. Send the designed questionnaire to the user terminal through methods such as email, online form, mobile application push, etc.

[0067] Collect the reply results of the user to the questionnaire, and use a natural language processing model to extract the sentiment polarity and keywords from the reply results.

[0068] Specifically, the user submits his reply through the questionnaire, then collect and organize these replies, and use a natural language processing (NLP) model to analyze the replies to extract the sentiment polarity (such as positive, negative, neutral) and keywords (such as warm, energy saving, intelligent) in the replies. Exemplarily, if the user hopes that the lighting in the living room is both warm and energy saving, and preferably intelligent control, then the extraction result: sentiment polarity = positive, keywords = warm, energy saving, intelligent control.

[0069] Based on the extracted sentiment polarity and keywords, adjust the matching thresholds for the corresponding features of the sentiment polarity and keywords to obtain the adjusted matching thresholds.

[0070] Specifically, according to the extracted sentiment polarity and keywords, analyze the user's degree of emphasis on different features, and adjust the thresholds during scenario template matching so that scenario templates that better match the user's preferences are more likely to be screened out. Exemplarily, if the user mentions "warm" multiple times and the sentiment polarity is positive, then lower the matching thresholds for scenario templates that do not match the "warm" feature, and at the same time increase the matching thresholds for scenario templates that highly match the "warm" feature.

[0071] Perform a secondary screening on the screened scenario templates according to the adjusted matching thresholds, and adjust the scenario templates after the secondary screening based on the corresponding features of the sentiment polarity and keywords.

[0072] Specifically, use the adjusted matching thresholds to perform another screening on the scenario templates initially screened out, retain the scenario templates with a matching degree exceeding the new threshold, and adjust the scenario templates after the secondary screening according to the sentiment polarity and keywords in the user's reply. Exemplarily, based on the extracted sentiment polarity and keywords, increase the types of lamps that the user potentially prefers and adjust the lighting layout to conform to the atmosphere expected by the user.

[0073] In the specific implementation process, extract the key feature parameters through the input design requirements and input them into the parameter allocation model to output the power distribution parameters of the current design.

[0074] In the specific implementation process, supplement the power distribution parameters of the current design into the adjusted scenario templates, and analyze their adaptability. The specific steps for screening out the scenario template with the highest adaptability as the lighting power distribution design drawing for the current design to display are as follows:

[0075] Supplement the generated power distribution parameters into the adjusted scenario templates to generate multiple candidate templates.

[0076] Specifically, supplement the generated power distribution parameters (such as line layout, circuit breaker specifications, cable specifications, etc.) into these adjusted scenario templates to form multiple candidate templates containing the new power distribution parameters.

[0077] Exemplarily, after adjustment, three scenario templates are obtained: Template A (office lighting), Template B (conference room lighting), and Template C (corridor lighting). Then, the power distribution parameters for office lighting are generated, including specific line layout, circuit breaker specifications, and cable specifications. Finally, these power distribution parameters are supplemented into Template A to form Candidate Template A.

[0078] Based on the preset evaluation conditions, evaluate the adaptability of each candidate template after adding new power distribution parameters respectively, and sort multiple candidate templates according to the size of the adaptability to obtain the sorting result.

[0079] Specifically, the preset evaluation conditions are as follows: Define the core evaluation dimensions of the candidate template, including voltage compatibility, load capacity matching degree, and safety specification degree, and set the evaluation coefficients for each core evaluation dimension. In this embodiment, the evaluation coefficients of voltage compatibility, load capacity matching degree, and safety specification degree are set to 0.4, 0.3, and 0.3 respectively. The calculation formula for its adaptability is: Adaptability = Voltage Compatibility * 0.4 + Load Capacity Matching Degree * 0.3 + Safety Specification Degree * 0.3; After evaluating the adaptability of each candidate template after adding new power distribution parameters, sort them in descending order to obtain the sorting result.

[0080] Based on the sorting result, screen out the scenario template with the highest adaptability as the optimal template.

[0081] Exemplarily, if the adaptability of candidate template A is 90, the adaptability of candidate template B is 7, and the adaptability of candidate template C is 60, then select candidate template A as the optimal template.

[0082] Use the optimal template as the lighting power distribution design drawing of the current design and display it to feedback to the user.

[0083] Specifically, use the optimal template as the lighting power distribution design drawing of the current design, and display it to the user for viewing through a terminal device (such as a computer, tablet, etc.). The user can see detailed information such as the line layout, circuit breaker specifications, and cable specifications, and can make further confirmation or adjustment according to needs, such as:

[0084] Generate the lighting power distribution design drawing of the current design from the screened scenario template with the highest adaptability, and display it to the user through the interface of a terminal device (such as a computer, tablet, or mobile phone); the user inputs the parameters that need to be adjusted in the lighting power distribution design drawing through an interactive method, and the parameter allocation model adjusts it individually to obtain the adjusted lighting power distribution design drawing; the user receives the adjusted lighting power distribution design drawing through the terminal device, and after confirmation, the adjusted lighting power distribution design drawing is the final lighting power distribution design drawing.

[0085] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices generate means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0086] By utilizing the historical lighting power distribution design drawing data, the present invention constructs a scene template database and a parameter allocation model, achieving highly efficient automation of the design process, greatly shortening the design cycle and improving the design efficiency. At the same time, it supports users to input design requirements interactively, customize and generate personalized power distribution parameters to meet the specific needs of different users and projects. Secondly, the present invention can screen out scene templates that highly match the user's needs and achieve optimization in design parameters. In addition, the parameter allocation model obtained through machine learning model training outputs optimized power distribution parameters, improving the quality and reliability of the design. The interactive mode of the user is used to finally adjust the parameters of the lighting power distribution design drawing, further enhancing the flexibility of the design and the user satisfaction.

[0087] Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be accomplished by instructing relevant hardware through a program, which can be stored in a computer-readable storage medium. The storage medium includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, magnetic disc storage, tape storage, or any other medium that can be used to carry or store data and is computer-readable.

[0088] It should also be noted that the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

Claims

1. An intelligent design interaction system based on a large model, characterized in that, The described design interaction system includes: A data processing module, which is used to obtain historical lighting distribution design drawing data for preprocessing, identify the scene features in the lighting distribution design drawing, and construct a scene template database with the scene features; A model construction module, which is used to extract key feature parameters from the historical lighting distribution design drawing data, use the key feature parameters as input and the distribution parameters of the lighting distribution design drawing as output to train a machine learning model, and obtain a parameter allocation model. The specific steps are as follows: Extract key feature parameters from the historical lighting distribution design drawing, specifically including lamp type, single lamp power, and lighting level; Use the key feature parameters as input features and the distribution parameters of the lighting distribution design drawing as output parameters, and mark the mapping relationship between the input features and the output parameters. Among them, the distribution parameters include line layout, circuit breaker specifications, and cable specifications; Adopt a machine learning model to train the machine learning model according to the marked mapping relationship between the input features and the output parameters to obtain a parameter allocation model; An information screening module, which is used to input design requirements through an interactive method, extract the scene information in the design requirements, and screen out all scene templates with a matching degree exceeding the matching threshold from the scene template database according to the scene information; A dynamic adjustment module, which is used to analyze the user's implicit preference emotions based on a questionnaire pre-answered by the user to dynamically adjust the matching threshold, perform secondary screening and adjustment on the screened scene templates, and obtain adjusted scene templates; A data analysis module, which is used to extract key feature parameters through the input design requirements and input them into the parameter allocation model to output the distribution parameters of the current design; An identification and generation module, which is used to supplement the distribution parameters of the current design into the adjusted scene template, analyze its adaptability, and screen out the scene template with the highest adaptability as the lighting distribution design drawing of the current design for display.

2. The intelligent design interaction system based on a large model according to claim 1, wherein The step of obtaining historical lighting distribution design drawing data for preprocessing, identifying the scene features in the lighting distribution design drawing, and constructing a scene template database with the scene features is as follows: Collect historical lighting distribution design drawing data from different projects and regions and perform preprocessing on it. Among them, the preprocessing includes image clarification processing and format unification; Use computer vision technology to identify the scene features in the historical lighting distribution design drawing data. Among them, the scene features include room type, lamp layout, and switch position; Deposit the identified scene features and their corresponding complete historical lighting distribution design drawings as entries into the scene template database.

3. The intelligent design interaction system based on a large model according to claim 1, wherein The step of inputting design requirements through an interactive method, extracting the scene information in the design requirements, and screening out all scene templates with a matching degree exceeding the matching threshold from the scene template database according to the scene information is as follows: The user inputs design requirements through an interactive method, including room type, lighting style, energy efficiency requirements, etc. Among them, the interactive method includes text box input or integrating a voice recognition module to convert voice into text; Extract the scene information in the design requirements and compare it with the entries in the scene template database to screen out all scene templates with a matching degree exceeding the preset matching threshold.

4. The intelligent design interaction system based on a large model according to claim 3, wherein Based on the questionnaire pre - replied by the user, analyze the user's implicit preference emotions to dynamically adjust the matching threshold, conduct a secondary screening and adjustment on the selected scenario templates to obtain the adjusted scenario templates. The specific steps are as follows: Set the questionnaire structure and send it to the user terminal. Among them, the questionnaire structure includes explicit demand collection and implicit demand collection; Collect the reply results of the user to the questionnaire, and use a natural language processing model to extract the sentiment polarity and keywords from the reply results; Based on the extracted sentiment polarity and keywords, adjust the matching threshold of the corresponding features of the sentiment polarity and keywords to obtain the adjusted matching threshold; Conduct a secondary screening on the selected scenario templates according to the adjusted matching threshold, and adjust the scenario templates after the secondary screening based on the corresponding features of the sentiment polarity and keywords.

5. The intelligent design interaction system based on a large model according to claim 4, wherein Supplement the currently designed power distribution parameters into the adjusted scenario templates, analyze their adaptability, and screen out the scenario template with the highest adaptability as the current designed lighting power distribution design drawing for display. The specific steps are as follows: Supplement the generated power distribution parameters into the adjusted scenario templates to generate multiple candidate templates; Through the preset evaluation conditions, evaluate the adaptability of each candidate template after adding the new power distribution parameters respectively, and sort the multiple candidate templates according to the size of the adaptability to obtain the sorting result; Based on the sorting result, screen out the scenario template with the highest adaptability as the optimal template; Use the optimal template as the current designed lighting power distribution design drawing and display it to the user.

Images