An intelligent design method for aircraft overall conceptual scheme based on large language model

Abstract

The application discloses an airplane overall concept scheme intelligent design method based on a large language model, and belongs to the field of airplane overall design. The method fully gives play to the capacity of the large language model in knowledge integration, multidisciplinary collaboration and optimization design. By constructing a knowledge base of overall parameters, engine parameters and aerodynamic layout parameters, and in combination with the powerful text understanding and reasoning capacity of the large language model, the method realizes the rapid generation and optimization of design. Compared with traditional methods and existing tools, the large language model can significantly improve design efficiency, optimize the design process and output high-quality parameters, thereby meeting the needs of modern airplane design for rapid response and innovative design. Advantages of the method are as follows. First, the large language model can efficiently process multidisciplinary knowledge and realize cross-field collaborative design. Second, the parameter estimation and design optimization capacity of the large language model is significantly improved, thereby providing more accurate support for design schemes. The method provides strong technical support for modern airplane design.

Description

Technical Field

[0001] This invention belongs to the field of aircraft overall design and relates to an intelligent design method for aircraft overall concept schemes based on large language models. Background Technology

[0002] The overall conceptual design of an aircraft is a crucial step in aeronautical engineering, directly impacting its performance, cost, and feasibility. Traditional design methods rely heavily on manual calculations and experience-based judgment, resulting in a cumbersome, time-consuming, and inefficient design process. Particularly during the conceptual design phase, designers need to integrate knowledge from multiple disciplines, including aerodynamics, structure, and propulsion, placing extremely high demands on their design capabilities. Furthermore, traditional methods are significantly lacking in multidisciplinary optimization and cross-departmental collaboration, making it difficult to generate high-quality design solutions quickly, thus limiting innovation and improving design efficiency.

[0003] In contrast, intelligent design methods based on large language models exhibit significant advantages. With their powerful knowledge integration capabilities and high computational efficiency, large language models can quickly complete complex tasks such as aircraft parameter estimation, engine selection, and aerodynamic layout design. This intelligent tool not only significantly reduces repetitive work in traditional design but also optimizes the design process and improves design quality through the deep integration of multidisciplinary knowledge. Especially under complex design requirements, large language models can overcome the limitations of traditional tools, generating superior design solutions through powerful reasoning capabilities and cross-domain knowledge fusion, injecting new vitality into aircraft concept design.

[0004] Therefore, developing an intelligent design method for aircraft overall concept schemes based on large language models can not only integrate multidisciplinary knowledge and optimize the design process, but also significantly improve design efficiency and scheme quality, meeting the urgent needs of rapid response and innovation in modern aircraft design. This method fully leverages the unique advantages of large language models in knowledge integration, multidisciplinary collaboration, and design innovation capabilities, providing an effective way to solve the pain points of traditional design methods, and has significant engineering practice value and broad application prospects.

[0005] Traditional aircraft overall concept design relies on tedious manual calculations and experience-based judgments, resulting in long design cycles, low efficiency, and significant shortcomings in multidisciplinary optimization and cross-departmental collaboration, making it difficult to quickly generate high-quality design solutions. While existing intelligent design tools improve efficiency, they still fall short in terms of systematicity, integration of multidisciplinary knowledge, and accuracy of design parameters.

[0006] This invention proposes an intelligent design method for aircraft overall concept schemes based on a large language model, fully leveraging the capabilities of large language models in knowledge integration, multidisciplinary collaboration, and optimized design. By constructing a knowledge base of overall parameters, engine parameters, and aerodynamic layout parameters, and combining this with the powerful text understanding and reasoning capabilities of the large language model, rapid design generation and optimization are achieved. Compared to traditional methods and existing tools, the large language model significantly improves design efficiency, optimizes the design process, and outputs high-quality parameters, thus meeting the demands of modern aircraft design for rapid response and innovative design. Its advantages are reflected in: firstly, the large language model can efficiently process multidisciplinary knowledge, enabling cross-domain collaborative design; secondly, its parameter estimation and design optimization capabilities are significantly improved, providing more accurate support for design schemes. This provides strong technical support for modern aircraft design. Summary of the Invention

[0007] The purpose of this invention is to provide an intelligent design method for aircraft overall concept schemes based on a large language model, aiming to solve the problems of tedious manual calculations, low efficiency, difficulties in multidisciplinary optimization, and complex cross-departmental collaboration in traditional aircraft overall scheme design. This method can quickly generate high-quality aircraft overall design schemes and significantly improve design efficiency and innovative design capabilities.

[0008] The technical solution of the present invention:

[0009] An intelligent design method for aircraft overall concept schemes based on a large language model includes the design of an overall parameter estimation module, an engine selection module, and a scheme determination and parameter output module. Specifically:

[0010] The overall parameter estimation module integrates historical aircraft data, performance parameters, and design specifications to construct a structured aircraft database and knowledge base consisting of the RAG knowledge base, model database, and design knowledge base. Users first input cruise altitude, Mach number, payload, range, acceleration climb, overload capacity, and design target constraints, as well as takeoff and landing characteristics, climb characteristics, and cruise characteristics requirements. The design requirements analysis module processes the input data to generate search suggestions. The search module retrieves design experience and knowledge from the structured aircraft database and knowledge base based on these suggestions. This retrieved design experience and knowledge are then input into the parameter estimation model for overall parameter trade-off estimation, quickly yielding takeoff gross weight, fuel weight coefficient, empty weight coefficient, thrust-to-weight ratio, wing loading, and bulk density. The resulting overall parameters are input into the aircraft gross weight calculation tool for weight verification. The aircraft carpet drawing tool generates weight characteristics, layout design requirements, propulsion system performance requirements, and other overall parameter reference values ​​based on the overall parameters. These results are fed back into the structured aircraft database and knowledge base to achieve model and scheme asset reserves and reuse. The overall parameter estimation module outputs comprehensive basic data.

[0011] Engine selection module: Based on the overall parameter estimation results, flight altitude, flight speed, and inlet type are input into the large model to query inlet performance and obtain the total pressure recovery coefficient; simultaneously, thrust requirements, cruise altitude, and speed, etc., are input into the engine thrust requirement correction module, combined with the engine outlet area obtained from the engine outlet area estimation module, to obtain the test thrust requirement; the test thrust requirement and total pressure recovery coefficient are input into the large model to recommend engines, obtain test performance and engine outlet area, and enter the installation performance calculation stage; a suitable engine type is selected from the structured aircraft database and knowledge base, considering thrust requirements, fuel efficiency, and compatibility, and outputs engine model, engine size, weight, takeoff installation thrust, fuel consumption rate, cruise installation thrust, and fuel consumption rate.

[0012] The scheme determination and parameter output module combines overall parameters and engine parameters, inputs relevant data into the aerodynamic layout scheme generation module, and searches the aerodynamic layout knowledge base in the structured aircraft database and knowledge base to match the aerodynamic layout form that is suitable for the current design requirements. It determines whether the layout is conventional, canard, or flying wing, and initially generates the corresponding aerodynamic parameters, including wing aspect ratio, sweep angle, root-to-tip ratio, horizontal tail area, and vertical tail area. The initially generated aerodynamic layout form and parameters are then input into the aerodynamic performance evaluation module. Combining wind tunnel test data, CFD simulation analysis results, and performance calculation models, the module performs a multi-dimensional evaluation of the scheme's lift-drag characteristics, handling stability, and cruise efficiency. If the evaluation does not meet the design requirements, the module returns to the aerodynamic layout scheme generation module to readjust the layout form and parameters until the evaluation is passed. Finally, the optimized overall design scheme is output, including the determined aerodynamic layout form, precise aerodynamic parameters, and a corresponding performance verification report.

[0013] The beneficial effects of this invention are as follows: The method of this invention provides an efficient, accurate, and innovative solution for aircraft overall concept design, significantly improving design efficiency and quality, and promoting the development of aircraft design technology. This invention provides an intelligent design method for aircraft overall concept schemes based on a large language model. By integrating a large language model and a multidisciplinary knowledge base, it achieves rapid parameter estimation, engine selection, and aerodynamic layout design, significantly improving design efficiency and scheme quality. This method can optimize the design process, achieve multidisciplinary optimization, and shorten the design cycle. The method is based on a large language model and knowledge base, and through a modular design workflow intelligent agent, it achieves intelligent optimization of overall parameter estimation, engine selection, and aerodynamic layout design. The method can automatically verify design parameters, match the optimal design scheme through multidisciplinary knowledge fusion, and propose design trade-off suggestions, significantly improving design efficiency and scheme quality. Attached Figure Description

[0014] Figure 1The overall flowchart of the intelligent design method for the overall concept scheme of aircraft provided by the present invention.

[0015] Figure 2 A flowchart illustrating the overall parameter estimation module provided by this invention.

[0016] Figure 3 The logic flowchart of the engine selection module provided by the present invention.

[0017] Figure 4 A detailed flowchart of the scheme determination and parameter output module provided by the present invention. Detailed Implementation

[0018] An intelligent design method for aircraft overall concept schemes based on a large language model includes an overall parameter estimation module, an engine selection module, a scheme determination and parameter output module, and a knowledge base query system, specifically:

[0019] (1) Overall parameter estimation module:

[0020] By integrating historical aircraft data, performance parameters, and design specifications, a structured aircraft database and knowledge base are constructed, consisting of the RAG knowledge base, the model database, and the design knowledge base.

[0021] The user inputs design target indicators and constraints (aircraft type, cruise altitude, cruise Mach number, payload, range, acceleration climb, overload capacity) and characteristic requirements (takeoff and landing characteristics, climb characteristics, cruise characteristics). The design requirement analysis module processes the above input data and generates search suggestions. The search module retrieves design experience and knowledge from the structured aircraft database and knowledge base based on the suggestions, and inputs the search results into the parameter estimation model to perform overall parameter trade-off estimation, quickly obtaining the first takeoff gross weight, fuel weight coefficient, empty weight coefficient, thrust-to-weight ratio, wing loading, and bulk density.

[0022] The estimated overall parameters are input into the aircraft gross weight calculation tool (large model calling tool) for weight verification. At the same time, the aircraft carpet map drawing tool (large model calling tool) generates weight characteristics, layout design requirements, propulsion system index requirements and other overall parameter reference values. The above results are fed back into the structured aircraft database and knowledge base to realize the asset reserve and reuse of models and schemes, and finally output comprehensive basic data.

[0023] (2) Engine selection module:

[0024] Based on the parameters output by the overall parameter estimation module, engine selection requirements (flight altitude, flight speed, thrust requirements, cruise altitude, speed, etc.) are extracted.

[0025] By inputting flight altitude, flight speed, and inlet type into the large model to query inlet performance, the total pressure recovery coefficient is obtained. At the same time, thrust-related data such as thrust requirement, cruise altitude, and speed are input into the engine thrust requirement correction module. Combined with the engine outlet area obtained from the engine outlet area estimation module, the bench thrust requirement is calculated.

[0026] The test bench thrust requirement and total pressure recovery coefficient are input into the large model to recommend an engine, and the test bench performance and engine outlet area are obtained, which then proceeds to the installation performance calculation stage. Engine types and performance parameters that meet the requirements are selected from the structured aircraft database and knowledge base, and the compatibility of single or multiple engines is evaluated. Taking into account thrust requirement, fuel efficiency and compatibility, the engine model, engine size, weight, takeoff installation thrust, fuel consumption rate, cruise installation thrust and fuel consumption rate are output, providing basic data for subsequent design.

[0027] (3) Scheme determination and parameter output module:

[0028] Integrate overall parameters and engine parameters to generate the design inputs required for aerodynamic layout.

[0029] Based on the design point thrust-to-weight ratio, wing loading, and bulk density output by the overall parameter estimation module, and the ground thrust and fuel consumption rate output by the engine selection module, the design takeoff gross weight, fuel weight coefficient, range-constrained takeoff gross weight requirement, and wing area are calculated. The relevant data are then input into the aerodynamic layout scheme generation module. By querying the RAG knowledge base containing overall parameters and aerodynamic layout parameters, and combining the reasoning ability of the large language model, the module matches the aerodynamic layout form (conventional layout / canard layout / flying wing layout) that is suitable for the current design requirements, and initially generates aerodynamic parameters such as wing aspect ratio, sweep angle, root-to-tip ratio, horizontal tail area, and vertical tail area.

[0030] The initially generated aerodynamic layout and parameters are input into the aerodynamic performance evaluation module. Combined with wind tunnel test data, CFD simulation analysis results, and performance calculation models, the lift-drag characteristics, handling stability, and cruise efficiency of the scheme are evaluated from multiple dimensions. If the evaluation does not meet the design requirements, the module returns to the aerodynamic layout generation module to readjust the layout and parameters until the evaluation is passed. The optimized overall design scheme is then output, including the determined aerodynamic layout, precise aerodynamic parameters, and a corresponding performance verification report.

[0031] (4) Knowledge base query system:

[0032] It provides historical aircraft data, performance parameters, design specifications, and engine performance parameters for the overall parameter estimation module, engine selection module, and scheme determination and parameter output module. It constructs a structured aircraft database and knowledge base consisting of a RAG knowledge base, a model database, and a design knowledge base to support rapid query and knowledge integration during the design process. At the same time, it receives design results from each module (overall parameter verification results, engine compatibility assessment results, aerodynamic layout schemes and verification reports, etc.) to realize the asset storage and reuse of models and schemes.

Claims

1. An intelligent design method for aircraft overall concept schemes based on large language models, characterized in that, This includes the design of the overall parameter estimation module, the engine selection module, and the scheme determination and parameter output module; specifically: Overall Parameter Estimation Module: Integrates historical aircraft data, performance parameters, and design specifications to construct a structured aircraft database and knowledge base consisting of the RAG knowledge base, model database, and design knowledge base. Users first input cruise altitude, Mach number, payload, range, acceleration climb, overload capacity design target constraints, as well as takeoff and landing characteristics, climb characteristics, and cruise characteristics requirements. The design requirements analysis module processes the input data to generate search suggestions. The retrieval module then searches the structured aircraft database and knowledge base for design experience and knowledge based on these suggestions. This retrieved design experience and knowledge are input into the parameter estimation model for overall parameter trade-off estimation, quickly yielding takeoff gross weight, fuel weight coefficient, empty weight coefficient, thrust-to-weight ratio, wing loading, and bulk density. The resulting overall parameters are input into the aircraft gross weight calculation tool for weight verification. The aircraft carpet drawing tool generates weight characteristics, layout design requirements, propulsion system indicator requirements, and other overall parameter reference values ​​based on the overall parameters. These results are fed back into the structured aircraft database and knowledge base to achieve model and scheme asset reserves and reuse. The overall parameter estimation module outputs comprehensive basic data. Engine selection module: Based on the overall parameter estimation results, the flight altitude, flight speed, and air intake type are input into the large model to query the air intake performance and obtain the total pressure recovery coefficient; at the same time, thrust-related data such as thrust requirement, cruise altitude, and speed are input into the engine thrust requirement correction module, and combined with the engine exit area obtained from the engine exit area estimation module, the bench thrust requirement is obtained. Input the test bench thrust requirement and total pressure recovery coefficient into the large model to recommend the engine, obtain the test bench performance and engine outlet area, and then proceed to the installation performance calculation stage. Select the appropriate engine type from the structured aircraft database and knowledge base, taking into account thrust requirements, fuel efficiency and compatibility, and output the engine model, engine size, weight, takeoff installation thrust, fuel consumption rate, cruise installation thrust and fuel consumption rate; The scheme determination and parameter output module combines overall parameters and engine parameters, inputs relevant data into the aerodynamic layout scheme generation module, and searches the aerodynamic layout knowledge base in the structured aircraft database and knowledge base to match the aerodynamic layout form that is suitable for the current design requirements. It determines whether the layout is conventional, canard, or flying wing, and initially generates the corresponding aerodynamic parameters, including wing aspect ratio, sweep angle, root-to-tip ratio, horizontal tail area, and vertical tail area. The initially generated aerodynamic layout form and parameters are then input into the aerodynamic performance evaluation module. Combining wind tunnel test data, CFD simulation analysis results, and performance calculation models, the module performs a multi-dimensional evaluation of the scheme's lift-drag characteristics, handling stability, and cruise efficiency. If the evaluation does not meet the design requirements, the module returns to the aerodynamic layout scheme generation module to readjust the layout form and parameters until the evaluation is passed. Finally, the optimized overall design scheme is output, including the determined aerodynamic layout form, precise aerodynamic parameters, and a corresponding performance verification report.

Images