Research On The Model And Data Hybrid-Driven Approaches For Parameters Optimization In Overall Design Of Aircraft

As a typical representative of high-end equipment in the field of aviation,aircraft research on its overall design method has important guiding significance for the innovative development of high-end equipment,and is also of great significance to the development of China’s manufacturing industry.The overall design problem of aircraft not only has the characteristics of high-tech equipment manufacturing,such as strong technical innovation,long development cycle,and large investment.It is further analyzed and found to have system complexity,dynamic demand,rapid response,and difficult to obtain data.In view of the above characteristics and problems,this paper fully considers the model-based systems engineering(MBSE)ideas and the advantages and applicability of data-driven technology.First,the technical framework for the model and data hybrid-driven overall design of aircraft is proposed.The key to the fusion of the two technologies is the design and optimization of the system parameters;then,the model-driven system parameter demand generation method and the data-driven system parameter optimization method were researched separately;finally,the application part of the plateau environment aircraft operation scenario was taken as an example to carry out research and analysis.The main work and innovations of the paper are as follows:(1)Construct a framework for the model and data hybrid-driven overall design of aircraft.In-depth analysis of the characteristics of the overall design problem of aircraft,research on the characteristics and applicability of model-driven technology and data-driven technology,and the relationship between them.The process points out the key research of hybrid-driven approach,the specific model-driven technology and data-driven technology applicable in the framework.(2)A multi-view model-driven demand generation method research on aircraft system parameters.There are many stakeholders involved in the demand identification of aircrafts.Drawing on the modeling ideas in the model-based system engineering methodology,a multi-view modeling framework suitable for the overall design of aircrafts is constructed,and model specifications and standards are clarified.By constructing the view product model of the overall design of the aircraft under this framework,combing the mapping relationship between the view products to determine the traceability relationship,the scene requirements outside the system are converted into system parameter requirements.(3)A system parameter optimization method based on multi-fidelity surrogate model is proposed.In the face of the major difficulties in the actual development process of aircraft,such as the small amount of data,the difficulty of obtaining high-precision data,the calculation accuracy and the calculation efficiency must meet the production requirements,etc.,you can first use a large amount of low-precision data to build a low-fidelity surrogate model,and then introduce a small amount of high-precision data to correct the low-fidelity data,thereby forming a multi-fidelity surrogate model parameter optimization method,which can effectively solve the above problems.This paper takes the aircraft operation scenario in the plateau environment as an example to conduct an application study,and the experimental results verify the feasibility and effectiveness of the framework and the proposed methods for the model and data hybrid-driven overall design of aircraft in this paper.At the same time,the research in this paper also provides ideas and solutions for more innovative development of high-end equipment.