航空学报 > 2020, Vol. 41 Issue (5): 623404-623404   doi: 10.7527/S1000-6893.2019.23404

飞行器气动外形数值优化与设计专栏

飞行器多学科耦合伴随体系的现状与发展趋势综述

黄江涛1, 刘刚1, 高正红2, 周铸1, 陈作斌1, 江雄1   

  1. 1. 中国空气动力研究与发展中心, 绵阳 621000;
    2. 西北工业大学 航空学院, 西安 710072
  • 收稿日期:2019-08-26 修回日期:2019-09-10 出版日期:2020-05-15 发布日期:2019-10-10
  • 通讯作者: 黄江涛 E-mail:hjtcyf@163.com
  • 基金资助:
    国家自然科学基金(11402288);国家重点研发计划(2016YFB0200704)

Current situation and development trend of multidisciplinary coupled adjoint system for aircraft

HUANG Jiangtao1, LIU Gang1, GAO Zhenghong2, ZHOU Zhu1, CHEN Zuobin1, JIANG Xiong1   

  1. 1. Computational Aerodynamics Institute, China Aerodynamics Research and Development Center, Mianyang 621000, China;
    2. School of Aeronautics, Northwestern Polytechnical University, Xi'an 710072, China
  • Received:2019-08-26 Revised:2019-09-10 Online:2020-05-15 Published:2019-10-10
  • Supported by:
    National Natural Science Foundation of China (11402288); National Key R&D Program of China (2016YFB0200704)

摘要: 多学科耦合伴随方法具有多学科耦合灵敏度计算量与各个学科设计变量个数均基本无关等优点,是一个值得关注的发展方向。面向气动、电磁、声学、结构、红外等与飞行器设计息息相关的学科,针对多学科耦合伴随方法的优势、现状、难点以及未来发展趋势开展研究与论述,系统性地分析了单一学科、多学科伴随方法的核心内容、关键技术与发展现状,对边界条件处理、交叉学科雅克比推导以及大型稀疏矩阵存储处理、求解等关键技术进行系统讨论,针对典型的关键环节和基础科学问题,给出了研究思路与解决方案,并进一步展望了多学科耦合伴随理论与应用发展趋势。希望能够为从事多学科伴随优化方法与应用的研究人员提供有意义的参考,促进多学科耦合灵敏度这一基础科学问题以及基于高保真度分析手段的多学科优化(MDO)技术的发展。

关键词: 多学科耦合伴随方法, 多学科优化, 耦合灵敏度, 交叉雅克比, 不确定度分析

Abstract: The multidisciplinary coupled adjoint system has many advantages, such as independence of multidisciplinary coupled sensitivity analysis from the number of design variables of various disciplines, and is thus an interesting development direction. This paper discusses the advantages, current situation, difficulties and future development trend of the multidisciplinary coupled adjoint system, including the disciplines of aerodynamics, structure, electromagnetism, acoustics, and infrared, which are closely related to aircraft design. The core contents, key technologies and current development situation of the single discipline and the multidisciplinary adjoint system are analyzed. The key technologies such as boundary condition processing, Jacobi derivation of multiple disciplines, and storage processing and solution of large sparse matrix are discussed systematically. The research ideas and solutions for respective key technologies and basic problems are given, and the development trend of multidisciplinary coupled adjoint theory and application is provided. Through this systematic discussion, we hope to provide meaningful suggestions for researchers engaged in multidisciplinary adjoint optimization and application, promote the research on multidisciplinary coupled system sensitivity, and the development of Multidisciplinary Optimization (MDO) technology based on high fidelity analysis.

Key words: multidisciplinary coupled adjoint method, multidisciplinary optimization, coupled sensitivity, cross Jacobi, uncertainty analysis

中图分类号: