航空学报 > 2025, Vol. 46 Issue (4): 330946-330946   doi: 10.7527/S1000-6893.2024.30946

基于多目标人工蜂鸟算法的研制保证等级分配

崔瑀欣1, 陆中1(), 周伽2   

  1. 1.南京航空航天大学 民航学院,南京 211106
    2.东方航空技术有限公司江苏分公司 生产控制中心,南京 211106
  • 收稿日期:2024-07-15 修回日期:2024-08-07 接受日期:2024-10-14 出版日期:2024-10-24 发布日期:2024-10-23
  • 通讯作者: 陆中 E-mail:luzhong@nuaa.edu.cn
  • 基金资助:
    国家重点研发计划(2023YFB4302403);中央高校基本科研业务费专项资金(NG2023003)

Development assurance level assignment based on multi-objective artificial hummingbird algorithm

Yuxin CUI1, Zhong LU1(), Jia ZHOU2   

  1. 1.College of Civil Aviation,Nanjing University of Aeronautics & Astronautics,Nanjing 211106,China
    2.Center of Production Control,Eastern Airlines Technic Co. ,Ltd. ,Jiangsu Branch,Nanjing 211106,China
  • Received:2024-07-15 Revised:2024-08-07 Accepted:2024-10-14 Online:2024-10-24 Published:2024-10-23
  • Contact: Zhong LU E-mail:luzhong@nuaa.edu.cn
  • Supported by:
    National Key Research and Development Program of China(2023YFB4302403);Fundamental Research Funds for the Central Universities of China(NG2023003)

摘要:

分配研制保证等级(DAL)是飞机系统研制过程中的一项重要工作,通常要求在满足DAL分配原则的基础上使得研制成本最小。构建了一种面向DAL分配的多目标优化模型,该模型将DAL分配原则和顶层失效状态概率要求分别作为定性和定量约束条件,以研制成本和系统重量最小化为优化目标,将为系统中功能或项目分配的DAL作为决策变量;在此基础上,提出了基于多目标人工蜂鸟算法的DAL分配方法。结合某飞机电传飞控系统给出了DAL分配实例,得到DAL分配的非支配解集。在相同测试条件下,与多目标粒子群算法相比,提出方法的运行时间缩短了17.59%,超体积指标(HV)提高了63.49%,表明提出方法能够快速收敛、求得的解集具有良好的分布性。

关键词: 系统安全性, 研制保证等级分配, 多目标优化, 多目标人工蜂鸟优化算法, 失效概率

Abstract:

Development Assurance Level (DAL) assignment is an important task in the process of aircraft system development. It is usually required to minimize development costs on the basis of satisfying the DAL assignment principles. A multi-objective optimization model for DAL assignment is constructed. The DAL assignment principles and the top-level failure condition probability requirement are taken as qualitative and quantitative constraints, respectively. The optimization objectives are the minimization of development cost and system weight. The DALs assigned to functions or items in the system are taken as decision variables. Besides, a method for DAL assignment is proposed based on the multi-objective artificial hummingbird algorithm. A case study is given based on a fly-by-wire system, and a non-dominated solution set for DAL assignment is obtained. Under the same test condition, the proposed method reduces the execution time by 17.59% and improves the HV metric by 63.49% compared with multi-objective particle swarm optimization, which indicates that the proposed method can converge quickly and the obtained solution set has good distribution.

Key words: system safety, Development Assurance Level (DAL) assignment, multi-objective optimization, multi-objective artificial hummingbird algorithm, failure probability

中图分类号: