系统工程与电子技术 ›› 2019, Vol. 41 ›› Issue (12): 2911-2918.doi: 10.3969/j.issn.1001-506X.2019.12.34

• 可靠性 • 上一篇    

基于双响应曲面的精密产品子集模拟可靠稳健优化设计

吴佳伟1, 宋华明1, 万良琪2, 黄甫1, 杨加猛3   

  1. 1. 南京理工大学经济管理学院, 江苏 南京 210094; 2. 南京航空航天大学经济与管理学院, 江苏 南京 211106; 3. 南京林业大学经济管理学院, 江苏 南京 210037
  • 出版日期:2019-11-25 发布日期:2019-11-27

Reliability-based robust design optimization for precision product by dual response surface methodology and subset simulation

WU Jiawei1, SONG Huaming1, WAN Liangqi2, HUANG Fu1, YANG Jiameng3   

  1. 1. School of Economics and Management, Nanjing University of Science and Technology, Nanjing 210094, China;
    2. College of Economics and Management, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China;
    3. College of Economics and Management, Nanjing Forestry University, Nanjing 210037, China
  • Online:2019-11-25 Published:2019-11-27

摘要:

针对不确定性场景下小失效概率的精密产品可靠性问题,提出了基于双响应曲面法(dual response surface methodology,DRSM)与子集模拟的可靠稳健优化方法。首先,建立精密产品多失效模式极限状态的RSM模型和以质量特性均值-方差为优化目标的功能函数DRSM模型。其次,在此基础上运用子集模拟法进行可靠性分析,将精密产品小失效概率描述成一系列较大的条件失效概率之积。最后,将极值事件的优化问题视作稀有事件的可靠性问题的特例,基于该转换思想采用子集模拟法将优化问题在可靠性问题的框架内进行求解。案例分析及验证结果表明所提方法的有效性。

关键词: 小失效概率, 可靠稳健优化, 子集模拟, 双响应曲面

Abstract:

To solve the problem of small failure probability in the precision product, a reliability-based robust design optimization method is proposed based on dual response surface methodology (DRSM) and subset simulation. Firstly, the limit state of failure modes is built based on RSM and performance function with mean-variance is built by DRSM. Secondly, the reliability analyze by the subset simulation, which convert small failure probability to large conditional failure probability. Then the larger conditional failure probability is estimated by density function of sampling. Finally, optimization problem is regarded as rare events for reliability problem. Subset simulation is used to solve the robust optimization within the reliability analysis framework. The results of numerical example demonstrate the effectiveness of the proposed method.

Key words: small probability failure, reliability-based design optimization, subset simulation, dual response surface methodology