A Reliability Design Method Based on Evidence Theory
-
摘要: 证据理论是一种用于描述不确定变量的非连续概率模型,本文提出一种基于证据理论的可靠性设计方法。在该方法中,将证据理论视为具有一定非精确概率分布的区间模型,运用区间模型的功能度量法,通过在约束函数中改变优化变量的取值摄动起作用约束中的不确定变量从而求得问题中不确定变量的最不可靠点,再综合运用序列优化策略与基于信赖域的响应面方法求得系统的最优解。算例表明,此方法能够解决基于证据理论的单学科及多学科可靠性设计问题并且具有较好的计算效率。Abstract: The evidence theory is a kind of discontinuous probability model for describing uncertainty variables. This paper proposes a reliability design method based on the evidence theory, which treats the evidence theory as an interval model with imprecise probability distribution. By using the function measurement with the interval model and through changing the optimization variables in the restraint function perturbation,we obtain the most unreliable points in the uncertainty variables. We obtain the optimal solution with the sequence optimization strategy and the response surface method based on trust region. The numerical examples show that the reliability design method can effectively solve the reliability design problems in a single discipline or multiple disciplines based on the evidence theory and has high computational efficiency.
-
Key words:
- design /
- efficiency /
- optimization /
- probability
-
[1] Agarwal H,Renaud J E,Preston E L. Uncertainty quantification using evidence theory in multidisciplinary design optimization[J]. Reliability Engineering and System Safety, 2004, 85:281 ~294 [2] Elishakoff I. Whys,Hows in Uncertainty Modelling. Probabil-ity,Fuzziness and Anti-Optimizati on[M]. Springer Verlag Wien Dezember, 1999:265 ~285 [3] 曹鸿钧, 段宝岩. 基于凸集合模型的非概率可靠性研究[J]. 计 算力学学报, 2005, 22(5):546 ~549 [4] 郭惠昕, 刘德顺, 胡冠昱, 梁亮. 证据理论与区间分析相结合的可 靠性优化设计方法[J]. 机械工程学报, 2008, 44(12):35 ~41 [5] Du X P. Uncertainty analysis with probability and evidence theo-ry[A]. International Design Engineering Technical Confer-ences[C],ASME:2006 [6] Zissimos P M,Zhou J. A design optimization method using evi-dence theory[J]. ASEM Journal of Mechanical Design, 2006, 128(7):901 ~908 [7] Agarwal H,Renaud J E,Preston E L. Trust Region Managed Re-liability Based Design Optimization Using Evidence Theory[A]. In:44th Structures, Structural Dynamics, and Materials Con-ference[C],AIAA;2002 [8] Du X P, Guo J, Beeram H. Sequential optimization and reliability assessment for multidisciplinary systems design[J]. Struct Mul-tidisc Optim 2008, 35:117 ~130 [9] 陶友瑞. 基于凸模型的多学科不确定性优化设计方法[D]. 长 沙:湖南大学, 2010
点击查看大图
计量
- 文章访问数: 164
- HTML全文浏览量: 18
- PDF下载量: 2
- 被引次数: 0