一种新的基本随机变量重要性测度指标体系

固体力学与飞行器总体设计

本期目录 | 过刊浏览 | 高级检索

前一篇 | 后一篇

一种新的基本随机变量重要性测度指标体系

李璐祎, 吕震宙, 李维

西北工业大学航空学院, 陕西西安 710072

收稿日期:2011-09-23 修回日期:2011-10-10 出版日期:2012-07-25 发布日期:2012-07-24
通讯作者: 吕震宙,Tel.: 029-88460480 E-mail: zhenzhoulu@nwpu.edu.cn E-mail:zhenzhoulu@nwpu.edu.cn
基金资助:
国家自然科学基金(51175425);航空科学基金(2011ZA53015);西北工业大学博士论文创新基金(CX201205)

A New Importance Measure Index System for Basic Random Variables

LI Luyi, LU Zhenzhou, LI Wei

School of Aeronautics, Northwestern Polytechnical University, Xi’an 710072, China

Received:2011-09-23 Revised:2011-10-10 Online:2012-07-25 Published:2012-07-24
Supported by:
National Natural Science Foundation of China (51175425); Aeronautical Science Foundation of China (2011ZA53015); Doctorate Foundating of Northwestern Polytechnical University (CX201205)

摘要/Abstract

摘要： 综合当前基于方差的重要性测度与矩独立重要性测度的优点,建立了一个新的随机变量重要性测度指标体系。该体系从输出响应量的均值、方差以及可靠度指标方面对随机变量的重要性进行了分析,进而根据不同的要求衡量基本随机变量对系统或模型输出的影响程度。给出了各个重要性测度指标的定义,并探讨了他们与现有的基于方差的重要性测度指标的关系。通过算例说明了所提新的重要性测度指标体系的优越性。结果表明:新指标体系中的指标不但可以反映旧的指标,而且还对其进行了修正,克服了基于方差的重要性分析中随机变量取不同实现值时对输出响应量的影响相互抵消的问题,从而更加合理地衡量随机变量的重要性;与当前矩独立的重要性分析方法相比,新的指标体系在继承其优点的基础上能够从不同侧面更加全面地对随机变量的重要性进行分析,因而具有更广泛的应用范围。

关键词: 基本随机变量, 重要性测度, 方差重要性分析, 矩独立重要性分析, 无条件响应量特征值, 条件响应量特征值

Abstract: By combining the advantages of the existing variance-based importance measure and those of the moment independent importance measure, a new measure index system is established for importance analysis of basic random variables. This system analyzes the importance of the basic variables comprehensively in terms of the mean, variance and reliability index of the output responses. Therefore, it can measure the importance influences of the basic random variables on the whole range of uncertainty of the output responses of the system or model in accordance with different requirements. After the definitions of the importance measure index system are given, their relationships with the existing variance-based importance measure are studied. Several examples are used to illustrate the advantages of the new measure index system. The results show that the indices of the new importance measure index system can not only reflect the old indices, but can also be viewed as their modified version. They can overcome the disadvantage of the old ones, i.e. the effects of different realizations of the basic random variables on the output response may counteract each other. Hence, the new measure index system is more reasonable. Comparison with the existing moment independent importance measure index shows that the new measure index system can analyze the importance of random variables comprehensively from different aspects, while the advantage of the moment independent importance method is inherited in the new measure index system. Therefore, the new measure index system has a wider applicability scope.

Key words: basic random variable, importance measure, variance-based importance analysis, moment independent importance analysis, characteristic value of unconditional response, characteristic value of conditional response

中图分类号:

V221
TB114.3

李璐祎, 吕震宙, 李维. 一种新的基本随机变量重要性测度指标体系[J]. 航空学报, 2012, 33(7): 1255-1264.

LI Luyi, LU Zhenzhou, LI Wei. A New Importance Measure Index System for Basic Random Variables[J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2012, 33(7): 1255-1264.

参考文献

[1] Satelli A. Sensitivity analysis for importance assessment. Risk Analysis, 2002, 22(3): 579-590.
[2] Helton J C, Davis F J. Latin hypercube sampling and the propagation of uncertainty in analyses of complex systems. Reliability Engineering and System Safety, 2003, 81(1): 23-69.
[3] Helton J C, Davis F J. Sampling-based methods. Saltelli A, Chan K, Scott E M. Sensitivity Analysis. New York: Wiley, 2000: 101-153.
[4] Saltelli A, Marivoet J. Non-parametric statistics in sensitivity analysis for model output: a comparison of selected techniques. Reliability Engineering and System Safety, 1990, 28(2): 229-253.
[5] Sobol I M. Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates. Mathematics and Computers in Simulation, 2001, 55(1-3): 271-280.
[6] Iman R L, Hora S C. A robust measure of uncertainty importance for use in fault tree system analysis. Risk Analysis, 1990, 10(3): 401-406.
[7] Xu C, Gertner G. Uncertainty and sensitivity analysis for models with correlated parameters. Reliability Engineering and System Safety, 2008, 93(6): 1563-1573.
[8] Xu C, Gertner G. Extending a global sensitivity analysis technique to models with correlated parameters. Computational Statistics & Data Analysis, 2007, 51(12): 5579-5590.
[9] Hao W R, Lu Z Z, Tian L F. A novel method for analyzing variance based importance measure of correlated input variables. Acta Aeronautica et Astronautica Sinica, 2011, 32(9): 1637-1643. (in Chinese) 郝文瑞, 吕震宙, 田龙飞. 基于方差的相关输入变量重要性测度分析新方法. 航空学报, 2011, 32(9): 1637-1643.
[10] Chun M H, Han S J, Tak N I. An uncertainty importance measure using a distance metric for the change in a cumulative distribution function. Reliability Engineering and System Safety, 2000, 70(3): 313-321.
[11] Liu H B, Chen W, Sudjianto A. Relative entropy based method for probabilistic sensitivity analysis in engineering design. Journal of Mechanical Design, 2006, 128(3): 326-333.
[12] Borgonovo E. A new uncertainty importance measure. Reliability Engineering and System Safety, 2007, 92(6): 771-784.
[13] Liu Q, Homma T. A new computational method of a moment-independent uncertainty importance measure. Reliability Engineering and System Safety, 2009, 94(7): 1205-1211.
[14] Chen J B, Li J. A note on the principle of preservation of probability and probability density evolution equation. Probabilistic Engineering Mechanics, 2009, 24(1): 51-59.
[15] Song B F. A study on method for computing structural system failure probability. Xi’an: Northwestern Polytechnical University, 1994. 宋笔锋. 结构体系失效概率计算方法研究. 西安:西北工业大学, 1994.
[16] Ishigami T, Homma T. An importance quantification technique in uncertainty analysis for computer models. Proceedings of the ISUMA’90, First International Symposium on Uncertainty Modeling and Analysis, 1990: 398-403.

E-mail：hkxb@buaa.edu.cn

关于我们

期刊社服务

专业学科

封面文章

友情链接

主管单位：中国科学技术协会主办单位：中国航空学会北京航空航天大学

一种新的基本随机变量重要性测度指标体系

A New Importance Measure Index System for Basic Random Variables

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

编辑推荐

Metrics

本文评价

[1]	张政, 周长聪, 戴志豪, 任哲先, 岳珠峰. 基于重要性测度降维的液压管路稳健优化设计[J]. 航空学报, 2018, 39(8): 421902-421902.
[2]	赵海龙, 岳珠峰, 刘伟. 矩独立重要性分析的Kriging代理模型方法[J]. 航空学报, 2016, 37(7): 2234-2241.
[3]	张磊刚, 吕震宙, 陈军. 基于失效概率的矩独立重要性测度的高效算法[J]. 航空学报, 2014, 35(8): 2199-2206.
[4]	郝文锐, 吕震宙, 田龙飞. 基于方差的相关输入变量重要性测度分析新方法[J]. 航空学报, 2011, 32(9): 1637-1643.