期刊文献+

基于变步长离散随机集的风险不确定性分析方法

Analysis method on risk uncertainty based on variable step discrete random set
下载PDF
导出
摘要 针对信息不一致、不完整下的风险评估不确定性难以刻画与传播问题,提出一种基于变步长离散随机集理论的风险混合不确定性分析方法。将各类不完整、不精确信息转化为随机集刻画框架,在随机集理论框架下建立了统一的混合不确定性传播模型,利用随机扩张原理,计算出风险的不确定性包络曲线。为解决不一致冲突信息的不确定性合成,采用D-S证据合成原则实现多源不确定性的融合。为减小不确定性传播截尾相对误差,提出一种不确定性变量分布的变步长离散随机集刻画策略,并给出了基于变步长离散随机集理论的混合不确定性传播实施步骤。通过一个质量-弹簧-阻尼非线性物理与现象响应模型,验证了方法的有效性和可用性。 In view of hybrid uncertainty presentation and propagation considering the dissonance and imprecision of information in risk assessment,a hybrid uncertainty analysis method based on variable step discrete random set theory was proposed. All kinds of incomplete and dissonant knowledge was represented with random set framework,a unified hybrid uncertainty propagation model was built using random extension principle,and uncertainty envelope curvesof risk was calculated at the same time. To solve the uncertainty combination problem of dissonant and conflict informations,D-S evidence combination principle was used to merge multisource uncertainty informations. For reducing the tail relative error,a variable step discrete random set presentation strategy of uncertainty variables was proposed,and the analysis procedure of hybrid uncertainty propagation was put forward based on variable step discrete random set theory. In conclusion,a physics and phenomena response model of a mass-spring-damper system was taken to verify the effectiveness and feasibility of the proposed method.
出处 《北京航空航天大学学报》 EI CAS CSCD 北大核心 2018年第2期295-304,共10页 Journal of Beijing University of Aeronautics and Astronautics
关键词 风险评估 随机集理论 D-S证据理论 混合不确定性 相对误差 risk assessment random set theory D-S evidence theory hybrid uncertainty relative error
  • 相关文献

参考文献1

二级参考文献20

  • 1邹湘军,孙健,何汉武,陈新,郑德涛,张平.虚拟环境中多Agent决策的冲突证据合成研究[J].系统仿真学报,2006,18(4):1010-1014. 被引量:9
  • 2Jim W Hall. Uncertainty-based sensitivity indices for imprecise probability distributions[J].Reliability Engineering and System Safety (S0951-8320),2006,(10-11):1443-1451.
  • 3Fulvio Tonon. Using random set theory to propagate epistemic uncertainty through a mechanical system[J].Reliability Engineering and System Safety (S0951-8320),2004,(03):169-181.
  • 4Ha-RokBae,Ramana V.Grandhi,RobertA.Canfield. Anappoximation approach for uncertainty quantification using evidence theory[J].Reliability Engineering and System Safety (S0951-8320),2004,(03):215-225.
  • 5Harish Agarwal,John E.Renaud,Evan L.Preston. Uncertainty quantification using evidence theory in multidisciplinary design optimization[J].Reliability Engineering and System Safety (S0951-8320),2004,(01):281-294.
  • 6J C Heton,J D Johnson,W L Oberkampf. Sensitivity analysis in conjunction with evidence theory representations of epistemic uncertainty[J].Reliability Engineering and System Safety (S0951-8320),2006,(10-11):1414-1434.
  • 7Philipp Limbourg,Etieunede Rocquigny. Uncertainty analysis using evidence theory-confronting level-1 and level-2 approaches with data availability and computational constraints[J].Reliability Engineering and System Safety (S0951-8320),2010,(03):550-564.
  • 8Oberkampf W L,Helton J C,Sentz K. Mathematical representation of uncertainty[A].USA:AIAA,2001.
  • 9Oberkampf W L,Diegert K V,Alvin K F. Variability,uncertainty,and error in computational simulation[A].USA:AIAA,1998.59-72.
  • 10Oberkampf W L,Deland S M,Rutherford B M. A new methodology for the estimation of total uncertainty in computational simulation[A].USA:AIAA,1999.

共引文献6

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部