首页> 外文会议>Asia Simulation Conference/International Conference on System Simulation and Scientific Computing >Modeling and Analysis of Common Cause Failure Based on Monte Carlo-Neural Network
【24h】

Modeling and Analysis of Common Cause Failure Based on Monte Carlo-Neural Network

机译:基于Monte Carlo-神经网络的常见原因失败的建模与分析

获取原文

摘要

Due to the dearth of the observed common cause failures, some models are limited. Applicability and shortcomings of the present models are pinpointed. Based on the idea of the shock models, this paper modifies the assumption in BFR model and indicates that the conditional failure probability of the components on its 'stress' is a random variable. A new model is represented according to the physical model of component failure—stress-strength interference model. Using Monte Carlo simulation and RBF neural network technique, the distributed type and parameters of the conditional failure probability of the components was obtained. The approach can make the best use of the few available observations on common cause failure to calculate any multiplicity failure rates of the system. Examples are given to illustrate its feasibility and precision of the computation.
机译:由于观察到的常见原因失败的缺乏,某些型号有限。本模型的适用性和缺点是精确定位的。基于震动模型的思想,本文修改了BFR模型中的假设,并表明组件在其“压力”上的条件失效概率是一个随机变量。根据部件故障应力强度干扰模型的物理模型表示新模型。使用Monte Carlo仿真和RBF神经网络技术,获得了部件的条件失效概率的分布式类型和参数。该方法可以充分利用少数可用的观察常见原因未能计算系统的任何多重故障率。给出了实施例来说明其可行性和精度的计算。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号