基于离散时间贝叶斯网络的动态故障树分析的改良方法

传统基于离散时间贝叶斯网络的动态故障树分析方法的计算时间和计算精度受时间分段影响极大。基于复合梯形积分方法分析传统方法的计算误差,提出改良的动态门转化方法,补偿其计算误差。以某测量系统为例,建立动态故障树和贝叶斯网络,验证改良方法的可行性和高效性。结果表明:改良方法在时间分段较小时,能得到精确的系统失效概率。改良方法补偿了传统方法的计算误差,提高了结果的计算精度和计算效率,适用于服从各种常见分布的复杂系统。

Improved method for dynamic fault tree analysis based on discrete time Bayesian network

The computational time and computational accuracy of the traditional dynamic fault tree analysis method based on discrete time Bayesian network are greatly affected by time segmentation. The calculation error of the traditional method based on the compound trapezoidal integral method is analyzed, and an improved dynamic gate transformation method is proposed to compensate its calculation error. Taking a measurement system as an example, the dynamic measurement tree and the Bayesian network are established to verify the feasibility and efficiency of the improved method. The results show that the improved method can obtain the accurate system failure probability when the time segment is small. The improved method effectively compensates the calculation error of the traditional method, improves the calculation accuracy and computational efficiency of the results, and is suitable for the complex system with all kinds of common distribution.