概率信息不完全系统的可靠度分析方法

Reliability analysis method of incomplete probability information system

针对仅有少数样本数据的概率信息不完全系统的可靠度分析问题,提出一种概率信息不完全系统的可靠度分析方法.首先将概率信息不完全的变量通过立方正态变换近似得到完全概率信息,并对已知少量样本数据的变量建立基于线性矩相关系数的等效相关系数确定方法;然后结合自适应降维近似模型和直接积分法,实现功能函数的低阶统计矩估计;以此为基础,引入基于立方正态变换的可靠度分析矩方法,进行概率信息不完全系统的可靠度分析.算例分析结果表明:该方法能够通过变量的有限样本数据建立变量的近似完全概率信息,并适用于概率信息不完全系统的可靠度分析.

Aiming at the problem of reliability analysis for the incomplete probabilistic information with a few sample data,a reliability analysis method of incomplete probability information system was proposed.First,the approximatively complete probability information of the variable was obtained by cubic normal transformation,and the equivalent correlation coefficient of the variables with a few sample data was derived through correlation coefficient of L-moments.Then,the low order statistical moments of the performance function were calculated by combining the adaptive dimension-reduction model with the direct integral method,and the reliability analysis method for incomplete probability information system was proposed by introducing cubic normal transformation based moment method.Examples analysis results show that the proposed method can establish the approximatively complete probability information of the variable with a few sample data,and is suitable for the reliability analysis of the incomplete probability information system.