改进的区间截断法及基于区间分析的非概率可靠性分析方法

A advanced interval-truncation approach and non-probabilistic reliability analysis based on interval analysis

从工程应用的角度出发 ,提出了用区间表示参数的不确定性时 ,线性系统的非概率可靠度指标 ,此可靠度指标在试验数据较少时比概率可靠度指标更合理 ;另外本文还提出了计算响应参量变化范围的区间截断法 ,研究了截断参数 t的选取范围 ,算例分析表明 ,当 t取为所有输入参数的最大相对变化量时 ,由区间截断法算得的结果近似等于精确解 ,而随着 t的增大 ,区间截断法的解逐渐平稳地趋近于由区间算术运算所求得的直接解。

From the view of engineering application, the non\|probabilistic reliability index of linear system is presented on the interval description of uncertainty parameter. And the non-probabilistic reliability index is more rational than the probabilistic one, as the experiment data are unavailable to set up the probability density function. Furthermore, an interval-truncation approach is investigated for computing the interval of response variable, and the selection of truncation parameter t is illustrated. As t takes a value that is equal to or little bigger than the maximum relative deviation of the input parameters to the corresponding central values, a solution with high precision can be obtained. The solution of the truncation interval is convergent to that calculated by the direct interval arithmetic method as the increase of t.