摘要
介绍了基于几何原理求解可靠指标、寻找设计验算点的具有较高实用价值的两种新算法： 图解渐近法和Monte Carlo递进法。推导了由标准正态变量反算常用分布类型的随机变量数值的计算公式，并且通过采取某些改善措施，简化了计算过程，大大提高了问题的求解效率。同时，两种算法可以结合使用，经算例验证，能够快速解决问题且满足高精度要求，说明了本文方法的可行性和有效性。
Two new methods, diagrammatical approximation and Monte Carlo Iteration, are introduced to search for design point and acquire reliability index. The two methods are developed based on geometrical principles. Several formulas are presented to calculate random variables of distributed type in accordance with standard normal variables. Improvements are made to simplify the computation process and enhance the efficiency of solution. If the two methods are combined, the solution efficiency will be even higher. Numerical examples are provided to demonstrate the feasibility of the methods. It is shown that satisfactory accuracy of solution can be achieved.
出处
《工程力学》
EI
CSCD
北大核心
2002年第3期159-165,共7页
Engineering Mechanics
基金
国家自然科学基金资助项目(50178011)
