摘要
对于材料性质和载荷具有不确定性结构进行疲劳寿命估计时,结构疲劳寿命往往是这些不确定性变量的函数.以凸分析和区间数学为理论基础,将这些不确定变量用椭球和区间定量化,基于Taylor级数展开,提出了近似估计结构疲劳寿命的非概率集合理论模型—凸模型方法和区间分析方法.它们克服了概率方法需要预先知道不确定变量的概率分布密度或大量统计数据的局限性,并且计算量小.通过数值算例,将凸模型方法、区间分析方法与概率方法进行了比较研究,数值计算结果表明了这两种非概率方法对线性及非线性形式的结构寿命估计均能提供令人相当满意的精度.
Fatigue lifetime of a structure with uncertainties in material properties and external loading is estimated. Fatigue lifetime of a structure, in general, is the function of these uncertain factors. Based on convex analysis and interval mathematics, quantifying the uncertainties as ellipsoidal and interval numbers, two new non-probabilistic set-theoretical models, which approximately estimate the fatigue lifetime through first-order Taylor series, are presented. The presented two methods not only overcome the shortcoming that probabilistic method needs a large number of data in advance, but also save the computation time. Numerical examples illustrate that the two methods can provide sufficient accuracy as probabilistic method for any kinds of lifetime functions, including linear or nonlinear.
出处
《固体力学学报》
CAS
CSCD
北大核心
2006年第1期91-97,共7页
Chinese Journal of Solid Mechanics
基金
国家自然科学基金委与中国工程物理研究院联合基金项目(10376002)
航天支撑技术基金项目(20032.2BH03)
国家杰出青年科学基金项目(10425208)资助
关键词
疲劳寿命估计
凸模型
区间分析
TAYLOR级数
不确定性
fatigue lifetime estimation, convex model, interval analysis method ,Taylor series, uncertainty
