随机模型参数不确定时结构失效概率的区间估计

在缺乏足够数据或信息不完整的情况下，给出失效概率的界限，比给出一确定的单值具有更高的可信度。本文用区间变量描述随机可靠性模型参数的不确定性，提出了考虑随机模型分布参数不确定性时，机械结构失效概率的区间估计方法。实际算例表明：结构的失效概率对随机模型参数的偏差非常敏感。说明在进行可靠性建模时，对模型的可靠性必须给予足够的重视。

Interval Estimation of Failure Probability of Structures In Uncertain Parameters of Stochastic Model

Uncertainties present everywhere in the physical and mathematical models. In most cases, it is difficult to describe the performance of systems using a precisely single number. But it is easy to describe the condition by using a range or an interval to represent it. In the reliability computation, when insufficient accurate data are available or the information is incomplete, a range representation of probability of failure may be more confident than the representation by a single number. In this paper, uncertainties of the parameters of stochastic model of reliability are described by intervals. A procedure is presented for the interval estimation of failure probability of mechanical structures when the distribution parameters are uncertain. It is shown by a numerical example that the probability of failure may be very sensitive to the possible errors of stochastic model. So enough attention must be paid to the reliability of stochastic modeling.