讨论了广义随机空间内随机参数服从正态分布情况下汽车零部件可靠性优化设计问题,在基本随机参数均值、方差和相关系数已知的情况下,应用可靠性指标计算的优化方法与优化技术相结合,避免了相关变量的独立变换以及求极限状态函数对基本变量的偏导数,并通过计算机程序给予实现,迅速准确地得到了汽车零部件的可靠性优化设计信息。计算结果表明,文中提出的方法是一种有效、实用的方法。
In this paper, the reliability optimization design of vehicle components is presented under the condition that the random parameters in generalized random space follow normal distribution. When the mean, variance and correlation coefficient of basic random parameters are known, applying optimization technique in reliability index calculation can avoid the orthogonal transformation on correlated variables and the calculation of partial derivatives of limit state function to basic variables, and hence greatly reduce the computation efforts. A corresponding program is developed for implementing the scheme so that the reliability optimization design information of vehicle components can rapidly and accurately obtained. Some examples are given, which demonstrate the effectiveness and practicality of the proposed method.