基于向量投影取样的改进加权响应面法

发布时间：2018-03-09 12:39

本文选题：响应面法　切入点：加权响应面法　出处：《科学通报》2017年17期 　论文类型：期刊论文

【摘要】：在加权非线性响应面法的基础上,通过向量梯度投影方法改进样本点的选择,提出一种基于向量投影取样的改进加权响应面方法.用向量梯度投影法获取新的试验样本点,再对这些取样点进行加权回归处理,赋予靠近极限状态曲面的样本点更高的权重来构造和更新迭代二次响应面函数,解决隐式极限状态函数的结构可靠性分析问题.该方法有效克服了传统响应面法计算结果受插值系数的影响,一定程度上提高了计算精度,可以得到相对更优的结果.算例分析验证了所提方法的合理性和有效性.
[Abstract]:On the basis of weighted nonlinear response surface method and vector gradient projection method, an improved weighted response surface method based on vector projection sampling is proposed, and a new test sample point is obtained by vector gradient projection method. Then these sampling points are weighted by regression, and the sample points close to the limit state surface are given a higher weight to construct and update the iterative quadratic response surface function. The problem of structural reliability analysis of implicit limit state function is solved. This method effectively overcomes the influence of interpolation coefficient on the results of traditional response surface method, and improves the accuracy of calculation to a certain extent. A better result can be obtained and the validity and rationality of the proposed method are verified by an example.
【作者单位】：空军工程大学理学院;空军工程大学航空航天工程学院;
【基金】：国家自然科学基金(51175510) 空军工程大学研究生创新基金资助
【分类号】：TB114.3

【相似文献】