论文部分内容阅读
该文推荐了一种基于可靠度的优化设计方法,该方法利用支持向量回归机进行两级响应面近似策略,将双层优化问题转换为一个单层优化问题,第一级称为分析响应面,用来拟合极限状态方程的近似表达式,第二级称为设计响应面,用来拟合设计变量和失效概率的关系式,然后借助一种全局的优化方法——粒子群算法求解该确定性的优化问题,该方法的有效性可以从三脚架基础算例中验证。
In this paper, an optimal design method based on reliability is proposed, which uses support vector regression machine to carry out two-level response surface approximation strategy, which transforms the double-layer optimization problem into a single-layer optimization problem, the first is called the analysis response surface , Which is used to fit the approximate expression of the equation of state of the limit. The second level is called the design response surface, which is used to fit the relationship between design variables and failure probability. Then, a global optimization method, particle swarm optimization The deterministic optimization problem, the validity of this method can be verified from the base tripod example.