Optimization of Variable Blank Holder Force Trajectory via Sequential Approximate Optimization with Radial Basis Function network
Optimization of Variable Blank Holder Force Trajectory via Sequential Approximate Optimization with Radial Basis Function network
复制标题
DOI：
--
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
S. Kitayama;K. Kita;K. Yamazaki
中科院分区：
文献类型：
--
作者：
S. Kitayama;K. Kita;K. Yamazaki
1. 摘要 本文提出了带有径向基函数（RBF）网络的序贯近似优化（SAO）。在SAO中，采样策略是重要问题之一。 RBF 网络在整个提议的 SAO 中得到使用。为了找到未探索的区域，构造了称为密度函数的新函数。通过最小化密度函数，在未探索区域周围添加新的采样点。所提出的 SAO 应用于拉深中可变压边力轨迹的优化。 2.
1. Abstract This paper proposed sequential approximate optimization (SAO) with radial basis function (RBF) network. In the SAO, the sampling strategy is one of the important issues. The RBF network is used throughout the proposed SAO. In order to find the unexplored region, new function called the density function is constructed. By minimizing the density function, new sampling points are added around the unexplored region. The proposed SAO is applied to the optimization of the variable blank holder force trajectory in deep drawing. 2.