点击次数：

论文类型：期刊论文

发表时间：2022-06-28

发表刊物：工程力学

卷号：30

期号：1

页面范围：454-462

ISSN号：1000-4750

摘要：In practical engineering applications, components of a steel frame are
   selected from a standard steel library according to the design code for
   steel structures. The cross-section sizes of components in a frame are
   non-continuous, thusly the optimized selection of standard cross-section
   types is a typical discrete optimization problem. When the number of
   cross-section types is large, using heuristic-based algorithms (such as
   genetic algorithm, etc) results in inefficiency of resolving this
   problem. In this paper, the high-dimensional Lagrange interpolation
   functions are introduced to make the discrete design problem be
   continuous. The model of cross-section selection optimization is
   formulated, and it can be solved by gradient-based method. In this
   model, design variables are reduced obviously, and only n variables are
   needed to describe 2n types of cross-sections. Compared with genetic
   algorithm, the numerical examples show that the calculation efficiency
   can be improved 1-2 orders of magnitude by the method proposed, and less
   cross-section types can be obtained under the condition of roughly equal
   performance with genetic algorithm, which satisfies the engineering
   applications well.

备注：新增回溯数据