基于SOFM实例推理系统的结构选型
SELECTION OF TARGET STRUCTURES BASED ON SELF-ORGANIZING-FEATURE-MAP
-
摘要: 本文以大跨空间结构和高层建筑结构两类大型复杂结构型式优选为研究内容,建立了基于自组织竞争神经网络(SOFM)检索机制的实例推理系统,并应用于结构选型决策;采用面向对象的设计思想建立方案集,采用自组织竞争神经网络进行实例检索和匹配,形成了一套有效的计算推理体系.最后给出一个高层建筑结构方案生成的算例,说明系统的工作过程.Abstract: This paper mainly discusses the selection of target structures for large scale complicated engineering structures such as high-rise buildings and long-span spatial structures. A case-based-reasoning system based on self-organizing-feature-map(SOFM)is put forward to solve this decision problem. The case library is built by object-oriented method, and the desired structural styles are searched or matched by SOFM. An example, inwhich a high-rise building is studied, illustrates the working process of the system.