复杂高层转换层结构的智能选型  

Form-Intellectual Selection of Transfer Story Structure in Complicated Tall Building Based on Artificial Neural Network

作　　者：林煌斌[1] 王全凤[1] 张云波[1] 

机构地区：[1]华侨大学土木工程学院,福建泉州362021

出　　处：《华侨大学学报（自然科学版）》2008年第3期427-431,共5页Journal of Huaqiao University(Natural Science)

基　　金：国家自然科学基金资助项目(50578066);福建省自然科学基金资助项目(E0410023)

摘　　要：分析复杂高层建筑结构转换层选型的主要控制因素,建立基于人工神经网络的复杂高层建筑结构转换层选型的数学模型.有针对性地收集国内外62个带转换层复杂高层结构工程实例,分别采用BP(Back-Propagation)算法和L-M(Levernberg-Marquart)算法进行复杂高层建筑结构转换层选型的研究.设计一种基于人工神经网络的复杂高层建筑转换层结构的智能选择,使单一设计人员通过向人工神经网络输入若干复杂高层建筑结构转换层的基本信息,便可在方案阶段解决高层建筑结构转换层结构型式选择.The primary control factors in structure selection of transfer story in complicated tall building are analyzed, and the mathematical model is established based on artificial neural network （ANN）. 62 domestic or foreign tall buildings with transfer story are collected, using traditional back propagation （BP） algorithm and Levernberg-Marquert （L-M） al- gorithm respectively, the complicated tall buildings with transfer story are investigated. The intelligent model to select the structure of transfer story in tall building is established. The model can help the engineers obtain the optimal structure of transfer story by inputting the essential information into artificial neural network.

关 键 词：高层建筑 转换层 结构选型 人工神经网络 

分 类 号：TU973.1[建筑科学—结构工程]