计算机与数字工程2019，Vol.47Issue(3)：508-512,5.DOI:10.3969/j.issn.1672-9722.2019.03.004
鲁棒独立分量分析在结构损伤特征提取中的应用
Application of Robust Independent Component Analysis in Structural Damage Feature Extraction
摘要
Abstract
A new ICA algorithm Robust Independent Component Analysis(RobustICA)is applied to the four-story steel frame scale model of the IASC-ASCE to extract modal natural frequencies in order to extract the damage modal frequencies of civil structures accurately. Firstly,the detection nodes near the excitation points are selected,then the independent components are ob?tained by blind source separation using RobustICA algorithm,and then the modal natural frequencies under different conditions are determined by spectrum analysis. The results show that the independence among the components separated by RobustICA algorithm is higher and the natural frequencies can be separated more precisely.关键词
IASC-ASCE结构模型/固有频率/RobustICA算法/盲源分离/频谱分析Key words
IASC-ASCE structure model/natural frequency/RobustICA algorithm/blind source separation/spectrum anal⁃ysis分类
建筑与水利引用本文复制引用
徐先峰,张华竹,段晨东..鲁棒独立分量分析在结构损伤特征提取中的应用[J].计算机与数字工程,2019,47(3):508-512,5.基金项目
国家自然科学基金项目"利用参量结构实现复杂信号环境下盲信号分离方法研究"(编号:61201407) （编号:61201407）
陕西省自然科学基础研究计划项目"单通道盲源分离应用与机械系统故障诊断关键技术研究"(编号:2016JQ5103) （编号:2016JQ5103）
长安大学中央高校基金项目"机械系统故障诊断病态问题研究"(编号:300102328202)资助. （编号:300102328202）
