首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于随机森林的建筑结构损伤识别方法
引用本文:周绮凤,杨小青,周青青等.基于随机森林的建筑结构损伤识别方法[J].振动.测试与诊断,2012,32(2):197-201.
作者姓名:周绮凤  杨小青  周青青等
作者单位:1. 厦门大学自动化系 厦门,361005
2. 厦门大学建筑与土木工程学院 厦门,361005
基金项目:福建省自然科学基金资助项目,中央高校基本科研业务费专项资金资助项目
摘    要:针对利用分类器对建筑结构进行损伤识别的问题,引入一种新的组合分类器算法——随机森林,提出基于小波包分解和随机森林的结构损伤识别方法。首先,采用小波包对结构在不同损伤程度和位置上的振动加速度信号进行分解,得到各个频带上的总能量;然后,利用各频带上能量值存在着差异性作为输入到分类器的特征向量;最后,训练随机森林模型并对建筑结构的损伤位置和损伤程度进行识别。应用该方法对一座8层剪切型钢框架结构进行损伤判别,并与BP神经网络和支持向量机方法进行对比,结果表明该方法具有较好的识别精度与稳定性。

关 键 词:损伤识别  随机森林  小波包分解  加速度信号

Damage Identification Technique for Building Structure Based on Random Forest
Zhou Qifeng,Yang Xiaoqing,Zhou Qingqing,Lei Jiayan.Damage Identification Technique for Building Structure Based on Random Forest[J].Journal of Vibration,Measurement & Diagnosis,2012,32(2):197-201.
Authors:Zhou Qifeng  Yang Xiaoqing  Zhou Qingqing  Lei Jiayan
Institution:1.Department of Automation,Xiamen University Xiamen,361005,China)(2.School of Architecture &Civil Engineering,Xiamen University Xiamen,361005,China)
Abstract:For the use of classifier in the structure damage diagnosis,a structural damage identification method is proposed based on wavelet packet decomposition and random forest which is a new combination classifier algorithm.Wavelet package decomposition is used to decompose the vibration acceleration signals of building structure with different damage degrees and locations.The energy sequences at different bands of frequencies decomposed by the wavelet packet decomposition are inputted to classifier as feature vectors.A random forest model is trained and used to identify the location and degree of injury.The method is applied for damage identification of an eight-story shear steel frame model.Experimental results show that the method has good recognition accuracy and stability compared with BP neural network and support vector machines.
Keywords:damage identification  random forest  wavelet package decomposition  acceleration signal
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《振动.测试与诊断》浏览原始摘要信息
点击此处可从《振动.测试与诊断》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号