摘要
为了剔除稳定图中的虚假模态和避免模态遗漏现象,提高模态参数识别的精确度,提出了一种基于滑动窗口和相似度的桥梁结构模态参数智能化识别算法。基于余弦相似原理提出了频率相似度和振型相似度,并依此构建置信度,以实现对系统真实阶次的自动化确定;引入改进集成经验模式分解算法,以消除响应信号内部的噪声信号,达到消除部分虚假模态的目的;接着引入滑动窗口以实现对响应信号的划分,并通过构建频率相似度、振型相似度以及阻尼比相似度实现多个窗口对应参数结果中同类模态的聚类处理,达到剔除虚假模态和避免模态遗漏的目的。最后将所提算法运用于实际斜拉桥结构的模态参数识别,并将识别结果与现场试验值以及有限元结果进行对比,结果表明,所提算法不仅能有效识别出频率结果还能识别出准确的模态振型图,能够实现桥梁结构模态参数的在线智能化识别。
In order to eliminate the false mode in the stability diagram and avoid the modal missing phenomenon,so that to improve the accuracy of modal parameter identification,an intelligent recognition algorithm which is based on recursive window and similarity was proposed for bridge structure modal parameters.Firstly,the frequency similarity and mode similarity were proposed based on the cosine similarity principle,and in accordance with them,the confidence was constructed to realize the automatic determination of the real order of the system.Secondly,the improved integrated empirical mode decomposition algorithm was introduced to eliminate the noise inside the response signal and to achieve the purpose of eliminating some parts of false modes.And then the recursive window was introduced to realize the division of the response signal,and the clustering processing of similar modalities in multiple window corresponding parameter results was realized by constructing the frequency similarity,the mode shape similarity and the damping ratio similarity.The clustering process achieves the purpose of eliminating false modes and avoiding modal missing at the same time.Finally,the proposed algorithm was applied to the real cable-stayed bridge structure,and its identification results were compared with the field test values and the finite element results.The results show that the proposed algorithm can not only effectively identify the frequency results but also identify the accurate modal shape patterns,that is,the proposed algorithm can realize on-line intelligent identification of modal parameters of bridge structures.
作者
陈永高
钟振宇
CHEN Yonggao;ZHONG Zhenyu(School of Civil Engineering and Architecture,Zhejiang Industry Polytechnic College,Shaoxing 312000,China;College of Civil Engineering and Architecture,Zhejiang University,Hangzhou 310058,China)
出处
《振动与冲击》
EI
CSCD
北大核心
2020年第16期196-204,共9页
Journal of Vibration and Shock
基金
浙江省教育厅科研项目(Y201941529)。
关键词
桥梁结构
智能化识别
随机子空间算法
模态参数
系统阶次
相似度
滑动窗口
bridge structure
intelligent identification
stochastic subspace identification
modal parameter
system order
similarity degree
recursive window
