基于解析模式分解和随机减量技术的桥梁模态参数识别  

Analytical modal decomposition-based random decrement technique for modal parameters identification of bridge

作　　者：杨勇 宁平华 王晟 华旭刚[2] 温青[2,3] YANG Yong;NING Pinghua;WANG Sheng;HUA Xugang;WEN Qing(Guangzhou Municipal Engineering Design&Research Institute,Guangzhou 510095,China;Key Laboratory for Wind and Bridge Engineering of Hunan Province,Hunan University,Changsha 410082,China;Department of civil engineering,Hunan University of Science and Technology,Xiangtan 411201,China)

机构地区：[1]广州市市政工程设计研究总院有限公司,广东广州510095 [2]湖南大学风工程与桥梁工程湖南省重点实验,湖南长沙410082 [3]湖南科技大学土木工程学院,湖南湘潭411201

出　　处：《铁道科学与工程学报》2022年第5期1355-1363,共9页Journal of Railway Science and Engineering

基　　金：国家自然科学基金资助项目(21708208);湖南省自然科学基金资助项目(2021JJ40208)。

摘　　要：大跨异型桥梁模态振型耦合严重,环境振动响应频率分布密集,直接利用随机减量技术(RDT)处理这类结构环境振动响应信号时,难以从随机减量信号中识别多模态的参数。由于解析模式分解(AMD)能有效将模态密集分布的振动信号分解为一系列单模态子信号,故结合解析模式分解和随机减量技术开展大跨异型桥梁模态参数识别研究。首先,对环境振动响应信号进行解析模式分解,获得一系列单模态子信号;然后,对每个子信号进行随机减量技术处理,提取随机减量信号,克服随机减量技术处理多模态信号的缺陷;最后,采用最小二乘法按照单自由度系统自由振动函数从随机减量信号中识别模态频率和阻尼比。通过三自由度振动系统仿真分析,研究该方法的可行性,分析随机减量技术参数对模态参数识别结果的影响。采用该方法识别了一座大跨曲线主梁人行拱桥的前4阶模态的频率和阻尼比,并与随机子空间法识别结果进行对比。结果结果表明:该方法可从多模态环境振动响应中有效识别结构的频率和阻尼比;阈值常数取1.5倍信号的均方根值、随机减量信号长度取10倍模态周期能更准确地识别固有频率和阻尼比。本文提出的方法能有效识别真实结构的模态参数,识别的前3阶竖弯模态阻尼比约0.45%,可为类似桥梁的阻尼比取值提供参考。Due to seriously coupling of the mode shape for the large-span special-shaped bridge,the frequency component of ambient vibration response is closed spaced,which leading to difficulty to identify several model parameter from the random decrement signal directly estimated by random decrement technology.Analytical modal decomposition can effectively decompose this signal into single-mode component sub-signals.Therefore,an analytical modal decomposition-based random decrement technique(AMD-based RDT)was presented to identify natural frequencies and damping ratios from ambient vibration response measured on large-span specialshaped bridge.First,the low-pass filtering measurement data was decomposed into a series of single-mode component sub-signals by AMD.Then,the random decrement technique was employed to extract single-mode random decrement signature from each sub-signal,coming over the disadvantage of RDT on dealing with the signal with closed spaced frequency components.Final,the natural frequencies and damping radios were estimated by fitting each random decrement signature to the theoretical free decay vibration function of one degree-of-freedom system by least square technique.A three-degree-of-freedom vibration system with closely spaced frequencies is simulated to optimally select the bisecting frequency,initial displacement,and duration of time.It demonstrates that the accuracy of the proposed method in identifying natural frequencies and damping ratios is remarkable when the bisecting frequency is the center of the two nearby frequencies.The threshold level of the initial displacement is 1.5 times of root mean square of sub-signal,and the duration of time is 10 times of the period of vibration of the sub-signal.The natural frequencies and damping radios of a curved steel-box girder arch bridge are identified by this method,demonstrating that which is viable in practical application.The identified modal damping radio for the first three vertical mode are about 0.45%,which can provide a reference for the sim

关 键 词：解析模式分解 随机减量技术 参数识别 最小二乘法 

分 类 号：U446.1[建筑科学—桥梁与隧道工程]