標題: 以EMD識別橋梁基本振動頻率並探討移動車輛阻尼、質量、速度之影響
Damping, mass, and velocity effects of a passing vehicle for indirectly measuring the first bridge frequency by EMD technique
作者: 李唯鈞
楊子儀
Lee, Wei-Chun
土木工程系所
關鍵字: 車橋互制;經驗模態分解法;總體經驗模態分解法;有限元素法;Vehicle-Bridge Interaction (VBI);Empirical Mode Decomposition (EMD);Ensemble Empirical Mode Decomposition (EEMD);Finite Element Method
公開日期: 2016
摘要: 在橋梁養護管理作業中,牽涉橋梁動態資訊之掌握,其中以橋梁頻率是最重要的動態參數。傳統上橋梁頻率之量測以直接量測法為主,仰賴固定於橋梁結構上之感測器以量測橋梁之振態反應,然而缺點為感測器安裝費時費力、量測系統之高維護成本、龐大量測數據之處理以及感測器受限在單一橋梁之使用等。有鑑於此,發展出間接量測法,以具機動性之量測車記錄通過橋梁之車輛反應以進行橋梁之動態分析。本研究使用1990年代發展之車橋互制理論,以單自由度之彈簧質量模擬量測車,並以有限元素法進行求解,車輛前進時之垂直慣性力與彈簧力以接觸力方式傳遞到橋梁造成振動,由於車橋耦合之故,橋梁振動將再次影響量測車之垂直振動,故透過分析量測車之頻譜即可得橋梁之頻率與振型,然量測車之行為受許多因素影響,諸如車流、地面粗糙度與車輛阻尼等,如何降低車輛頻譜中之雜訊反應以忠實反應出橋梁之行為,為當前研究之重要課題。本研究探討車輛阻尼對於橋梁動態行為量測之影響,在不同鋪面粗糙度下,透過車輛阻尼之放置以降低車頻振幅,進而識別出橋梁的頻率,此外,為提升頻譜識別之效果,採用經驗模態分解法與總體經驗模態分解法分別進行頻譜分析。
The indirect approach for measuring bridge frequencies from a passing vehicle has become one prosperous method recently in that it provides mobility and liability to set up sensors while it is relatively cheap to maintain the instruments. Following the theory of vehicle-bridge interaction (VBI) developed in the 1990s, the vehicle is modeled as a sprung mass moving over a simply supported beam, and the VBI system is solved by the finite element method. The contact force induced by the vehicle causes the vibration of the bridge while bridge vibration turns back to the vehicle. Therefore, one can identify the bridge frequencies from the spectral analysis of the vehicle. However, there exist many factors affecting identification of bridge frequencies such as existing traffic, road surface roughness, and damping of vehicle. How to effectively reduce the noises in the frequency spectra is of urgent need. This study investigates the effect of vehicle damping on the identification of bridge frequency while considering road surface roughness. To further enhance the results of identification, both the empirical mode decomposition (EMD) and ensemble empirical mode decomposition (EEMD) are adopted in the spectral analysis.
URI: http://etd.lib.nctu.edu.tw/cdrfb3/record/nctu/#GT070251238
http://hdl.handle.net/11536/139665
Appears in Collections:Thesis