標題: 以具雙質量車輛模型識別橋梁頻率與頻譜分析
Two-Mass Vehicle Model for Extracting Bridge Frequencies and Spectral Analysis
作者: 陳柏豪
楊子儀
Chen, Bo-Hao
Yang, Judy P.
土木工程系所
關鍵字: 車橋互制;簧下質量;有限元素法;希爾伯特-黃轉換;Vehicle-bridge interaction (VBI);Unsprung mass;Finite element method;Hilbert-Huang transform (HHT)
公開日期: 2017
摘要: 在對於橋梁健康狀態進行評估與檢測時,橋梁頻率為其中一個重要參數,傳統上量測橋梁頻率的主要方法是透過直接量測法,即將數個感應器安裝於橋梁,透過量測橋梁各點之動態反應以計算橋梁頻率。有鑑於直接量測法費時費力,近二十年發展出間接量測法,透過車橋互制之理論,從量測車得到橋梁之反應。在車橋互制模型中,傳統之車輛模行僅考慮單一質量以代表整輛量測車,然量測車之簧下質量乃實際接觸地面之質量,故本研究提出具雙質量之車輛模型,以忠實反應橋梁之動力行為。在數值分析中,考慮橋梁鋪面粗糙度存在之情況,進一步探討簧下質量對於橋頻識別之影響,以期做為未來量測車設計與選用之參考;在頻譜分析部分,除了以傳統的FFT(fast Fourier transform)頻譜進行橋梁峰值識別外,亦使用HHT(Hilbert-Huang transform)處理頻譜訊號,以比較兩者對於擷取橋頻之差異與可行性,另外進一步將訊號以平均方式處理,以將粗糙度所造成之雜訊影響降低,提高橋頻之識別。
To assess or monitor the health of bridges, the most important parameter is the bridge frequency. Traditionally, the direct approach is the main way to measure bridge frequencies, which relies the installation of several sensors on a bridge and calculates bridge frequencies by measuring dynamic responses obtained at sensor locations. Since the direct approach wastes time and manpower, the indirect approach was proposed and developed over the past twenty years. Through the theory of vehicle-bridge interaction (VBI), the bridge responses can be obtained from the dynamic responses of the test vehicle. In the model of VBI, the entire test vehicle is simulated as a single mass conventionally. However, the unsprung mass of the test vehicle is the actual component in direct contact with the ground. As such, a two-mass vehicle model is proposed to truly reflect the dynamic behavior of the bridge in this study. In the numerical analysis, for the bridge pavement subject to roughness condition, the effects of unsprung mass on the identification of bridge frequencies is investigated, which serves as a reference in designing the test vehicle in the future. Concerning the spectral analysis, both FFT(fast Fourier transform) and HHT(Hilbert-Huang transform) are applied to process the spectrum signals. The difference between the two methods and the feasibility of the methods are discussed. Furthermore, the signals are processed by averaging to reduce the effect of road surface roughness and enhance the identification of bridge frequencies.
URI: http://etd.lib.nctu.edu.tw/cdrfb3/record/nctu/#GT070451228
http://hdl.handle.net/11536/142515
顯示於類別:畢業論文