濾波技巧於橋梁頻率間接量測法之應用

Application of Filtering Techniques to Indirect Measurement of Bridge Frequencies

李育謙(Yu-Chien Li)

指導教授 ： 楊永斌

國立臺灣大學/工學院/土木工程學研究所/碩士(2011年)

https://doi.org/10.6342/NTU.2011.02189

摘要

橋梁頻率是一個非常重要的橋梁動力參數，不論是橋梁的破壞檢測、安全監測甚至數值模型的更新，均需以橋梁頻率為基礎。傳統橋梁量測方式是直接將感應器安裝在橋梁上，經由感應器來記錄橋梁的動力反應，最後再透過分析所記錄之資料，以得到橋梁的頻率，這樣的橋頻量測方式，稱之為橋頻直接量測法。近年來，有一種與前者迥異的橋頻量測法被提出，其概念是採用移動車輛通過橋梁時，利用安裝在車體上的感應器，來記錄車體的動力反應，間接地從車體動力反應中擷取出橋梁頻率，此種方法亦稱之為間接量測法。間接量測法相較於直接量測法，有著經濟、快速與可攜式之優勢，且當天然災害(如地震)發生後，可更快速地將所獲得橋梁頻率，並提供給工程師作為模型更新及安全評估之參考依據。
本研究之目的在於從理論與數值模擬的角度，來探討路面粗糙度對橋量頻率間接量測法之影響，與利用濾波技巧和隨機子空間識別法，來提升橋頻間接量測法之品質，並初步探討利用希爾伯特轉換(Hilbert Transform，HT)從移動車輛反應中擷取橋梁模態振型的可行性。而在此所採用的濾波工具有著傳統帶通濾波器、奇異譜分析(Singular Spectrum Analysis，SSA)以及以帶通濾波器為基底的奇異譜分析(Band Pass-filter based Singular Spectrum Analysis，BP-SSA)。根據本文各章的實例分析，吾人將針對各個方法和效應提出具體之結論。

關鍵字

橋梁頻率 ； 橋頻間接量測法 ； 路面粗糙度 ； 希爾伯特轉換 ； 奇異譜分析 ； 隨機子空間識別法

並列摘要

Bridge frequency is an essential dynamic parameter in bridge engineering. It is one of the parameters that have been used in the areas of structural damage detection, structural health monitoring and structural model updating.
Traditionally, the bridge frequencies are measured from the dynamic response of a bridge via the vibration sensors directly mounted on the bridge. Such an approach is referred to as the direct approach.
In recent years, another concept is proposed for measuring the bridge frequencies from the dynamic response recorded of a vehicle during its passage over the bridge of concern. Such an approach is called the indirect approach.
Compared with the direct approach, the indirect approach has the advantages of being more economic, fast, and movable. Whenever a natural disaster (e.g., an earthquake) occurs, this approach can quickly provide the information of bridge frequencies to field engineers for preliminary safety assessment.
The objective of this study is to investigate the effect of road surface roughness from a theoretical point of view via numerical simulation, to refine the indirect approach for extracting the bridge frequencies by filtering techniques and Stochastic Subspace Identification technique, and to investigate the feasibility of extracting bridge mode shapes from the dynamic responses of a vehicle via Hilbert transform. Furthermore, several signal filtering techniques are adopted in this study: the traditional band-pass filter, Singular Spectrum Analysis and band pass-filter based Singular Spectrum Analysis. Various numerical examples will be conducted using each technique with the conclusions drawn.

並列關鍵字

bridge frequencies ； indirect approach for measuring the bridge frequencies ； road surface roughness ； Hilbert transform ； Singular Spectrum Analysis ； Stochastic Subspace Identification technique

參考文獻

吳峰賓(民國99年)，「考慮路面粗糙度與車輛非線性勁度於橋頻間接量測法之模擬分析」，國立台灣大學土木研究所碩士論文，楊永斌教授指導。

陳韋帆(民國99年)，「橋梁頻率間接量測法之實驗研究」，國立台灣大學土木研究所碩士論文，楊永斌教授指導。

Alonso, F.J., Del Castillo, J.M., Pintado, P. (2004), “Application of singular spectrum analysis to the smoothing of raw kinematic signals,” Journal of Biomechanics, 38: 1085-1092.

Abdel-Ghaffer, A.M. and Scanlan, R.H. (1985), “Ambient vibration studies of Golden Gate Bridge: I. Suspended structure,” Journal of Engineering Mechanics, 111: 462-482.

Broomhead, D.S. and King, G.P. (1986), “Extracting qualitative dynamics from experimental data,” Physica D, 20: 217-236.