軌道運輸具有便捷、快速等特性,已成為民眾最依賴的大眾運輸工具。然而,現今列車的行駛速度越來越快,若軌道結構系統稍有損傷,將有可能造成重大災情,更加突顯軌道結構系統安全維護的重要性。 本文應用希爾伯特-黃轉換來分析鋼軌的振動訊號,此理論的創建“經驗模態分解法”具有完整性、幾乎正交性及可適性。可適性可表達原訊號之物理特性,藉以處理非線性及非穩態的訊號。因為這種方法的特性,再加上路堤段、隧道段的鋼軌振動訊號相當複雜,屬於非線性且非穩態的訊號,因此本研究主要探討希爾伯特-黃轉換應用於解析現地鋼軌振動訊號之可行性與適用性,並由頻譜圖的結果來檢視軌道結構系統的完整性。 本研究於分析過程中,加入了正交性指數的檢核,增進經驗模態分解結果的正確性。研究結果顯示,分解的效果十分良好,且藉由希爾伯特-黃轉換可確實解析現地鋼軌的振動訊號,獲得鋼軌振動頻率,並由頻譜圖中獲知扣件遭鬆脫後,鋼軌振動頻率變化的情況。
The track transportation has convenient and rapid properties. Today, it has already become a part of indispensable in communities popular. Hence, it accentuates the importance of railway system maintenance. In this article, by using series of Hilbert-Huang transform to analyze the vibration signal of rail, this data analysis method, based on the empirical mode decomposition, have completeness、orthogonal and adaptively. In terms of adaptively, this method can analyze nonlinear and non-stationary signal. Based on this method properties, the rail signal of embankment and tunnel segment belong to nonlinear and non-stationary signal is very complicated, consequently, this research mainly inquires into Hilbert-Huang transform to apply in analyze rail vibration signal in situ of the feasibility and the applicability, and by the result of spectrum to examine the integrity of railway structure system. Adding the orthogonal index to check the decomposed result in this research, improve the accuracy of empirical mode decomposition result. In terms of result, the effect of decomposition is goodness, and by Hilbert-Huang transform can really analyze the vibration signal of rail in situ.