應用無線感測儀器與微振試驗評估橋梁基礎淘刷之研究

Abstract

隨著氣候變遷的加劇，極端氣候對基礎設施的破壞亦愈行嚴重。其中，橋梁基礎的淘刷攸關橋梁安全甚鉅，但頻繁發生的特大暴雨對橋梁基礎造成沖刷，嚴重影響橋梁安全。與傳統的有線或水下測量儀器不同，透過量測橋梁上部結構的動力參數，可以反映橋墩基礎與土壤互制系統的特性，間接推估橋梁基礎淘刷的嚴重程度，同時也可避免儀器於颱風季節遭受損壞。惟測量所得之動力參數可能受到不明來源的外在環境干擾，而以快速傅立葉轉換(Fast Fourier Transformation，FFT)處理所得之加速度數據後，有錯誤識別橋梁自然頻率的風險。因此本研究嘗試以短時傅立葉轉換（short-time Fourier transform，STFT）延伸發展出的方法去除測量所得加速度數據中之干擾因素，以更正確且穩定地識別橋梁的自然頻率。 本研究採用無線感測儀器、MMF KB12VD加速度計進行微振量測試驗，後續則使用Arduino Uno介面控制版、Xbee無線傳輸模組等開源軟體(Open Source)，進行資料蒐集與傳輸，再以MATLAB 作為資料處理及示波器，將測得之加速度值進行數位濾波並嘗試以「短時傅立葉轉換沿時間軸加總法」 (STFTCATS)，取代快速傅立葉轉換(FFT )識別橋梁的自然頻率。本研究選取了臺北市基隆河上的成美長壽橋、南湖大橋及景美溪上的萬壽橋所量測到的微振訊號加以分析，確認「短時傅立葉轉換沿時間軸加總法」(STFTCATS)相對於快速傅立葉轉換(FFT )較能識別橋梁自然頻率，且較為穩定。 本研究另選取了基隆河上的南湖大橋以有限元素法SAP2000軟體建立模型，並以土壤彈簧模擬土壤與基礎間的互制行為，藉著逐層解除基礎土壤彈簧的束制，獲得橋梁自然頻率與基礎淘刷深度關係圖。結果顯示，橋梁之自然頻率確隨淘刷之深度而降低。經由適當地設定邊界條件，模擬與現場實測之自然頻率差異約為+0.19%，模擬與現場實測之淘刷深度差異約為+5.50%。顯示所獲得之自然頻率-淘刷深度關係圖應具實用性。 本研究所提供之基礎淘刷評估適用於災害發生之前後，用以與基礎不受淘刷時之自然頻率做一比較以作為預警或災後復舊之參考。另宜配合基樁耐洪能力分析和極限承載能力之評估，以訂定適當之頻率警戒值。

As climate change becomes more severe, so infrastructure is being damaged more by extreme weather. In particular, extreme rainstorms are occurring more often and are scouring the foundations of bridges, seriously affecting their safety. Unlike traditional wired or underwater measuring instruments, measuring the dynamic parameters of the bridge superstructure can reflect the characteristics of the soil-foundation interaction and indirectly estimate the scour of bridge foundation and so avoid the device being destroyed during typhoon season. However, the measured dynamic parameters may be affected by external environmental interference from unknown sources, and there is a risk of misidentifying the natural frequency of the bridge after processing the acceleration data obtained with Fast Fourier Transformation (FFT). Therefore, this study attempts to use a method developed by the short-time Fourier transform (STFT) to remove interference factors in the measured acceleration data to more accurately and stably identify the natural frequency of the bridge. In the study, a wireless sensing instrument was used to measure bridge ambient vibration, it uses an accelerometer and an open-source Arduino Uno microcontroller board and XBee wireless transmission module for data collection and transmission, with MATLAB used for the subsequent data processing, including digital filtering and identifying the natural frequency of the bridge using the short-time Fourier transform cumulate along time sequence (STFTCATS) method to replace the Fast Fourier Transform (FFT). This study selected the ambient vibration tests by the Chengmei Changshou Bridge on the Keelung River and the Wanshou Bridge on the Jingmei River in Taipei City for analysis, and confirmed that STFTCATS is better than FFT in identifying the natural frequency of bridges and is more stable. This study also selected the Nanhu Bridge on the Keelung River to build a model using the finite element method SAP2000 software, and used soil springs to simulate the interaction between the soil and the foundation. By releasing the restraints of the foundation soil springs layer by layer, the simulated scour depth–natural frequency relationship was obtained. The results show that the natural frequency of the bridge does decrease with the scour depth. By appropriately setting the boundary conditions, the difference in natural frequency between simulation and field measurement is approximately +0.19%, and the difference in scour depth between simulation and field measurement is approximately +5.50%. Displaying the simulated scour depth–natural frequency relationship should be practical. The scour assessment provided by this study is applicable before and after a disaster, and can be compared with the natural frequency when the foundation is not scoured as a reference for early warning or post-disaster recovery. In addition, it should be combined with flood-resisting capacity analysis and the assessment of ultimate bearing capacity of foundation piles to determine appropriate frequency warning standard.