TDR錯動變形物理模型與資料分析改良

Abstract

錯動變形監測可提供與邊坡安全性有關的重要訊息，尤其對於山勢陡峭，降雨湍急的台灣，更需發展有效的邊坡監測系統。時域反射法(time domain reflectometry, TDR)技術應用於大地工程監測已行之有年，其透過電磁波傳送於預埋地層內之同軸纜線，當地層滑動造成纜線之幾何形狀改變所傳回之反射訊號以分析滑動位置，且可透過網路進行遠端自動化、一機多功與低成本等優勢更符合工程上實際需求。然而，現地存在纜線電阻、纜線-灌漿材料-外填土壤三者材料互制與剪力弱帶寬度等影響因子而使得TDR錯動變形量化分析仍具有難度，其中纜線電阻之影響已能利用傳輸線理論合理考慮，但材料互制之影響尚需進一步探討。 本研究將著重於材料互制之探討，於實驗室建立一套與現地邊界條件極相近之物理模型：大型直剪儀，據以了解纜線-灌漿材料-外填土壤複合材料受剪後TDR反射訊號與剪力盒位移量之關係，據以了解如何提昇感測靈敏度及量化分析之可行性。利用大型直剪儀研究結果提出TDR錯動變形監測安裝標準程序及量化分析，提供實務應用的參考；最後，透過訊號處理方法提昇TDR波形辨識度，提早偵測滑動，並以監測實務常用警戒值概念建立一門檻值，當TDR波形超過警戒門檻值，可據以調整監測頻率，進行後續較密集的觀測。 本研究充分探討不同纜線、灌漿配比與不同勁度外填材料組合下TDR纜線受剪反應，結果顯示較適用於軟弱土壤監測之纜線種類為外導體為網狀(braided cable)的RG-8纜線型，其外層橡膠包覆能防止纜線受蝕破壞，相對較軟的外導體可較早察得滑動，重要的是在不同勁度外填材料下仍有極相似的靈敏度，即纜線初始剪動後相對變形可被量化，將有助於簡化現地應用難度。另外，TDR反射訊號經訊號處理去雜訊後，可至少提早約2mm察得滑動，利用不同纜線試驗結果，初步訂出可供實務應用參考的警戒門檻值。

Time domain reflectometry (TDR) technology has become a valuable tool for detecting displacements and locating shear planes in rock or soil slopes. It is based on transmitting an electromagnetic pulse into a coaxial cable grouted in rock or soil mass and watching for reflections of this transmission due to cable deformity induced by the ground deformation. Early detection of localized shear deformation in soft soils and quantifying the shear displacement using TDR remains a challenging work. The TDR response due to localized shear deformation is affected by cable resistance, soil-grout-cable interaction, and shear bandwidth. While the effect of cable resistance can be taken into account by electromagnetic theory, the effects of soil-grout-cable interaction and shear bandwith needs further studies. The objective of this study is to develop a physical model that mimics the field condition such that the effect of soil-grout-cable interaction can be experimentally investigated. The physical model mainly consists of a large direct shear device to avoid boundary effects in radial and axial directions and a loading system that applies overburden and shear force. Different combinations of soil, grout, and cable form soil-grout-cable composites inside the shear box. The effect of soil-grout-interaction on the TDR response is experimentally investigated with emphasis on how to enhance the sensitivity to shear displacement and quantify the shear displacement from TDR response. In addition, a post singal processing technique is proposed to expedite and automate detection of shear displacement. The results show that, while the solid-outer-conductor cable (e.g. P3-500) is suitable for monitoring in rocks, a braided cable (e.g. RG-8) is desired in soils for early dectection of localized shear displacement. Furthermore, the relationship between the reflection magnitude and shear displacemtn in a RG-8 cable is relatively independent of the soil stiffness, sugguesting the possibility of quantifying shear displacement after it is detected. The experimental results show that the grout, although stiffer than the surrounding soil, does not “protect” the cable from deformity. A stiff brittle grout can be used for all soil conditions. The proposed signal processing technique allows detection of shear displacement at least 2 mm earlier than visual inspection. A shearband device is suggested to be added to the shear box in future study.