時域反射方法於地層下陷監測技術研發及可行性評估

Abstract

台灣的地層下陷問題已存在50多年，地層下陷監測方法大致分為兩大類，包含地表量測及地下監測，目前台灣主要採用方法分別對應為衛星定位固定站與磁環式監測井量測。本研究的主要目的為改善現有的地層下陷監測方法的限制，如單井僅能單一深度量測、無法自動化、量測準確度因人員不同有所誤差等，以時域反射方法(TDR)技術為基礎研發一可自動化、單井多點位量測、高精度、高重覆性的地層下陷監測元件。 TDR沉陷計的主要構件由金屬套管、纜線及感應環所組成，研發概念為一放大版同軸纜線，利用同軸纜線及金屬套管為媒介，當電磁波由傳輸段至量測段時，外導體內徑將內縮使其剖面改變，特徵阻抗將縮小，使用TDR方法擷錄波形時可以發現感應環內環處有明顯的反射訊號，以反應沉陷量變化，若地層發現下陷，可配合初始波形差異分析，定量提供感應環特徵波形偏移量。 TDR沉陷計內部元件的選擇，如內、外導體的選擇，外導體為金屬套管(內徑78mm)。以金屬棒為內導體在空氣中有明顯的感應環訊號，但水中衰減很快，且現地施作上需對金屬棒車公、母螺紋才能解決搭接問題。若以纜線內金屬材質為內導體，纜線無搭接問題，亦有包覆層可防鏽、增加訊號傳遞長度，本研究選用纜線為同軸纜線(P3-500)鋁殼為內導體，感應環在水中及空氣中皆有顯著的反射訊號提供分析。感應環材質及尺寸的選擇，在金屬環內徑(25mm、23mm、20mm及16mm)及塑膠環內徑(21mm、18mm)選擇一個良好反射訊號且可以垂直移動的感應環，經過比較以金屬環內徑23mm為最佳尺寸，另外我們發現若增加感應環，總走時將會增加，意指當電磁波的速度通過感應環後將有些微的改變，可配合初始波形差異分析，提供感應環特徵波形偏移量(走時差)，當感應環位移3cm(真實量)時，量測平均值2.92cm及標準偏差0.04223。

Monitoring methods for land subsidence are typically classified into underground monitoring and surface monitoring. Current monitoring practice in Taiwan is mainly via magnetic ring monitoring well (underground) and continuous Global Positioning System (GPS) measurement on fixed station (surface). The main purpose of this study is to improve the shortcomings of existing land subsidence monitoring techniques, such as single depth measurement limitation for a monitoring well, lack of automated monitoring potential, measurement accuracy inconsistencies due to operator errors. Therefore, this study developed a pilot land subsidence monitoring approach based on Time Domain Reflectometry (TDR), which aimed at automated subsidence monitoring, multiple depth measurement for a monitoring well, monitoring components with high accuracy and reproducibility. TDR subsidence meter (TDRSM) are comprised of metal casing, waveguide cable, and sensing ring. Initial prototype of TDRSM is to adopt the concept of an enlarged coaxial cable. By using coaxial and metal casing as transmission medium, when electromagnetic (EM) waves travel from transmission segment to sensing segment, reflection is induced at the sensing ring due to characteristic impedance drop caused by the cross-sectional area reduction of the outer conductor’s inner diameter (ID). Observing captured waveform from TDR instrument, strong reflection signals are found at the sensing segment and settlement amount can be deduced from the waveform variation. In the event of land subsidence, subsidence amount can be deduced quantitatively from the waveform offset differential analysis between the initial and current characteristic sensing waveforms. Selection of TDRSM component would influence the reflection signal intensity of the sensing ring and its corresponding analysis approach. This study adopted metal casing (fixed ID of 78 mm) as the outer conductor, whereas the optimal configuration of TDRSM is explored from the combination of inner conductor and sensing ring. Configuration of metal rod as inner conductor generated clear reflection signal in air but attenuated rapidly in water. Extra metal threads are required on the rods to facilitate extended connection on site. However, no connection issue would be encountered by using cable as inner conductor instead, with added benefits of rusting protection and signal transmission length increment due to the cable coating layer. This study hence selected aluminum casing in P3-500 coaxial cable as the inner conductor, reflection signals obtained in both air and water are sufficiently strong for waveform analysis. Sensing ring material and dimension combination are determined from the best configuration which produced good reflection signal and could move vertically with the subsided strata. This study examined the configuration using metal ring (ID of 25mm, 23mm, 20mm and 16mm) and plastic ring (ID of 21mm and 18mm). After series of experiment, best sensing ring configuration is found as metal ring with ID of 23mm. With additional sensing rings, total travel time would increase as a result of slight variation in EM wave velocity after passing the sensing rings. From waveform offset differential analysis between initial and current characteristic sensing waveforms, the deviated amount of the characteristic waveform are computed. In case of actual sensing ring offset by 3 cm, TDRSM measured data resulted in average of 2.92 cm with standard deviation of 0.04223 cm.