Deformation Monitoring of Retaining Walls using Terrestrial Laser Scanning

Abstract

Although the replacement cost of retaining walls is relatively low compared to other infrastructure, leading to minor research interest in their inspection, the probability of failure and collapse in retaining walls is significantly higher than in other infrastructures. Before any failure occurs, certain defects tend to be observed in retaining walls. The common theme of the observed defects in many historical cases is a change in the geometry of the retaining walls; either on a global scale (e.g., tilt and lateral displacement) or locally (e.g., bulge and cracks).

Terrestrial Laser Scanner (TLS) is one of the main contactless techniques that have been researched in monitoring the geometric deformations in retaining walls. Although the TLS often had a small error in monitoring geometric deformations (i.e., a few millimeters), some studies showed larger error (i.e., a few centimeters), suggesting that the tolerance of the TLS has not yet been fully investigated for detecting small geometric deformations, such as those for the serviceability limits of retaining structures. Factors such as the scanning range, angle of incidence, as well as the material and color of the scanned surface have already been studied regarding their influence on the accuracy of TLS measurements. However, their impact on the accuracy of deformation estimation between point-clouds at two epochs is yet to be investigated. The aim of this research is to propose a list of strategies to achieve an accuracy of 1 − 2mm in monitoring the global and local geometric deformations in retaining walls.

Two experiments are conducted to simulate global and local geometric deformations using two prototypes of retaining walls. The experimental assessment focuses mainly on investigating the effect of (i) the type and amplitude of the deformation and (ii) scanning parameters (i.e., scanning distance and angle of incidence), and (iii) data analysis method on the deformation estimation using the TLS. This experimental work is done for two cases: (i) when the reference point-cloud was at the same location as the deformed cloud, and (ii) when the reference and deformed clouds are taken from different scanning positions. Furthermore, a case-study of monitoring the seasonal movement of retaining walls is conducted to examine the performance of TLS in monitoring small geometric deformations of retaining walls.

The main contribution of this research lays on listing strategies and recommendations to achieve accuracy of within 1−2mmwhile using the TLS for monitoring small geometric deformations in retaining walls. The thesis contributes also to knowledge of the behaviour of retaining structures through the analysis for the case-study. Although it was shown in previous research that many cases had collapsed because of the additional hydro-static pressure in the retained soil, the results in this research suggests that the solar radiation had larger impact on the seasonal deformation than the change in the hydro-static pressure caused by the fluctuation in water level of the canal. Additionally, it was observed that the parts of the sheet piles that were exposed to the sunlight for longer period and/or at the peak of the solar radiation had larger seasonal deformation.