ABSTRACT
This paper compares excavation-induced and construction-related ground settlements measured during the construction of Aksel Møllers Have station, part of the Cityringen (M3) metro line in Copenhagen, Denmark, with prediction methods. In particular, the analysis focuses on empirical prediction methods currently employed in design and a selected analytical solution. These methods can relate movements to the excavation/rock depth or the wall deflection level (referred to as the ground loss approach). In this case history: the station’s box excavation was supported by secant pile walls as the temporary retaining system; the ground conditions comprise stiff glacial deposits and Copenhagen Limestone; the monitoring system at this station included levelling points at the ground surface and inclinometers at the wall. First, settlement troughs from analytical and empirical methods are analysed in terms of shape and magnitude associated with typical design parameters, with a discussion of the relevance of the ground loss approach. Next, predictions are compared with measured settlements at the ground surface. This quantifies the level of overestimation of predicted settlements associated with the current design practice or the use of the measured wall deflection for settlement prediction.
