Abstract

Large-diameter reinforced concrete caissons are a widely-adopted solution for underground storage and attenuation tanks, and launch and reception shafts for tunnel boring machines. The sinking process presents a number of challenges including maintaining verticality of the caisson, controlling the rate of sinking, and minimising soil-structure frictional stresses through the use of lubricating fluids. A bespoke monitoring system has been developed at University of Oxford to provide early warning of adverse responses during the sinking phase (e.g. excessive soil-structure interface friction). The monitoring system was trialled on a recent pilot project in the UK involving the construction of a 32 m internal diameter, 20 m deep reinforced concrete caisson. This seminar will describe the monitoring system that was developed and its impact on the construction process of the pilot project. The potential role of machine learning algorithms in construction monitoring will also be explored.