Analysis and Optimization of Transverse Post-Tensioning in Immersed Tunnels

Abstract

Due to the recent increase in traffic capacity requirements, the need for larger spans in immersed concrete tunnels has become a pressing matter. Although commonly used in Europe, reinforced concrete has a structural capacity limit when it comes to the transverse span length. The implementation of post-tensioning could allow for longer spans and a reduction in the overall concrete used. This technique is however rarely used in industry due to the complexity of applying it to underwater environments and the varied loading conditions an immersed tunnel element is subjected to. When implementing post-tensioned tendons, the curvature creates additional distributed loads which compensate for the high hydrostatic pressures and backfilling weight present at the final immersed stage of the tunnel. Due to the absence of these loads during the initial stages, prior to transportation, high tensile stresses are exhibited which can lead to severe cracking. Crack mitigation can be achieved by implementing permanent additional reinforcements in the opposite face of the post tensioning tendon. Additionally, another method is installing a temporary system connecting top and bottom slabs to replicate the final loading conditions. The objective of this research is to investigate the governing limitations of implementing transverse post-tensioning and evaluate when it is a structurally viable option. Using finite element modelling, a linear analysis was carried out to evaluate the behavior of the structure after implementing post-tensioning loads in the cross section and to identify the critical areas. The assumptions for the analytical moment distribution were found to overestimate the rigidity of the structure and made for larger moments at the wall-top slab connections, which were adjusted for the remaining parts of the study. An analysis into the effect of adding post-tensioning in the lower bottom slab revealed a substantial improvement in the final stage stress distribution, and was observed to be in full compression. A nonlinear analysis was used to provide insight into the global structural behaviour for both final immersed and dry dock stages. The final immersed stage exhibited linear elastic behaviour, whereas, the onset of cracking was observed at the top of midspan at the dry dock stage. In a further analysis of the critical dry dock stage, the relation between partial prestressing, curvature of post-tensioning tendons and the effects on the cracking behaviour at midspan was explored. The results showed that when maximum curvature and percentage of prestressing are simultaneously present, the crack width limit is reached. When slightly lowering either of these parameters, a substantial decrease in the amount of reinforcement is needed to mitigate these cracks. Lastly, a case study was carried out on the Fehrmanbelt Fixed Link to compare two methods to mitigate cracks at midspan in the dry dock: additional reinforcement and temporary tendons. This study found that implementing transverse post-tensioning was feasible when reducing the curvature of the tendons, reducing amount of prestressing and implementing additional reinforcement, which helped increase structural capacity of the critical areas in the dry dock stage.