Reliability based optimization of steel-fibre segmental tunnel linings subjected to thrust jack loadings

Neu, G.E. and Edler, P. and Freitag, S. and Gudžulić, V. and Meschke, G.

Volume: 254 Pages:
DOI: 10.1016/j.engstruct.2021.113752
Published: 2022

The circular lining in mechanized tunnelling consists of concrete segments, which are exposed to different loading cases during tunnel construction. One of the most critical loading condition is the thrust jack force, which is induced to the lining segments by the Tunnel Boring Machine (TBM) during construction. Experimental campaigns showed that steel fibre reinforced concrete is suitable for bearing such loads and could replace conventional reinforcement schemes. In this contribution, a numerical model is presented, which allows to directly track the influence of important design parameters such as fibre type, fibre orientation, fibre content and concrete strength on the structural response of steel-fibre reinforced segments. For this purpose, submodels on the single fibre level are combined into a crack bridging model considering the fibre orientation and the fibre content. The submodels are integrated into a finite element model to perform numerical structural analyses. Two validation examples demonstrate that the modelling approach is capable to predict the failure loads as well as the crack development of fibre reinforced specimens subjected to localized loads. Finally, an optimization procedure is carried out to determine a robust and cost-effective design of a fibre reinforced segmental linings. A hybrid reinforcement scheme consisting of two layers of fibre reinforced concrete is employed in order to provide an improved material utilization. The thickness of the segment and the fibre content are optimized taking uncertainties of the material parameters and construction tolerances as uncertain a priori parameters into account. A sufficient load bearing capacity and serviceability performance under all possible conditions is ensured by the consideration of accepted failure probabilities as constraints in the optimization task. © 2022 Elsevier Ltd

« back