Finite element analysis of asymmetrical deep braced excavation
Lim, Tracy Cyne Jeat
Date of Issue2017-05-15
School of Civil and Environmental Engineering
In most deep braced excavation analyses, for convenience, half of the excavation cross section is modelled and analysed with the assumption of symmetrical geometry and horizontal soil stratigraphy across the excavation site. However, in reality, the actual ground conditions at the excavation site are not uniform, but may vary across the section. The focus of this project is to conduct back analysis using the finite element program PLAXIS 2D to study the importance of carrying out full excavation finite element analysis by incorporating the actual excavation conditions, which include the variation in soil stratigraphy across the excavation and the difference in wall depths for asymmetrical deep braced excavation. Case study on the collapse of cut and cover tunnel excavation adjacent to the Nicoll Highway was considered due to the comprehensive nature of data. To consider asymmetrical excavation geometry, full excavation analysis involving two walls was carried out. In addition, half mesh analysis of north and south wall with the assumption of symmetrical excavation geometry and horizontal soil layers were also conducted to have a comparison with the full excavation analysis. Half mesh analyses adopting method A and C were carried out and the results showed that method A is an unsafe method of computation as the design undrained shear strength of clay is being underestimated. Lateral wall displacements, wall bending moments and strut loads were obtained from each analysis to highlight the differences between method A and C, north and south wall, lastly the half and full mesh analysis. The findings show that it is important to fully incorporate the actual excavation conditions at the excavation site in a finite element analysis as the full excavation analysis considering the actual excavation conditions obtained larger lateral wall displacements, wall bending moments and strut loads. To provide a comparison, even when the most conservative soil profile was adopted, the convenient half excavation method is inadequate in obtaining actual behaviours of the wall and soil at both sides of the excavation. Hence, when the soil layers are inclined and the soil stratigraphy at both walls is distinctively different, full excavation analysis should be carried out instead.
Final Year Project (FYP)
Nanyang Technological University