ABSTRACT
Deep excavations inevitably induce stress relief and soil movement, which may Cause damage to load-bearing structures which often historical buildings. Although various researchers have investigated the effects of excavation on adjacent brick Inasonry " wall, the mechanism of crack development caused in the wall due to excavation-induced stress relief and ground deformation requires more investigation.
This study investigates the interaction between multi-propped deep excavations and an existing isolated brick masonry wall three-dimensionally and quantifies induced settlement, slope, and mechanism of development or cracks-pattern in the brick masonry wall. Three series of the numerical parametric studies were carried out to investigate the responses of the existing brick masonry wall at different depths of deep excavation in sand. In the first series of the parametric study, the effects of mufti-propped excavation brick masonry wall (constructed parallel to excavation) are investigated. In the second series, the effects of the brick masonry wall orientation (i.e. perpendicular to the excavation) were investigated by varying excavation depths, wall locations with respect to excavation, sand density, and load on the wall. The effects of twin excavations on the settlement and cracks pattern mechanism in an isolated brick masonry wall were examined in the third series of this study. Twin excavations were simulated one after the other in cases Of PerpPerp and ParPerp. The twin excavations were perpendicular to the brick masonry wall in the case of PerpPerp, whereas the first excavation was parallel and the second excavation Was perpendicular to the brick masonry wall in the case of ParPerp. In addition, another analysis was performed with the change of construction sequence of twin excavation in case PerpPar in which the first excavation and the second excavation were carried out perpendicular and parallel to the brick masonry wall, respectively. To accurately capture the excavation-induced ground deformation, path- and strain-dependent soil stiffness at small strains were taken into account in each numerical analysis by adopting an advanced hypoplastic sand model coupled with the inter-granular strain formulation. Centrifuge test results reported by [II were used to validate the soil model and model parameters. The brick masonry wall was modelled with a micro-modelling technique. The traction- separation behavior of the cohesive element was employed to model the mortar joints.
The computed results revealed that significant differential settlement (with the maximum settlement at the mid-part of the footing) was induced in the wall (parallel to the excavation) due to excavation-induced stress relief. The deformed profile of the wall behaved as a simply supported beam which results in bending of the wall at its middle portion at the end of the excavation. This induced tensile stresses at the lower portion of the wall. As a result, the crack width developed in the foundation and lower portion of the wall is larger than that in the upper layers of the wall. Negligible footing settlements are observed when the wall is located at a distance exceeding 0.48/1e (excavation depth) from the basement. By increasing the relative density from 30% to 90%, the induced tooting settlement decreases by 72%. In the perpendicular case, the settlement in the portion Of the wall closest to the diaphragm wall (i.e., deep excavation) was larger than that in farthest during the excavation. The settlement decreased significantly in the portions away from the excavation. The portion of the wall (()Sv/BS12) bends downward, behaving like a cantilever beam. As a result, cracks were developed in the upper brick courses of the wall. The different loads applied on the top of the brick masonry wall have significant effects on the responses of the wall due to excavation in each series of the parallel and perpendicular walls. During sequential twin excavations, the maximum settlement after twin excavation in the case of ParPerp is 37% larger than that in the case PerpPar. In the case of PerpPerp, the twin excavations resulted in the hogging deformation profile, causing the cracks at the top of the wall. In the case of PerpPar, the second excavation resulted in the closing of cracks developed on the top of the wall due to the first excavation.
Keywords: Brick masonry wall; multi-propped deep excavation; sequential twin
excavations twin excavation; micro-modelling; differential settlement; Cracks