Assessing and Updating Damage Probabilities for a Deep Excavation in Mexico City Soft Soils
- PDF / 9,844,331 Bytes
- 18 Pages / 595.276 x 790.866 pts Page_size
- 52 Downloads / 167 Views
ORIGINAL PAPER
Assessing and Updating Damage Probabilities for a Deep Excavation in Mexico City Soft Soils C. J. Sainea-Vargas1
•
M. C. Torres-Sua´rez2
Received: 1 August 2019 / Accepted: 12 December 2019 Indian Geotechnical Society 2019
Abstract This paper presents the results of numerical and probabilistic modeling of a deep excavation in Mexico City soft soils, concerning the estimation of potential damages in neighboring buildings. Constitutive model hardening soil small strain was used to model soft soil behavior, and ref selected parameters Eref 0 and E50 were represented as random variables and random fields. A combination of response surface method and random variable and random field-based tridimensional finite element modeling was employed to assess the excavation performance in terms of damage potential indexes. Results for the two representations of parameter spatial variability were obtained and compared for the different stages in the construction sequence. In the considered case, when uncertain parameters are modeled using highly anisotropic random fields, the assessed damage probability values differ considerably to the ones of random variable-based modeling. The distributions of calculated damage probabilities are also different for the buildings located around the excavation when changing the soil variability representation. On the other hand, calculated ground movements from semiempirical methods and actual measurements were integrated to obtain damage potential indexes for the excavation stages. The prior building damage probabilities and additional estimations are combined in a Bayesian framework to update the initial estimations. This approach allows also
& C. J. Sainea-Vargas [email protected] 1
Escuela de Ingenierı´a Civil, Universidad Pedago´gica y Tecnolo´gica de Colombia, Avenida Central del Norte No. 39-115, Tunja, Boyaca´, Colombia
2
Facultad de Ingenierı´a, Universidad Nacional de Colombia, Av. NQS No. 45-03, Bogota´, Colombia
updating soil constitutive parameter values, resulting in equal prior and posterior mean values of parameters Eref 0 and Eref 50 , but reducing its uncertainty. Keywords Bayesian updating Building damage Deep excavations Numerical-probabilistic modeling Random fields Soft soils
Introduction The design and construction of deep excavations in soft soils in urban areas are subjected to considerable difficulties, given its unfavorable mechanical behavior, which may generate structural damages during its execution [1]. Mexico City (CDMX) lacustrine zone imposes problematic conditions to geotechnical projects due to the presence of soft soils, materials of high compressibility and low shear strength, subjected to high-level water tables and regional subsidence related to water pumping in deep aquifers. The properties of these materials have been studied intensively since the second half of the twentieth century [2–4]. The lacustrine sediments in CDMX are complex mixtures of crystalline minerals and amorphous materials with microfossils co
Data Loading...
