See discussions, stats, and author profiles for this publication at: [Link]
net/publication/334576003
Reliability-based analysis of braced excavation in unsaturated soils considering
conditional spatial variability
Article in Computers and Geotechnics · July 2019
DOI: 10.1016/[Link].2019.103163
CITATIONS READS
4 225
2 authors:
Amir Gholampour Ali Johari
8 PUBLICATIONS 20 CITATIONS
Shiraz University of Technology
108 PUBLICATIONS 613 CITATIONS
SEE PROFILE
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Finite element programming View project
Geotechnical Earthquake Engineering View project
All content following this page was uploaded by Ali Johari on 19 July 2019.
The user has requested enhancement of the downloaded file.
&RPSXWHUVDQG*HRWHFKQLFV
Contents lists available at ScienceDirect
Computers and Geotechnics
journal homepage: [Link]/locate/compgeo
Research Paper
Reliability-based analysis of braced excavation in unsaturated soils
considering conditional spatial variability
⁎
A. Gholampour, A. Johari
Dept. of Civil and Environmental Engineering, Shiraz Univ. of Technology, Shiraz, Iran
A RTI C L E I NFO A B S T R A C T
Keywords: This study presents a practical approach for reliability analysis of braced excavation in spatially varied un-
Braced excavation saturated soils. To take full advantage of the measured data, random fields are conditionally simulated and
Reliability analysis implemented in a finite element analysis. In this approach, the uncertainties involved in the soil-structure in-
Unsaturated soils teraction are considered by performing the analysis for perfectly rough and perfectly smooth interfaces leading
Conditional simulation
to upper and lower bounds on the full range of solutions. To assess the influence of the unsaturated condition on
the braced retaining system (i.e., wall and struts), the probabilistic analysis of excavation is carried out with and
without considering suction. The results demonstrate that applying the unsaturated state not only causes sig-
nificant changes in the statistical parameters of wall responses; it also moves the location where the maximum
values occurred.
1. Introduction slip lines method or scaled boundary method [14,15]. These studies
illustrate that the spatial variability is an essential factor in the relia-
Braced excavations are frequently utilized for construction of un- bility-based design of geotechnical problems and ignoring this factor
derground facilities in urban environments. This type of earth sup- can result in either overestimation or underestimation of the failure
porting system usually consists of two main components (i.e. wall and probability.
struts) for ensuring soil stability. In braced excavation analysis, the wall Although the unconditional random field provides an applicable
deflection is becoming a crucial design issue, and failure to control this approach to estimate soil spatial variability, it discards the values and
deformation can cause considerable damage to adjacent buildings and the locations of measured data and uses only the general statistics (e.g.,
utilities [1–4]. mean, standard deviation and covariance). Neglecting the measured
Excavations are commonly performed above the groundwater data may increase the simulation variance of random fields and leads to
where the soil state is unsaturated [5–7]. Matric suction, which is de- a conservative design which shows a waste of site investigation effort.
fined as the difference between pore air pressure and pore water On the other hand, the conditional random field takes full advantage of
pressure, plays a major role in shear strength behaviour of unsaturated values and locations of the measured data and reduces the degree of
soils. This state variable may be estimated by textural properties which uncertainty. Few researchers have addressed the application of the
can affect the soil behaviour almost as much as mechanical properties. conditional random field in soil problems and they have all focused on
Although textural properties are normally available in soil reports, slope stability analysis [16–18].
these parameters carry many uncertainties. In this paper, the application of conditional random fields is ex-
The level of uncertainty is mostly influenced by the inherent spatial tended into the reliability analysis of braced excavation in unsaturated
variability of the subsoil. One attempt to assess the soil spatial varia- soils which is known as a complicated soil-structure interaction pro-
bility is developing a technique called the Random Finite Element blem. The excavation analysis is carried out with a finite element-based
Method (RFEM), which uses the unconditional random field theory in program provided in MATLAB. In order to consider the uncertainties
conjugation with the Finite element method (FEM) [8]. There is ex- involved in interface modelling, the solutions are obtained for the
tensive literature in the application of RFEM in probabilistic geo- perfectly rough and perfectly smooth interfaces leading to upper and
technical engineering [9–13]. Application of unconditional random lower bounds on the wall responses. The Sequential Gaussian
fields has also been studied with other numerical techniques such as the Simulation (SGS) method is used to generate conditional random fields
⁎
Corresponding author.
E-mail address: Johari@[Link] (A. Johari).
[Link]
Received 20 April 2019; Received in revised form 1 June 2019; Accepted 5 July 2019
;(OVHYLHU/WG$OOULJKWVUHVHUYHG
View publication stats