Beruflich Dokumente
Kultur Dokumente
139]
TECHNICAL NOTE
In 2014, the authors of this paper considered predictions of reinforcement tension in a piled
embankment from British standard BS 8006 published in 2010 and the 2012 amended version, and
compared the results with finite-element model predictions. In keeping with BS 8006, any contribution
from the subsoil beneath the embankment was ignored. The present paper extends that earlier work by
also considering the potentially beneficial contribution of a lightly overconsolidated clay subsoil layer,
both in the finite-element predictions and as a simple modification to the BS 8006 predictive method.
It is assumed that there is no working platform (granular) material below the pile cap level, and that
the water table in the subsoil does not drop, since either of these factors would be likely to significantly
reduce the ability of the subsoil to carry load from the embankment. As anticipated, the subsoil support
reduces reinforcement tension. When compared to the finite-element results the proposed modified BS
8006 prediction is quite accurate. The BSI 2012 modified prediction is best (but sometimes slightly
unconservative), whereas the BSI 2010 modified prediction is conservative in all cases considered.
KEYWORDS: embankments; finite-element modelling; geosynthetics
INTRODUCTION
Zhuang et al. (2012) reported three-dimensional finiteelement (3D FE) modelling of arching in an unreinforced
piled embankment, dependent on key geometrical variables
(Fig. 1)
a = the square pile cap dimension
s = the centre-to-centre pile spacing for a square grid
He = the embankment height
Zhuang & Ellis (2014) considered predictions of reinforcement tension in a piled embankment from BS 8006 (BSI,
2010, 2012), comparing the results with 3D FE results. Based
on BS 8006 (BSI, 2010, 2012) any contribution from the
subsoil beneath the embankment was ignored.
Han & Gabr (2002) performed a numerical study on reinforced piled embankments, including the underlying subsoil;
however, an axisymmetric analysis was used. Stewart & Filz
(2005) also considered the effect of subsoil using numerical
analysis, concluding that this should be a factor in design,
but without considering how this might be achieved. DGGT
(2011) and Van Eekelen et al. (2012) considered the elastic
response of the subsoil in analytical models. However, the
models are complex, and the potentially very important
effect of the subsoil preconsolidation stress is not considered.
This technical note extends the 3D FE studies of
Zhuang et al. (2012) and Zhuang & Ellis (2014), also
596
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright ICE Publishing, all rights reserved.
Pile cap
Embankment stress e
supported by combined
action of the reinforcement
(r) and subsoil (s)
Potential arching
mechanism
WT s r
Subsoil
(a)
s/2
Finite-element model
vertical
boundaries
(b)
Tensile
reinforcement
Pile (continues
down)
597
FINITE-ELEMENT ANALYSES
Figure 2 shows a schematic diagram of the FE model,
based on the boundaries shown in Fig. 1(b).The approach
is the same as Zhuang et al. (2012) and Zhuang & Ellis
(2014), but explicitly modelling the subsoil, pile cap and pile.
Table 1 summarises the analyses undertaken. Meshes contained approximately 35 000 to 190 000 elements, and the
sensitivity to mesh size was checked, similarly to Zhuang &
Ellis (2014).
Constitutive modelling of the embankment as a dry
elasticperfectly plastic material was identical to Zhuang
et al. (2012) and Zhuang & Ellis (2014), Table 2.
As in Zhuang & Ellis (2014), a single layer of biaxial
tensile reinforcement was positioned 100 mm above the base
of the embankment. Compared to separate upper and
lower orthogonal layers, the tension result corresponds to
an average, which would be more than an upper layer and
less than a lower layer (Love & Milligan, 2003).
The 100 m thick soft subsoil was modelled using modified
Cam Clay (MCC; Wood, 1990), Table 3, with hydrostatic
groundwater pressure from the water table at the surface of
the subsoil. It was assumed that the clay had previously
experienced a small vertical preconsolidation stress (vp)
of 10 kN/m2 in excess of the in situ vertical effective stress.
The pile cap and pile were modelled as an elastic material
(concrete) with Youngs modulus of 30 GN/m2, and Poisson
ratio of 020. They were assumed to be either smooth or
rough in terms of the interface friction angle with the
surrounding soil ()
Smooth: = 0
Rough: = 10 for the subsoil (clay); = 15 for the
(granular) embankment material (on the top face of the cap)
At the start of the analyses only the subsoil, pile cap and pile
were present. The in situ effective stress in the subsoil was
specified based on the MCC parameters and preconsolidation stress.
The embankment was then constructed in layers, similarly
to Zhuang et al. (2012) and Zhuang & Ellis (2014). During
this process the soft clay subsoil was treated as undrained
(considered pragmatic and conservative). The clay subsoil
was then allowed to consolidate by way of drainage at the top
and bottom boundaries.
Monitoring of pore water pressure confirmed that the
anticipated excess pore pressure (He) was generated during
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright ICE Publishing, all rights reserved.
598
Embankment, He
Vertical faces
restrained against
normal movement
s/2
Embankment
Subsoil, Hs = 100 m
Subsoil
Reinforcement
(100 mm above
base of
embankment)
05 m
a/2
1/4
pile
cap
1/4
pile
030/2 = 015 m
Restraint in all
directions at base
of subsoil
(a)
(b)
Fig. 2. Finite-element mesh geometry: (a) overview; (b) enlarged detail at embankmentsubsoil interface
Table 1. Geometry and reinforcement stiffness
Pile cap,
a: m
10
05
10
10
CC cap
spacing, s: m
20
125
25
35
Embankment
height, He: m
20,
15,
30,
50,
65,
35,
65,
65,
100
65
100
100
Reinforcement
stiffness, J: MN/m
10, 30
10, 30
30, 100
100
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright ICE Publishing, all rights reserved.
599
Youngs
modulus: MN/m2
Poisson
ratio,
Cohesion
intercept, c: kN/m2
Friction angle,
: deg
Kinematic dilation
angle at yield: deg
050
25
020
30
170
030
010
330
0772
020
15
vp
10
Low
He
and s
One-dimensional prediction
20
40
60
y: mm
(a)
s = 35 m:
He = 50 m He = 65 m He = 100 m
30
20
vp
10
0
0
100
200
y: mm
One-dimensional prediction
300
400
(b)
ACKNOWLEDGEMENTS
The financial support of the National Natural Science
Foundation of China (grant no. 51478166), the Fundamental
Research Funds for the Central Universities (grant nos.
2015B17814, 2015B06014 and 2015B25914) and the
Scientific Research Foundation for the Returned Overseas
Chinese Scholars, State Education Ministry is acknowledged.
sa
4J
Eliminating Trp between the two equations and re-arranging
y 3
s a
WT
74 J
sa
a
Substituting WT = sr (equation (5) in main text) and
re-arranging
y 3
74 a J y 3
r
A
s s a s a
sa
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright ICE Publishing, all rights reserved.
600
H&R (2010) NSS
H&R (2012) NSS
FE NSS
100
100
80
Trp: kN/m
Trp: kN/m
80
H&R (2010) SS
H&R (2012) SS
FE SS sm
FE SS ro
60
40
60
40
20
20
0
0
6
H: m
10
100
100
80
80
60
10
40
60
40
20
20
0
0
6
He: m
10
(c)
6
He: m
10
10
(d)
300
300
200
200
Trp: kN/m
Trp: kN/m
(b)
Trp: kN/m
Trp: kN/m
(a)
6
H: m
100
100
0
0
6
He: m
(e)
6
He: m
10
6
He: m
(f)
800
Trp: kN/m
600
400
200
0
0
10
(g)
Fig. 4. Variation of reinforcement tension (Trp) with embankment height (He): BS 8006 (BSI, 2010, 2012) predictions (including proposed
modification for subsoil) and FE results: (a) a = 10 m, s = 20 m, J = 10 MN/m; (b) a = 10 m, s = 20 m, J = 30 MN/m; (c) a = 05 m,
s = 125 m, J = 10 MN/m; (d) a = 05 m, s = 125 m, J = 30 MN/m; (e) a = 10 m, s = 25 m, J = 30 MN/m; (f) a = 10 m, s = 25 m,
J = 10 MN/m; (g) a = 10 m, s = 35 m, J = 10 MN/m
74 a J
s s a
A has units kN/m2. r will vary with plan location, but is here
nominally considered to be an average value satisfying vertical
equilibrium in equation (1).
a
b
c
He
Hs
J
ks
NOTATION
A
s
Trp
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright ICE Publishing, all rights reserved.
v
vp
e
r
s
REFERENCES
Abusharar, S. W., Zheng, J. J., Chen, B. G. & Yin, J. H. (2009). A
simplified method for analysis of a piled embankment reinforced
with geosynthetics. Geotextiles and Geomembranes 27, No. 1,
3952.
BSI (2010). BS 8006-1: Code of practice for strengthened/reinforced
soils and other fills. London, UK: BSI.
BSI (2012). BS 8006-1: Code of practice for strengthened/reinforced
soils and other fills, incorporating corrigendum 1. London, UK:
BSI.
DGGT (Deutsche Gesellschaft fr Geotechnik e.V.) (2011).
Recommendations for design and analysis of earth structures
using geosynthetic reinforcements EBGEO (transl. A.
Johnson). Berlin, Germany: Wilhelm Ernst & Sohn.
Ellis, E. A. & Aslam, R. (2009). Arching in piled embankments:
comparison of centrifuge tests and predictive methods part 2
of 2. Ground Engng 42, No. 6, 2831.
601
Downloaded by [ Purdue Univ Lib TSS] on [13/09/16]. Copyright ICE Publishing, all rights reserved.