Actes du colloque - Volume 4 - page 820

3484
Proceedings of the 18
th
International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013
to calculate the settlement of foundation. Assuming that the
basis load - settlement curves can be fitted by using the
hyperbolic curve, we create the hyperbolic curve method, as
formula (1):
As a and b are the parameters of test, the nonlinear tangent
modulus of soil can be calculated as hyperbolic curve formula
(1), as formula (2):
bs a
s
p
(1)
0
2
)
1(
t
u
f
t
E
p
p R E

(2)
0
t
is the initial tangent modulus, p is the additional
stress of soil,
u
is the ultimate capacity of foundation soil,
f
is the damage ratio coefficient which is similar to that of
the Duancan-Chang model.
E
p
R
While the tangent modulus
t
E
along the depth is solved
by formula(2), we could use layer-wise summation method to
calculate the settlement. It is a good method to determine the
parameters of soil because the soil is undisturbed and the
nonlinear properties of soil are well considered, this method can
calculated the nonlinear settlement of soil accurately.
Assuming soil layer is
j
h
load increment
i
p
, the
amount of compression of this soil layer is as formula (3):
ij
j
i
ij
E
h
p
s
  

(3)
is the additional stress of distribution coefficient,
and
ij
is the tangent modulus on the soil layer corresponding
to the load p. After the amount of compression of each layer is
calculated, according to layer-wise summation method, the total
settlement under added load is
as formula (4):
E
(4)
 
n
j
ij
i
s
s
1
When using this method to calculate the nonlinear
settlement of foundation, the key is to determine three
parameters, including the tangent modulus value
t
of the each
foundation layer of soil, and the strength parameters - c and φ
required in Formula (2) calculation as well as the initial tangent
modulus
, which are simple.
E
0
t
Please put one open line before a Figure (centered, see Figure 1)
or a Table (use the Figure tag or use paintbrush).
E
1.2
Foundation settlement calculation parameters determined
by plate loading test
As to the p-s curve measured by the plate loading test, it
can inversely calculate the internal friction angle φ of
foundation settlement parameters, cohesion c and the initial
tangent modulus
0
t
, resulting in a more reasonable
calculation parameters.
E
Loading test site is located in the Riverside Campus,
Texas A & M University, with 0 to 10.5m depth of sand and
black stiff clay as the sublayer. The main physical and
mechanical parameters of the indoor tested silty sand are shown
in Table 1.
Tab.1 physical mechanic index of ground soil
(kN/m
3
) w (%)
Gs
e
c
(kPa)
(°) depth(m)
0
34.2
0.6
15.6
5.0
2.66 0.75
0
36.4
3.0
A total of five plate loading tests are conducted in the site,
and the plates are square with width of 1~3m. The test is carried
out at depth of 0.76 m.
We use two methods to forecast the p-s curve, e.g. the
first method uses the aforementioned fitted hyperbolic
parameters a and b to directly draw p-s curve based on the
assumption
[2]
that p-s curve complies with the hyperbolic
model; the second method is the aforementioned tangent
modulus method: first calculate the tangent modulus
t
under
different loads as formula(2), then calculate the amount of
compression of each layer as(3), and lastly make summary of
the foundation settlement under different load by layer-wise
summation as (4), and draw p-s curve
[4 ]
E
When calculating the amount of compression of each
layer by tangent modulus method, the ultimate bearing capacity
of each layer is determined by the Terzaghi formula (5). The
cohesion
0
c
, and the internal friction angle φ is inversely
calculated by loading test data in Formula (5).
c
q
u
cN qN BN P
 
2
1
(5)
By using the above two methods, the p-s curves of the
five plates are drawn respectively, and compared with the
actually measured data shown in Figures 2 to 6.
Fig.2 The P-S curve of 5#footing(1.0m×1.0m)
Fig.3 The P-S curve of 2# footing(1.5m×1.5m)
Fig.4 The P-S curve of 4# footing(2.5m×2.5m)
1...,810,811,812,813,814,815,816,817,818,819 821,822