08-09 Exam Paper

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

NATIONAL UNIVERSITY OF SINGAPORE

FACULTY OF ENGINEERING

EXAMINATION FOR
(Semester II: 2008-2009)

CE6102 - GEOTECHNICAL ANALYSIS

April/ May 2009 - Time allowed: 2 hours

INSTRUCTIONS TO CANDIDATES

1. This examination paper contains TWO(2) questions and comprises FIVE(5) printed
pages.

2. Answer ALL questions.

3. All questions carry equal marks.

4. This is an “OPEN BOOK” examination.

.../2
-2- CE6102

Question 1

(a) An axisymmetric finite element model, comprising two constant strain triangles, is used
to simulate a quarter section of a triaxial soil specimen, as shown on Figure 1-1. The
nodes are labeled N1, N2, N3 and N4, and the elements E1 and E2.

y
z x

N4 N3 x
E2
76 mm 38 mm
E1
N1 N2

19 mm

38 mm

(a) soil specimen (b) axisymmetric FE model


Figure 1-1

The element stiffness matrices of E1 and E2 may be simply denoted as

x x x x x x y y y y y y
x x x x x x  y y y y y y 
 
x x x x x x y y y y y y
[k E1 ] =   and [k E2 ] =  
x x x x x x y y y y y y
x x x x x x y y y y y y
   
 x x x x x x   y y y y y y 

in which the individual x and y terms do not necessarily represent the same values.

(i) Write down the global stiffness matrix [K G ] of the FE model after assembly of the
element stiffness matrices. Your matrix should be expressed in terms of x and/or y
(for nonzero values) and 0 (for the zero terms). Assume that the global degrees of
freedom are numbered according to the nodal numbering, that is, in the order 1X,
1Y, 2X, 2Y, …., etc.
[8 marks]

.../3
-3- CE6102

(ii) Indicate and explain clearly how you would modify the global stiffness matrix [K G ]
to incorporate the boundary fixities associated with modeling the load-controlled
consolidated undrained (CU) triaxial test, without changing the size of the matrix.
You should also indicate any corresponding modification, if required, to the applied
load vector on the RHS.
[6 marks]

(iii) Indicate and explain clearly how you would modify the global stiffness matrix [K]
to incorporate the boundary fixities associated with modeling the strain-controlled
consolidated undrained (CU) triaxial test, without changing the size of the matrix.
You should also indicate any corresponding modification, if required, to the applied
load vector on the RHS.
[6 marks]

(b) The equation for the elasto-plastic (tangent) stress-strain matrix for a work-hardening
plastic material is given by the relation

 T
a b De 
D ep  D e    T 


T
 b D e a c H a 

The yield surface for modified Cam-clay is given by

F = q2 – M2 [p’ (p 0 ’-p’)] = 0

where M is the frictional constant, and p 0 ’ is the current consolidation pressure.

Assume the following modified Cam-clay parameters:


 = 0.1,  = 0.02, M = 1.0,  = 0.25
and the initial isotropically consolidated conditions:
e o = 1.0, p’ = p 0 ’ = 100 kPa, q = 0

(i) Using the definitions given in the lecture notes, and working in terms of the
volumetric and shear strains  v and  s , evaluate the elastic stress-strain matrix D e .
[6 marks]
3(1  2)
( Note that G = K.)
2(1  )
(ii) Using the definitions given in the lecture notes, and working in terms of stress path
parameters p’ and q, and volumetric and shear strains  v and  s , evaluate the
T
vectors a , b and the matrix c H . You should first express your answers in terms
of the various quantities before calculating them numerically.
[8 marks]

.../4
-4- CE6102

(iii) Evaluate numerically the elasto-plastic stress strain matrix D ep .


[6 marks]
(iv) Using the D ep matrix obtained in (iii), calculate the axial strain d y experienced by a
normally consolidated soil specimen when it is subjected to a deviator stress dq of
50 kPa under undrained triaxial compression (CU test).
[4 marks]
(v) Using the D ep matrix obtained in (iii), calculate the axial strain d y experienced by a
normally consolidated soil specimen when it is subjected to a deviator stress dq of
50 kPa under drained, 1-D compression (that is, the specimen is constrained
laterally).
[6 marks]

Question 2

(a) For an embankment of height H and width B, what would be the appropriate extent of
the lateral boundary for a drained and undrained analysis to be valid? Give brief reasons
for your answers. How would you check that your mesh is large enough for the problem?
[7 marks]

(b) For a strip footing of width B, how far below the footing should your FEM model be
extended for an accurate analysis with a Mohr-Coulomb model? Why is it not an
advantage to make the model much deeper than required to get good predictions? How
would you use a Mohr-Coulomb model to get good predictions of long-term footing
settlements in stiffer soils?
[7 marks]

(c) There are several methods to model undrained response in FEM analysis. These are
known as Method A, B and C. Describe briefly what are the inputs of each of these
method for a Mohr-Coulomb model.
[6 marks]

(d) What are the settings of soil parameters and pore water pressure conditions you would
apply to a FEM model for an undrained excavation of a single propped wall for a Mohr-
Coulomb soil using Method A, B and C in the Plaxis program. The soil parameter
is: c'  5 kPa,  '  30o ,   20 kN / m3 , E '  20,000 kPa, '  0.3
Show clearly the values of your input for the soil parameters and initial stresses (Ko) for
each of these methods. Assume the soil is over-consolidated with OCR=2, and GWT is at
ground surface.
[12 marks]

.../5
-5- CE6102

(e) A site consists of a layer of loose sand above a Normally Consolidated (NC) soft marine
clay with the following properties:
Loose sand:   20 kN/m3, thickness=10m
NC soft marine Clay: c'  0 kPa,  '  20o ,   15 kN/m3, thickness=15m
Assuming GWT is at the ground surface, what is the undrained strength at the top and
bottom of the soft clay in compression and extension failure using the Mohr-Coulomb
model? How do these values compare with measured data correlations of cu=0.22  'v ?
[12 marks]

(f) Explain why there is a critical time step for accurate consolidation analysis? If the
average element size in a mesh is given by:
( xmax  xmin )( ymax  ymin )
le  , and for fine mesh setting n c =400
nc
What is the critical time step for a 15-noded triangle mesh of 100m x 50m, if the soil has
property is: E '  20,000 kPa,  '  0.3, and k y  1x10 7 m/s?
What would be the crtical time step for a 6-noded triangle mesh of same size. What is the
significance of a critical time step?
[6 marks]

- END OF PAPER -

You might also like