ONE DAY SEMINAR ON GEOTECHNICAL ASPECTS IN INFRASTRUCTURE(Organized by IGS Chennai Chapter and Thiagarajar College of Engineering, Madurai)
Date: 25.01.2014
Use of Geotechnical Software in Infrastructure Projects
Dr. Subhadeep BanerjeeAssistant ProfessorAssistant Professor
Geotechnical Engineering DivisionDepartment of Civil EngineeringIndian Institute of Technology MadrasChennai 600 036, India
How to handle a geotechnical problem ?Acquire land.
Reconnaissance of the sitesite.
Geologic history.
D t il dDetail survey and leveling.
Courtesy: Burland (1987)
How to handle a geotechnical problem ?
Sched le detail soilSchedule detail soil testing.
Field: SPT, Borehole, , ,plate load test etc.
Laboratory: Index h
Courtesy: Burland (1987)
property, strength, compressibility etc.
How to handle a geotechnical problem ?
Modeling:g
physical: not always possible
analytical: difficult for complex problem
Courtesy: Burland (1987)
numerical: FEM, BEM, FDM etc.
User Programmer
Analyst Engineer
Some Common Geotechnical Analysis Software
Plaxis (FEM) Pl i i f th t l ft d b t h i l Plaxis is one of the most popular software used by geotechnical consultants. it is preferred because of its user-friendly nature.
Plaxis is quite handy for plane stress problems.
There are lot of geotechnical features (such as anchors, geogrids, tunnels etc.) in-built in Plaxis. So just click and play….
Diff t tit ti d l f i li ti Different constitutive models for various applications.
Limitations It is a “Blackbox” type software.
It is so easy to use that the person who has no proper training in geotechnical engineering may possibly run a Plaxis analysis by simplygeotechnical engineering, may possibly run a Plaxis analysis by simply following some tutorials.
It is difficult to model non-conventional geotechnical problems (such l t i bl i l t i t )as, large strain problems, irregular geometries etc.)
Flexibility in modelling is absent.
FLAC d ib li it fi it diff l ti
FLAC (FDM) FLAC describes an explicit finite difference solution.
It can be used for rock mechanics problem as well.
It also has specific features such as structural elements e g to It also has specific features, such as structural elements, e.g., to represent anchors, piles, rock bolts or tunnel support, capabilities for thermal and hydro-mechanical analysis.
It i ti l l f l f d i bl It is particularly useful for dynamic problems.
Limitations It is NOT so simple in use.
It is difficult to model complicated geometries.
Dynamic analysis may sometimes encounter convergence problem
G St di t i i k f t h i l d
GeoStudio (Limit equilibrium) GeoStudio contains various packages for geotechnical and geoenvironmental applications:
1. Slope/W (Slope stability analysis)
2. Sigma/W (Load deformation analysis)
3. Seep/W (Seepage problems)
4. Quake/W (Dynamic analysis)
5. Temp/W (Geothermal analysis)
6. Air/W (Air flow analysis)
7. CTRAN/W (Contaminant transport)
8. Vadose/W (Vadose zone and soil cover analysis)
It is based on simple limit equilibrium method.Limitations
p q
The solution sometimes overestimates soil strength.
It is difficult to model complicated geometries.
It i l FEM
ABAQUS (FEM) It is a general purpose FEM.
It can model solid and water as two phases.
ABAQUS-explicit analysis is particularly useful for dynamic problems ABAQUS-explicit analysis is particularly useful for dynamic problems.
It offers various flexibilities in modelling.
It is NOT easy to use More suited for research
Limitations
It is NOT easy to use. More suited for research.
For large problems with many nodes ABAQUS analysis may encounter memory problem.
Convergence for highly non-linear problems is not easy to achieve.
ANSYS (FEM)
It is a general purpose FEM for mostly mechanical engineering applications.
It is quite efficient in dynamic analysis.
It is NOT suited for geotechnical problems
Limitations
It is NOT suited for geotechnical problems.
Couple flow analysis can not be done.
Limited number of constitutive models.
There are few moreThere are few more,
SASSI: Useful for soil-structure interaction analysisy
LS DIANA: Particularly suited for blast loadings.
SageCrisp: Couple flow analysis can be doneSageCrisp: Couple flow analysis can be done.
There are many more…..
Modern Finite Element SoftwareModern Finite Element Software
Ease of UseEase of Use≠
Ability to Use
especially true for geotechnical softwareespecially true for geotechnical software
Consequence
There may be many more if…
Problem 1: Stability Analysis of Slopes
Problem 1: Stability Analysis of Slopes
RL 107.00RL 107.00RL 104.00
RL 100.00 RL 96.70
RL 93 40
Ash Core Ash Core
4 m
7 m `Earthen Cover
RL 93.40RL 90.00
`Fly Ash Embankment ` Earthen Embankment
2.251
4m
5 m4 m
RL 80.00
The typical section of a dyke of NALCO ashpond at Angul OdishaThe typical section of a dyke of NALCO ashpond at Angul, Odisha
Plaxis model
The analysis was carried out using PLAXIS ver. 8.
Plane strain model with 15 noded triangular elements.
The base of the embankment is assumed as fixed base. The base of the embankment is assumed as fixed base.
The sides are horizontally restrained.
Material properties
Elevation Cohesion (kPa)
Angle of friction (⁰)
Unit weight (kN/m3)
B l 80 RL (F d ti ) 50 30 18Below + 80m RL (Foundation ) 50 30 18
+80 to +90 m RL (within starter dyke) 50 20 18
+80 to +90 m RL (Ash deposit) 5 35 1380 to 90 ( s depos t) 5 35 3
+90 to +100 m RL (Ash deposit) 5 30 13
+100 to +107 m RL (Ash deposit) 0 30 13
Material properties
Calculation stages
Results
FOS
Results
Total increment
Failure surface: It is not the conventional slip circles
Problem 2: Analysis of Excavation andProblem 2: Analysis of Excavation and Support Systems
Problem statement
The excavation carried up to a depth 11m from ground The excavation, carried up to a depth, 11m from groundsurface, was roughly rectangular in plan with dimensions of100m x 26m.
The excavation area was circumference by a 5-storybuilding on the north side and roads on all other sides.g
Problem statement
The sheet piles were driven to a depth of 30m below theground surface to support the excavation.
There were six levels of internal struts of three different There were six levels of internal struts of three differentsizes.
Steel H Piles were driven down to the bedrock at Steel H-Piles were driven down to the bedrock athorizontal 1.5m grid spacing within the excavation site.
26Sh il llSix no. of struts
11
26mSheet pile wall
11m
30m
H-Piles @ 1.5mspacing
Problem statement
The soil profile consisted of six different layers,
1.5m thick layer of fill
20.50 m thick marine clay layer,
9m thick silty clay layer,
7m thick medium stiff clay layer,
7m thick sandy silt layer and 7m thick sandy silt layer and
weathered rock (~5m).
Height of sheet pile
ll 30
SHEET PILE WALL
depth of excavati
0.0-1.5
wall, 30m
excavation 11mMARINE CLAY
STRUTSNo. excavation l l i
-22
SILTY CLAY
STRUTS
levels, six-31
38
SILTY CLAY
MEDIUM STIFF CLAYH-Piles-38
-45SANDY SILT
WEATHERED ROCK
H-Piles
-50WEATHERED ROCK
Excavation stages
Stages Construction sequences
1 Pile installation considering surcharge of 10 kPa for existing structure
2 Sheet pile driving up to depth of 30 m below ground
3 Excavation up to -1.4 m and installation of strut 1 at -1.0 m with a preload of 100 kN
4 Excavation up to -4.5 m and installation of strut 2 at -3.5 m with a preload of 150 kN
5 Excavation up to -6.0 m and installation of strut 3 at -5.25 m with a preload of 200 kNof 200 kN
6 Excavation up to -7.5 m and installation of strut 4 at -7.25 m
7 Excavation up to -9.25 m and installation of strut 5 at -8.75 m
8 Excavation up to -11 m and installation of strut 6 at -10.25 m
Struts
-1 H 350x350x12x19
Ground
1
-3.5 H 350x350x12x19
H 350x350x12x19
-5.25 H 400x400x13x21
-7.25-8.25 2H 400x400x13x21
2H 400x400x13x21
-10.25
2H 400x400x13x21
Excavation level
Ground water table
The initial position of the ground water table can be taken asground itselfground itself.
However, with each stage of excavation GWT will be lowered toth ti l l th i idthe excavation level on the passive side.
Initial GWT
At excavation stages
Fill Marine Clay Silty Clay Med. Stiff Clay Sandy Silt Weathered rockrock
Type ?? ?? ?? ?? ?? ??
kN/ ³ 18 00 15 00 18 00 16 00 18 00 21 00s gunsat kN/m³ 18.00 15.00 18.00 16.00 18.00 21.00
gsat kN/m³ 18.00 15.00 18.00 16.00 18.00 21.00
Eur’ kN/m² 13000 2344 13000 18900 26000 230000pe
rtie
n 0.200 0.200 0.200 0.200 0.200 0.200
c’ kN/m² 10.00 1.00 5.00 1.00 1.00 300.00
j' ° 30 00 20 00 22 00 22 00 35 00 35 00oil p
ro
j ° 30.00 20.00 22.00 22.00 35.00 35.00
y ° 5.00 0.00 5.00 0.00 5.00 0.00
Einc’ kN/m²/ 0.00 473.00 0.00 473.00 0.00 0.00
So
m
yref m 0.000 0.000 0.000 0.000 0.000 0.000
Rinter. 0.50 0.50 0.50 0.50 0.50 0.50
esNo. Identification EA
kN/mEI
kNm²/mW
kN/m/m
1 Sh il ll 3 85E6 47196 0 00 0 15pert
ie
1 Sheet pile wall 3.85E6 47196 0.00 0.15
2 H Piles (H 344x354x131 kg/m) 2.4E6 18768 0.90 0.15
prop
No Identification EA Horizontal spacingural
No. Identification EAkN
Horizontal spacing
1 H 350x350x12x19 3599730 5.00
2 H 400x400x13x21 4527090 5.00ruct
u
3 2H 400x400x13x21 9054000 5.00Str
ResultsLateral deflection of wall
00 50 100 150 200 250
Bending moment
0-400 -300 -200 -100 0 100 200 300 400
Vertical settlement
20
00 10 20 30 40
(a) (b) (c)
-10
-5
-10
-5
-60
-40
-20
-20
-15
Dep
th (m
)
-20
-15D
epth
(m)
-100
-80
Settl
emen
t (m
m)
-25
Exc. Stage 1Exc. Stage 2Exc. Stage 3Exc. Stage 4
Exc. Stage 5
-25
-160
-140
-120S
-35
-30
Deflection (mm)
Exc. Stage 6
-35
-30
Bending moment (kN-m/m)
-200
-180
Distance from wall (m)
Drain-undrain analysisDrain undrain analysis
What is drain/undrain behaviour?
• Undrain: Excess pore pressure are not allowed to dissipatep p p
• Drain: Excess pore pressure completely dissipated
How to choose drain/undrain?How to choose drain/undrain?
• Short term problem: undrainpeg. earthquake, blast etc.
• Long term problem: draineg. excavation, tunneling etc.
Analysis in PLAXIS
Undrained Behaviour
MethodPlaxis
Material Material ModelParameters Computed
Stet od ate aSetting
ate a ode StressesStrength Stiffness
A Undrained Mohr-Coulomb C', Ø' (Effective)
E', ' (Effective)
Effective stressand pore pressure
C Ø 0 E' ' Eff ti tB Undrained Mohr-Coulomb Cu, Øu = 0 (Total)
E', ' (Effective)
Effective stressand pore pressure
C Non-porous Mohr-Coulomb Cu, Øu = 0 (Total)
Eu, u = 0.495 (T t l)
Total stress(Total) (Total)
D As in Method A, for other soil models (HS, SS, SSC)
Drained BehaviourDrained Behaviour
Drained Mohr-Coulomb, other models
C', Ø' (Effective)
E', ' (Effective)
Effective stress.Pore pressure
specifiedby user
Look at a Simple Problem of Single Propped Wallp g pp
PARAMETRS Used in Method A
SWITCH SOIL TYPE: UNDRAINED (EFFECTIVE Stress Analysis)
PARAMETRS Used in Method BSWITCH SOIL TYPE: UNDRAINED (EFFECTIVE Stress Analysis)
Use Advance button, and set t thstrength
increase with depth = 3.75 kN/ 2/kN/m2/m obtained from Ko=0.5 in
)('sin2
)1( 'v
Kocu
PARAMETRS Used in Method C
SWITCH SOIL TYPE: NON-POROUS (TOTAL Stress Analysis)
Use Advance button, and set strengthstrength increase with depth = 3.75 kN/ 2/kN/m2/m obtained from Ko=0.5 in
)('sin2
)1( 'v
Kocu
Modeling of Pore Water Pressures
• Method A, Use Z-Water TableMethod A, Use Z Water Table
• Method B Use Z-Water TableMethod B, Use Z-Water Table
• Method C No Water Table place phreatic line• Method C, No Water Table, place phreatic line at the base of mesh
Modeling of Ko Condition
• Method A, For NC soils, Ko =1-sin’• For OC soils use OCRKK NC• For OC soils use • Method B, same as Method A
M th d C i T t l St
OCRKK NCoo
• Method C is Total Stresses:
uK hh 'uuK
v
h
v
hOT
'
Results of Deformed Mesh
Results of Yielded Zones
Results of Wall Deflection
Results of Wall Bending Moments
Problem 3: Three dimensional FEM:Problem 3: Three-dimensional FEM: Piles under lateral load
Problem 3: Three-dimensional FEM: Piles under lateral load
Three-dimensional (3-D) numerical model of the field pile lateral load test
(U t l 2011)(Urano et al., 2011)
The analysis was carried out using ABAQUS ver 6.10
• Soil Layer
– Hypoelastic Soil Model
RCC St l il d R ft• RCC or Steel pile groups and Raft
– Linearly elastic materials
Details of Field Test
ABAQUS Modeling
Structured mesh generated g
20-noded quadratic brick elements (C3D20R)
- Reduced integration-type elements
3-noded quadratic space beam elements (B32) 3-noded quadratic space beam elements (B32)
Used symmetry
Assigned proper boundary conditions
3D Soil-Pile-Raft Model in ABAQUS
Pile E (kN/m2)
Depth (m)
Soil Profile
Eo(kN/m2) Raft E (kN/m2)
= 2 5x107
Pile E (kN/m2)= 3.9x107
1.9 Fill 69160
1.5 loam 66500
= 2.5x10
1.4 Clay 39900
2.1 Clayey sand 31920
0 9 Sandyclay 1542800.9 Sandy clay 154280
0.55 Clay 154280
2.15 Medium sand 252700
Boundary Conditions
Back Left Boundary Boundary
UY = 0
BoundaryUY = 0
Plane of RightSymmetry
UY = 0
Right Boundary
UY = 0
Bottom SurfaceUx = UY = Uz = 0
Soil Model
Hypoelastic Soil Model
Stiffness reduction curve by Vucetic and Dobry, 1991
1
0.8
0.4
0.6
G/G
max
0
0.2PI=200
. PI=100
. PI=50
. PI=30
. PI=15 PI=NP
1.0E-06 1.0E-05 1.0E-04 1.0E-03 1.0E-02 1.0E-01
Cyclic shear strain, γ0 (%)
Pile Model
Piles were modeled using solid elements.Pile
Bending moment can not be measured from solid elements.
3-noded quadratic space beam elements.
Flexible beam
Flexural rigidity (scale down by 106)
Bending moment (scale up by 106)
Fl ibl b l thFlexible beam along thepile central axes
Input Loadp
500
600
700
200
300
400
500
oad
(KN
) Lateral Load
0
100
200
0 50 100 150 200 250
Lo
Load application point
0 50 100 150 200 250
Time (min)
point
Results and Discussion
0-200 -100 0 100 200
Bending moment (kNm)
-2
Fi ld
6
-4FieldAnalysis By Urano
et al.Analysis By
-8
-6
(m)
ABAQUS
-10
Dep
th
B di t l th il l thBending moment along the pile length
Pile–Raft–Reinforcement Body
Depth (m)
Soil Profile
Eo(kN/m2)(m) Profile (kN/m2)
1.9 Fill 69160
1 4 loam 665001.4 loam 66500
1.5 Clay 39900
2 1 Clayey fine 319202.1 sand 31920
0.9 Sady clay 1542800.55 Clay 154280
Medium
Grouted soil (Reinforcement Body)
2.15 Medium sand 252700
Pile, Raft and Reinforcement Body
Modulus of Elasticity (E0) of Reinforcement Body 1.5x106 kN/m2
With Reinforcement Body
0-200 -100 0 100 200
Bending moment (kN-m)
-2
0
A l i B U
-4
FieldAnalysis By Urano et al.
Analysis By
-6
RBy y
ABAQUS
-8
epth
(m)
-10 De
Position of Reinforcement Body
NRB
New Position for Reinforcement Body
RBReinforcement Body By
Urano et al.
Position for Reinforcement Body
New Position for Reinforcement Body
Reinforcement Body By Urano et al.
Effect of Position Grout
-200 -100 0 100 200
Bending moment (kN-m)
2
0
Analysis By ABAQUSNew Grout position
-4
-2
Test results with Old Grout position
Analysis By ABAQUSWith Old Grout position
-6
-8D
epth
(m)
-10
D