(매우중요) KL 11

Design of Retaining Wall
and Support System for

Deep Basement
Ir. TAN Yean Chin

G&P Geotechnics Sdn Bhd
www.gnpgroup.com.my
March, 2014
INTRODUCTION
Deep basement construction

Urban areas for parking space
Infrastructures, e.g. KVMRT
Riskassociated with deep basement

construction high!
FAILURES of DEEP EXCAVATION
SOIL
PARAMETERS
SOIL PARAMETERS
Some important soil parameters

related to retaining wall and support
system design:
Shear strength parameters (su, φ’ & c’)
Soil permeability
Soil stiffness
SOIL PARAMETERS
Shear strength parameters (φ’ and c’)
Commonly obtained from CIU for effective
stress strength parameters
Commonly obtained using in-situ test
methods (e.g. field vane shear test) for
total stress strength parameters, su
Finite element analysis – requires proper
understanding of the constitutive soil
models
SOIL PARAMETERS
Soil permeability
Important in order to choose modelling in
drained or undrained condition
In-situ tests (rising, falling or constant head)
recommended
Values obtained from tests should always be
compared to published values (e.g. BS8004)
BS8004: 1986
SOIL PARAMETERS
Soil stiffness
Important parameters for retaining wall design
BUT difficult to obtain reliably
In Malaysia, sometimes based on empirical
correlations
Laboratory tests unreliable and values
obtained significantly smaller than
appropriate values for retaining wall design
Designer should be aware of small-strain
nature of retaining wall design
SOIL PARAMETERS
Soil stiffness
Seismic tests or seismic piezocone appears
promising – future direction
Basis of empirical correlations should be
understood – e.g. local soil conditions,
constitutive model used, etc.
Example, correlations in Kenny Hill formation
using hardening soil model of PLAXIS
software
DESIGN OF
RETAINING WALLS
DESIGN CONSIDERATIONS
Overall stability
Basal heave failure
Hydraulic failure
Axial stability
Finite element analysis
Ground movement associated with
excavation
OVERALL STABILITY
To ensure sufficient wall embedment

Overturning of wall and overall slope
stability
Adequate factors of safety (e.g. 1.4 for
high-risk-to-life structures)
EUROCODE 7
Overall stability
Basal heave failure
Hydraulic failure
Axial stability
excavation
BASAL HEAVE FAILURE

Critical for deep excavation in soft ground
Analogous to bearing capacity failure, only in
reverse
Available methods:
Terzaghi (1943) – shallow and wide excavations
Bjerrum & Eide (1956) – deep and narrow and
excavations
Moment equilibrium - adequate for routine design
MOMENT
EQUILIBRIUM
METHOD
BASAL HEAVE FAILURE

Required factor of safety for moment
equilibrium method
1.2 (Kohsaka & Ishizuka, 1995)
Vertical shear resistance along retained
ground shallower than the excavation is
ignored
Overall stability
Basal heave failure
Hydraulic failure
Axial stability
excavation
HYDRAULIC FAILURE
Base instability caused by piping
Seepage due to high groundwater level
Available methods
Terzaghi’s method
Critical hydraulic gradient method
HYDRAULIC FAILURE CHECKS
HYDRAULIC FAILURE
Terzaghi’s method recommended
Based on latest research by Tanaka &
Verruijt (1999)
Factor of safety required – 1.2 to 1.5
HEAVING DUE TO ARTESIAN PRESSURE
HYDRAULIC FAILURE
Heaving due to artesian pressure
Factor of safety – 1.0 to 1.2

Smaller FOS sufficient as it did not
consider shear strength or adhesion
strength of the ground and retaining wall
Overall stability
Basal heave failure
Hydraulic failure
Axial stability
excavation
AXIAL STABILITY
Simple check but is often overlooked
Factor of safety
required – 2.0
FINITE ELEMENT
ANALYSIS
FINITE ELEMENT ANALYSIS

Soil-structure interaction important for
deep basement retaining wall design
Commercial finite element software easily
available and very user friendly (e.g.
PLAXIS, CRISP, etc.)
Understanding and proper use
important!!!
Results using the same soil parameters
and same software by different people!
FINITE ELEMENT ANALYSIS

Some important considerations in FEM:
Locations of the boundaries of the problem
Details of mesh
Modelling of stages of construction
Modelling of interfaces
Use of suitable constitutive soil model
Use of appropriate soil parameters, especially
empirical parameters
GEOMETRICAL DATA
Provision for over-excavation
for cantilever walls - 10% of the wall height above
excavation level, limited to a maximum of 0.5m
for a supported walls - 10% of the distance between
the lowest support and the excavation level, limited to
a maximum of 0.5m
Can be reduced when excavation surface can
be controlled reliably throughout the excavation
works
GEOMETRICAL DATA
Water levels
Flood level should be taken into consideration
for flood prone areas
Surcharge
Minimum surcharge of 10kPa – for
construction loads and unforeseen
circumstances
CONSTITUTIVE SOIL MODELS

Various constitutive soil models, e.g.
Mohr-Coulomb, Cam Clay, Hardening Soil,
Soft Soil, etc.
Proper understanding and limitations of
each model important!
Incorrect use of soil models in Nicoll Highway!
Overestimation
of shear
strength!!!
Stress path using

Mohr-Coulomb Stress path
model of real soil
FEM ANALYSIS OF LIMIT STATE
Eurocodes have replaced British

Standards
Current design of retaining wall for deep

basement mostly based on “working state
design”
Three (3) schemes to perform limit states

design using FEM:
(A) Perform all the calculations with design (factored)
values of ground and action parameters
(B) Simulate the whole stress history using ground
parameters at characteristic levels
- check the safety at the relevant stages
(C) Simulate the whole stress history using ground
parameters at characteristic levels and multiplying
these values by the load factor (which then in fact
acts as a model factor on the effects of actions).
Important to evaluate limit state:

Part of the reason for factors of safety is to
cover human error
Limit state analyses investigate unrealistic
states, especially in the Ultimate Limit State
(ULS) analysis
TO ENSURE THAT LIMIT STATE IS

SUFFICIENTLY UNLIKELY TO OCCUR
GROUND MOVEMENT
ASSOCIATED WITH
EXCAVATION
GROUND MOVEMENT
Deep excavation include a substantial
component of horizontal strain
Damage estimation should include both

angular distortion and horizontal strain
GROUND MOVEMENT INDUCED BY
DEEP EXCAVATION

DEEP EXCAVATION

DEEP EXCAVATION
Gap
(75mm) Settlement
Void
(150mm)
Damage criteria by Boscardin & Cording, 1989

CATEGORY ANGULAR HORIZONTAL

DISTORTION, STRAIN,
β (x 10-3) εh (x 10-3)
Negligible < ~ 1.1 > 0.5
Very slight ~ 1.1 < β < ~ 1.6 0.5 < εh < 0.75
Slight ~ 1.6 < β < ~ 3.3 0.75 < εh < 1.5
Moderate ~ 3.3 < β < ~ 6.7 1.5 < εh < 3.0
Severe > ~ 6.7 > 3.0

Before excavation
After excavation
h
β = ∆/L
∆
L = length of building εh = h/L
E.g.
Building length = 10m
Allowable lateral movement = 15mm
(εεh = 1.5 x 10-3)
CASE HISTORIES
Case History 1 :
32m Deep Excavation for
KVMRT – Blue Line
(Cochrane Underground Station)
Rock Stabilisation Measures

Construction Stages 1-
1-4
Construction Stages 5-
5-8
Case History 2 :
28.5m Deep Excavation for
Berjaya Times Square,
Jalan Imbi
Berjaya Times Square

Excavated depth
24.5m - 28.5m (6-level basement)
Retaining wall
1.2m thick diaphragm walls
Support system
Prestressed Ground Anchors
TYPICAL SUBSOIL PROFILE

FINITE ELEMENT MODELLING

Wall Relative Lateral Displacement (mm) Wall Relative Lateral Displacement (mm) Wall Relative Lateral Displacement (mm)
-10 0 10 20 30 40 -10 0 10 20 30 40 -10 0 10 20 30 40
0 0 0
Stage 1
Stage 2
Excavate to R.L.35.0m
-10 -10
-10
Stage 3
-20 -20 Excavate to R.L.27.0m
Depth(m)
Depth (m)
-20
Depth (m)
-30 -30
-30
-40 -40
Measured Profile -40
FEM Back Analysed

-50 -50
Wall Relative Lateral Displacement (mm) Wall Relative Lateral Displacement (mm) Wall Relative Lateral Displacement (mm)
-10 0 10 20 30 40 -10 0 10 20 30 40 -10 0 10 20 30 40 -50
0 0 0
-10 -10 -10
Stage 4
Stage 4
-20 -20
Stage 6 -20
Depth (m)
Depth (m)
Depth (m)
-30 -30 -30
-40 -40 -40
WALL A -50 -50 -50

Berjaya Times Square

Hardening Soil Model of PLAXIS able to
model the problem sufficiently accurate
From FEM back-analysis, the correlations

between soil stiffness (E’) and SPT ‘N’ as
follows:
E’ = 2000*SPT‘N’(kN/m2)
E’ur = 3*E’ = 6000*SPT’N’(kN/m2)
Human Size
Case History 3 :
30m Deep Excavation for
Solaris Dutamas, KL
Solaris Dutamas
Typical
Simplified
Borelog Profiles
Initial basement wall

Original CBP
Wall with
Temporary
Ground
Anchors
Temporary
Ground Anchor
Original CBP
Wall with
Temporary
Ground
Anchors
Typical finite element model (PLAXIS)

Measured
Predicted ground movement

Soil nails subjected to verification and
proof test based on FHWA, 1998
Criteria on creep movement
Criteria on theoretical elongation
Based on verification tests on

preliminary soil nails
Ground-grout bond = 3 to 5*SPT-N (in kPa)
THANK YOU
Q&A
Y.C. TAN
www.gnpgroup.com.my
March, 2014

(매우중요) KL 11

Uploaded by

Copyright:

Available Formats

(매우중요) KL 11

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(매우중요) KL 11

Uploaded by

Copyright:

Available Formats

Design of Retaining Wall

and Support System for

Ir. TAN Yean Chin

Deep basement construction

Riskassociated with deep basement

Some important soil parameters

To ensure sufficient wall embedment

BASAL HEAVE FAILURE

BASAL HEAVE FAILURE

Heaving due to artesian pressure

Factor of safety – 1.0 to 1.2

FINITE ELEMENT ANALYSIS

FINITE ELEMENT ANALYSIS

CONSTITUTIVE SOIL MODELS

Stress path using

FEM ANALYSIS OF LIMIT STATE

Eurocodes have replaced British

Current design of retaining wall for deep

FEM ANALYSIS OF LIMIT STATE

Three (3) schemes to perform limit states

FEM ANALYSIS OF LIMIT STATE

Important to evaluate limit state:

TO ENSURE THAT LIMIT STATE IS

Damage estimation should include both

GROUND MOVEMENT INDUCED BY

GROUND MOVEMENT INDUCED BY

Damage criteria by Boscardin & Cording, 1989

CATEGORY ANGULAR HORIZONTAL

Slight ~ 1.6 < β < ~ 3.3 0.75 < εh < 1.5

Moderate ~ 3.3 < β < ~ 6.7 1.5 < εh < 3.0

Severe > ~ 6.7 > 3.0

Rock Stabilisation Measures

Berjaya Times Square

TYPICAL SUBSOIL PROFILE

FINITE ELEMENT MODELLING

-20 -20 Excavate to R.L.27.0m

Measured Profile -40

FEM Back Analysed

-10 -10 -10

WALL A -50 -50 -50

Berjaya Times Square

From FEM back-analysis, the correlations

Initial basement wall

Typical finite element model (PLAXIS)

Predicted ground movement

Based on verification tests on

You might also like