On the Simplified Methods Liquefaction Potential of

: Twenty Years Development of

Speaker: Jin

Prof., Civil Engineering Department, National Central University, Taiwan

Geotechnical Division Head, NCREE

Methods for Assessing Liquefaction Potential of Soils

Development of HBF Method

Jin Hung Hwang

Prof., Civil Engineering Department, National Central University, Taiwan

Geotechnical Division Head, NCREE

Flow Chart to Solve Liquefaction Problems

Start

Liquefaction

triggering

1

2

3

4

5

No

Yes

Simplified

method

5

6

Flow instability

Ground improvement

+

Deep foundation

End

Yes

Yes

No

Flow Chart to Solve Liquefaction Problems

Assessing consequences

1. Liquefaction potential index (LPI)

2. Post-liquefaction settlement

3. Reduction coefficients for kh and pu

4.Flow pressure and displacement

5. Building settlement due to liquefaction5. Building settlement due to liquefaction

6. floating check

Acceptable?

Yes

No

Outline

�Introduction to the simplified methods commonly used in the world

�Framework of NCREE-HBF method and its main �Framework of NCREE-HBF method and its main features

�Comparison Studies of the simplified methods

�Conclusions and suggestion

Outline

Introduction to the simplified methods commonly used

HBF method and its main HBF method and its main

Comparison Studies of the simplified methods

Conclusions and suggestion

Introduction

�The pioneer works on the simplified stress based method�Seed & Idriss (1971); Seed & Idriss (1979); Seed et al.(1985)

�Other followers in USA:Youd et al.(2001); Cetin et al.(2004); Idriss and Boulanger (2004); Boulanger and Idriss(2014)

� The similar works in Japan� The similar works in Japan� Iwasaki et al.(1982); JRA(1990,1996);

AIJ(2001)

�A new simplified method called HBF was developed in Taiwan�Basing on Chi-Chi earthquake cases

�A cyclic resistance ratio curve of HyperBolic

�Verification by local and worldwide liquefied and non

Introduction

The pioneer works on the simplified stress based method(1979); Seed et al.(1985)

Other followers in USA:Youd et al.(2001); Cetin et al.(2004); Idriss and Boulanger (2004); Boulanger and Idriss(2014)

Iwasaki et al.(1982); JRA(1990,1996); Tokimatsu and Yoshimi(1983);

A new simplified method called HBF was developed in Taiwan

HyperBolic Function(HBF) is used

Verification by local and worldwide liquefied and non-liquefied data

The Framework of the Simplified Assessing Soil Liquefaction Potential

CRRKKKCRR MFC ⋅⋅⋅== =

σασ

Black Box

MSFrA

CSRd

v

v

MFC

165.0 max ⋅⋅⋅

′⋅

== =

σσ

ασ

MCRRThe key relation is

The Framework of the Simplified Method for Assessing Soil Liquefaction Potential

5.7=Performance Check

MSF

15.7=

Verification by liquefied and

non-liquefied earthquake cases

sandclean for )(N vs. 6015.7=M

Parameters in the black box used in liquection triggering analysis

Description Parameter

CRRCyclic resistance curve

SPT borehole correction factors

Overburden stress

Diving energy

Rod length

Borehole size

Sampler

C

C

C

C

CSampler C

Fines correction FC

CRR stress adjustment

CRR initial stress adjustment

Magnitude scaling factor

Shear stress reduction factor

K

K

MSF

r

∗ :Neglectable

Parameters in the black box used in liquection triggering analysis

Parameter Notes

5.7CRR sandclean for )( vs CS6015.7 NCRR

NC

EC

RC

BC

C

NCCCCCN sBREN ⋅=601)(

∗∗∗

SC

(%) vs)(N 601 FC∆(%)FC

601601601 )(N)N()(N ∆+=CS

σK

αK

MSF

dr

5.7CRRKKCRR ⋅⋅= ασ

Number of cycles

Visco-elastic seismic response

∗∗

∗

Neglectable

Seed et al. Method �The pioneer and original

works (1971, 1979, 1983, 1985)

�All the parameters need to refer figures

Peak ground acceleration

maxa

to refer figures

�Innovative and intrinsic framework to solve problems of soil mechanics

�It is still good enough now since it was established 48 years ago

Earthquake induced cyclic

shear stress ratio

CSR ×= 65.0

Peak ground accelerationEarthquake

magnitudeSPT Eff. overburden Fines content

Ground propertiesEarthquake

maxwM 60N )kPa( vσ ′ (%) FC

( ) ( ) 60601 NCN vN ×σ′=規模修正因子

refer figureNC

dr

11.1)5.7

( −= wMMSF

15.7 ( NvsCRRrefer figure

5.7CRR

Cyclic resistance ratio

MSFCRRCRR ×=5.7

CSR

CRRFS =

Earthquake induced cyclic

shear stress ratio

d

v

v rg

a ×′σ

σmax

refer figure15.7

NCEER Method(2001)

�Based on Seed method (1985) with complicated equations to fit original CRR curves, stress reduction factor and fines correction factor correction factor

�This method is nearly the same as Seed et al’s method.

�The proposed method has been discussed by many experts in a NCEER workshop(1997)

�A representative of US method

Earthquake induced cyclic

Peak ground

acceleration

Moment

magnitudeSPT

Ground propertiesEarthquake

maxawM 60N

( ) 60'601 NP

Nv

a ×=σ

Magnitude

scaling factor

56.2)5.7

( −= wMMSF

32 +++= gxexcxaCRR

601601)()( NN

csβα +=

[0.5

190(76.1

0

=−=

=

ααα

EXP

[2.1

1000/(99.0

0.15.1

=+=

=

βββ

FC

%35

%355

%5

><

FC

FC

FC

≦

≦

%35

%355

%5

><

FC

FC

FC

≦

≦

Fines content

(%) FC

Eff. overburden

)kPa( vσ ′

Safety factor

CSR

CRRFS =

Earthquake induced cyclic

shear stress ratio

d

v

v rg

aCSR ×

′×=

σσmax65.0

Cyclic resistance ratio

MSFCRRCRR ×= 5.7

06714.3,05673.1,04285.3

,04136.6,03578.9,03721.4

,01248.1,02844.4,)(

1

601

4325.7

−=−−=−−=−=−=−−=

−−=−==++++

+++=

EhegEf

EeEdEc

EbEaNx

hxfxdxbx

gxexcxaCRR

CS

)001210.0006205.005729.04177.0000.1(

)001753.004052.04113.0000.1(25.15.0

5.15.0

zzzz

zzzr

d +−+−++−=

JRA Method (1996)

�Peak shear stress ratio

�Neglecting influence of earthquake magnitude

�Resistance depending on earthquake type

Design seismic coefficient

Peak shear stress ratio

earthquake type

�Conservative for soils with high N value

�Non-conservative for soils with high fines content and little plasticity

Sandy soils need to be assessed to obtain their liquefaction potential

(1)water table less than 10m below G.L. and saturated sandy soils within 20m below G.L.

(2)fines content FC≦35% 或FC＞35% with plasticity index Ip＜15%.

(3)mean grain D50≦10mm with 10% grain size D10≦1mm.

Design seismic coefficient

SPT Effective overburdenFines content ,

Peak shear stress ratio

zr

krL

d

v

vhcd

015.01−=′

××=σσ Earthquake type I： 0.1=wc

)4.0...(

)4.01.0...(

)10...(

0.2

67.03.3

0.1

L

L

Lw

R

R

.R

Rc

<≤<

≤

+=

Lw RcR ⋅=Cyclic resistance ratio

(%) FC

(mm) 50D (mm) 10D N )kPa( vσ ′

hck

Earthquake type II：

Safety factor

L

RFS =

)4.0...(0.2 LR<

( )

≤<

−××+= − )N...(14

14)...(N

14106.17.1/0882.0

7.1/0882.0

a

a

5.46

aa

a

LNN

NR

Sandy soils：

211 cNcNa +=

70σ'

170

v

1 +×= N

N

≤<≤

<≤

−+=

)...(60%

60%)...(10%

10%)...(0

1)20(

50)40(

1

1

FC

FC

FC

FC

FCc

≤<≤

−=

)...(10%

10%)...(0

18)10(

02

FC

FC

FCc

Gravelly soils：

( )[ ]15010 2log361.01 NDNa ××−=

AIJ Method (2002)

�Based on the method proposed by Tokimatsu and Yoshimi(1983)

�Equivalent shear stress ratio

�The function of cyclic resistance

1

2

�The function of cyclic resistance curve is also simple

�The framework is similar to that of USA

�Fines correction is non-conservative

Peak ground accelerationMoment

MagnitudeSPT Eff. overburden Fines content

Ground propertiesEarthquake

<≤<≤<

≤<≤

−×+−×+

−×=

%50

%20

%10

1%5

5%

11

)20(1.08

)10(2.06

)5(1.2

0

∆

FC

FC

FC

FC

FC

FC

FC

FC

N f

A72N (%) FC)kPa( 0vσ ′

wM

( )72'721

98NN

v

×=σ

Sandy soils susceptible to liquefaction

1.Alluvial saturated soil layers within 20m below G.L. and with fines content <35%

2.Soils with fines content>35% and with clay content<15% or with plasticity index Ip<

Pa

tf/m2

Safety factor

fa NNN ∆+= 721)(

Earthquake induced

average shear stress ratio

zr

rg

AMCSR

d

d

v

vw

015.01

)1(1.00

0

−=′

−=σσ

Cyclic Resistance ratio

CRR

CSR

CRRFL =

Cycli

c r

esis

tance

rati

o

CR

R

Adjusted N value (Na)

3020100

0.1

0

0.2

0.3

0.4

0.5

0.6Shear strain γ=5%

+=

80

16

100

162565.0

aNN

CRR

NCREE- HBF

�Based on the liquefied and non-liquefied cases of 1999 Chiearthquake

�Use of HyperBolic Fuction (HBF) to fit cyclic resistance ratio curve�Use of HyperBolic Fuction (HBF) to fit cyclic resistance ratio curve

�The equations used are as elegant and simple as possible

�Soils with Ip>7 are regarded as clayliquefaction

Method (2019)

liquefied cases of 1999 Chi-Chi

(HBF) to fit cyclic resistance ratio curve(HBF) to fit cyclic resistance ratio curve

The equations used are as elegant and simple as possible

>7 are regarded as clay-like soils and unsusceptible to

Framework of HBF Methodand Its Main Features

Feature 1: All the formula used to

calculate cyclic resistance ratio (CRR),

stress reduction factor (rd), correctionstress reduction factor (rd), correction

factor of fines content (Ks),

magnitude scaling factor (MSF), and

overburden stress correction factor

(CN) are as simple and elegant as

possible so that engineers can prevent

calculation errors or coding mistakes

that commonly encountered in

practices.

Method

maxawM 60N )kPa( vσ ′

8.1)

5.7(

−= wMMSF

( ) 60'601 NP

Nv

a ×=σ

07.01

10

0.1

10

×+=

>=

≤

FCk

FCIf

k

FCIf

s

s

(%) FC

d

v

v rg

aCSR ×

′×=

σσmax65.0

mzm

mz

z

zrd

2010

10

03.02.1

01.00.1

≤<≤

−−

=

MSFCRRCRR ×= 5.7

CSR

CRRFS =

39/)(1

)(0035.008.0

)()(

601

6015.7

601601

cs

cs

scs

N

NCRR

NkN

−×+=

×=

The Main Features of HBF Method

Feature 2: The method fits the CRR (Mw=7.5) curve of clean sand by using a hyperbolic function

The Main Features of HBF Method

The Main Features of HBF Method

Feature 3: The HBF method believesindex PI>7 are clay-like materialsexhibit cyclic softening behavior underexhibit cyclic softening behavior underin consistent with the suggestion byas the definition of clay soil inCasagrande (1948), and the studycohesive soils (Hwang et al., 2000).

The Main Features of HBF Method

believes the cohesive soils with plasticitymaterials which will not liquefy and only

under seismic shaking. This criteria isunder seismic shaking. This criteria isBoulanger and Idriss (2006) as wellthe plasticity chart proposed byon the cyclic resistance of Taipei

The Main Features of HBF Method

Feature 4: A lot of liquefied cases ofplastic ML soils in central Taiwanobserved during Chi-Chi earthquake (et al., 2004) and then resulted in severalexperimental researches on these non-plasticexperimental researches on these non-plasticsoils (Huang et al., 2007; Chen et al, 2014The conclusion is the correction ofcontent on CRR is suggested to be aconservative, especially for non-plasticsoils.

The Main Features of HBF Method

of non-were

(Uengseveralplastic

0.3

0.4

0.5

CR

R

NCEER

JRA

AIJ

HBF

(N1)60cs=15

plastic2014).Fines

a littleplastic ML

0.0

0.1

0.2

0 20 40 60 80 100

CR

R

細料含量 (%)

A Comparison of Cyclic Strength Curve

Method Cyclic strength curve (CRR

NCEER(2001) CRR7.5 = 134 − �N1 60cs + �N135

CRR7.5 = exp ��N1 60 �1+0.004FCCetin et al.(2004)

HBF(2012)

Boulanger and

Idriss (2014)

CRR7.5 = 0.08 + 0.0035�N1 1 − �N1 60cs

CRR7.5 = exp ��N1 60cs14.1 + ��N

CRR7.5 = exp ��N1 60 �1+0.004FC

A Comparison of Cyclic Strength Curve

Cyclic strength curve (CRR7.5)

�N1 60cs135 + 50�10�N1 60cs + 45 2 − 1200

FC −29.53 ln �Mw −3.7 ln σ vP a %+0.05FC +16.85+2.7Φ−1�0.15,0,1 13.32 (

� 60cs cs /39

��N1 60cs126 (2 − ��N1 60cs23.6 (3 + ��N1 60cs25.4 (4 − 2.8*

FC −29.53 ln �Mw −3.7 ln σ vP a %+0.05FC +16.85+2.7Φ−1�0.15,0,1 13.32 (

A Comparison of Stress Reduction Factor

Method Stress

NCEER(2001) rd = 1.0 − 0.41131.0 − 0.4177z0.

z < 20m, rd = 11Cetin et al.(2004)

HBF(2012)

Boulanger and

Idriss (2014)

z < 20m, rd = 11

rd = 1 − 0.01rd = 1.2 − 0

rd =α�z = −1.012β�z = 0.106

A Comparison of Stress Reduction Factor

Stress reduction factor (rd)

4113z0.5 + 0.04052z + 0.001753z1.5.5 + 0.5729z − 0.006205z1.5 + 0.00121z2

1 + −23.013−2.949amax +0.999Mw +0.0525VS12m∗16.258+0.201e0.341 �−z +0.0785 V s12m∗ +7.586

1 + −23.013−2.949a +0.999M +0.0525V∗

34

1 + −23.013−2.949amax +0.999Mw +0.0525VS12m∗16.258+0.201e0.341 �−z +0.0785 V s12m∗ +7.586

1 + −23.013−2.949amax +0.999Mw +0.0525VS12m∗16.258+0.201e0.341 �0.0785 V s12m∗ +7.586

01z from z�m ≤ 10 0.03z from 10 < z�m ≤ 10

= exp9α�z + β�z M:012 − 1.126sin z11.73 + 5.133%

106 + 0.118sin z11.28 + 5.142%

A Comparison of Fines Content Correction

Method Fines content correction (C

NCEER(2001)α = 0; β = 0 α = exp91.76 − �190α = 5; β = 1.2

CFINESCetin et al.(2004)

HBF(2012)

Boulanger and

Idriss (2014)

Ks = 1 Ks = 1

CFINES

�N1∆�N1 60 =

A Comparison of Fines Content Correction

Fines content correction (CN)

�N1 60cs = α + β�N1 60 for FC ≤ 5 190/FC2 :; β = 0.99 + FC 1.5

1000 for 5 < FC ≤ 35 for FC > 35

= 1 + 0.004FC + 0.05 FC�N %

34

�N1 60cs = Ks�N1 60 for FC ≤ 10+ 0.07√FC − 10 for FC > 10

= 1 + 0.004FC + 0.05 FC�N1 60 %

Lim:5%≤FC≤35%

� 1 60cs = �N1 60 + ∆�N1 60

= exp 1.63 + 9.7FC +0.01−C 15.7FC +0.01D2%

COMPARISON STUDIES OF THE HBF METHOD�A total of 669 sets of SPT-based cases were

�302 sets of Chi-Chi earthquake data collected by Hwang and Yang (2001)

�367 sets of data collected by Youd et al. (1997

� Summary of comparison using success rate of prediction

Success rate Seed (1985) NCEER (2001)

FC≤10 96% 97%

Liquefied

FC≤10 96%(143/149) 97%(145/149)

10<FC≤30 88%(136/155) 92%(142/155)

FC>30 91%(63/69) 97%(67/69)

Non-

liquefied

FC≤10 59%(62/105) 57%(60/105)

10<FC≤30 88%(130/147) 86%(126/147)

FC>30 91%(40/44) 82%(36/44)

Liquefied 92%(342/373) 95%(354/373)

Non-liquefied 78%(232/296) 75%(222/296)

Total 86%(574/669) 86%(576/669)

COMPARISON STUDIES OF THE HBF METHODbased cases were collected

data collected by Hwang and Yang (2001)

et al. (1997)

Summary of comparison using success rate of prediction

AIJ (1983) JRA (1996) JRA (1990) HBF (2005)

99% 99% 87% 95%99%(147/149) 99%(147/149) 87%(130/149) 95%(142/149)

90%(140/155) 97%(150/155) 81%(126/155) 98%(152/155

94%(65/69) 97%(67/69) 81%(56/69) 99%(68/69

50%(53/105) 43%(45/105) 61%(64/105) 64%(67/105

77%(113/147) 67%(99/147) 79%(116/147) 80%(117/147

75%(33/44) 75%(33/44) 64%(28/44) 75%(33/44

94%(352/373) 98%(364/373) 84%(312/373) 97%(362/373

67%(199/296) 60%(177/296) 70%(208/296) 73%(217/296

82%(551/669) 81%(541/669) 78%(520/669) 87%(579/669

A comparison of Microof Soil Liquefaction Potential

�Using liquefaction potential index(PL) by Iwasaki et al. (1978)

� The area of different liquefaction potentials in Taipei basin by different methodsbasin by different methods

Method

Liquefaction potential

Low Middle

Area

HBF (2005) 101.3/39% 95.0/36%

NCEER (2001) 111.7/43% 98.0/37%

JRA (1996) 76.3/29% 104.6/40%

Micro-zonation MapSoil Liquefaction Potential in Taipei basin

Liquefaction potential Total

Middle High

Area(km2)/Percentage

/36% 65.6/25% 261.9/100%

/37% 52.1/20% 261.9/100%

/40% 80.9/31% 261.9/100%

A comparison of Microof Soil Liquefaction Potential in Taipei basin

HBF(2005) NCEER(2001)

A comparison of Micro-zonation Mapof Soil Liquefaction Potential in Taipei basin

NCEER(2001) JRA(1996)

Conclusions and

�All the methods are acceptable in engineering practice.

�Each method is an integrated system. You can not replace one of the elements in A method with that in B method.

�The prediction accuracy of new methods are not necessarily better than the old methods.

�The prediction accuracy of more complicated methods are not necessarily better than simple methods

�The main features and the merits of HBF method are

�The HBF method has more simple and elegant formula

�The performance of HBF method is as good as NCEER method

�It is suggested that the HBF method can be an acceptable alternative for assessing liquefaction potential in practice

Conclusions and suggestions

All the methods are acceptable in engineering practice.

Each method is an integrated system. You can not replace one of the elements in A method with that in B method.

The prediction accuracy of new methods are not necessarily better

The prediction accuracy of more complicated methods are not simple methods.

main features and the merits of HBF method are illustrated

simple and elegant formula

performance of HBF method is as good as NCEER method

method can be an acceptable alternative for assessing liquefaction potential in practice beside USA and Japan

Thank you very muchThank you very muchfor

your attentions

Thank you very muchThank you very muchfor

your attentions