ssrn-4385752

Download as pdf or txt
Download as pdf or txt
You are on page 1of 25

Settlement of Combined Piled Raft Foundation of a Nuclear Power

ed
Plant in Non-liquefiable and Liquefiable Soils

iew
Author 1
Bal Krishna Maheshwari
Professor, Dept. of Earthquake Engineering
Indian Institute of Technology Roorkee

ev
Uttarakhand, 247667, India
Email - bk.maheshwari@eq.iitr.ac.in
*Corresponding Author

r
Author 2
Mohd. Firoj
Research Scholar
Department of Earthquake Engineering
er
pe
Indian Institute of Technology Roorkee
Uttarakhand, 247667, India
Email: mfiroj@eq.iitr.ac.in
ot
tn
rin
ep
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 Settlement of Combined Piled Raft Foundation of a Nuclear
2 Power Plant in Non-liquefiable and Liquefiable Soils

ed
3 Bal Krishna Maheshwari1*, and Mohd. Firoj2

iew
4 1Professor,Dept. of Earthquake Eng., Indian Institute of Technology Roorkee, Uttarakhand,
5 247667, India. Email - bk.maheshwari@eq.iitr.ac.in
6 2Research Scholar, Department of Earthquake Engineering, Indian Institute of Technology

7 Roorkee, Uttarakhand, 247667, India. Email: mfiroj@eq.iitr.ac.in


8 *Corresponding Author

v
9 Abstract

re
10 The settlement analysis of a nuclear power plant (NPP) resting on the combined pile raft
11 foundation (CPRF) under static and seismic loads is very important. However, it has not been
12 yet reported in the literature as mostly NPPs are founded on raft foundation rather than on
13 CPRF. In this paper, settlement has been examined in detail. The geometrical nonlinearity at
14
er
the interface of raft-soil and pile-soil is considered using the coulomb friction. The piles are
15 rigidly connected to the raft. The nonlinear behaviour of the soil in the non-liquefiable soil is
16 modelled using elasto-plastic Drucker Prager cap model and CycLiq model for the liquefiable
pe
17 soil. For this, user defined material (UDM) is implemented in the FLAC3D material library.
18 Thus, a complete 3D finite difference model of a NPP considering raft-pile-soil-structure
19 interaction (RPSSI) for nonlinear settlement analysis is developed. After successful validation
20 of the present CPRF numerical model with the literature, it has been used for further study
21 under real earthquake acceleration–time history. Next CPRF with NPP structure is studied
ot

22 under static loading which is further extended for the nonlinear time domain settlement
23 analysis. The effects of variation in various parameters i.e., spacing between piles, cohesion,
24 friction angle and Poisson's ratio of soil on the settlement are examined and found to be
tn

25 significant. The predominant frequency of the earthquake significantly affects the response of
26 the CPRF. It was inferred that the interface nonlinearity has greater significance in the
27 liquefiable soil as compared to the non-liquefiable soil. Higher differential settlement in the
28 liquefiable soil is attributed to the reduction in the stiffness of the soil. The findings of this
rin

29 study could be useful to frame the general guidelines for the economical design of NPP-CPRF
30 systems.
31
32 Keywords: Seismic Settlement, Combined Piled Raft Foundation, Nuclear Power Plants, Soft
ep

33 Soil, Liquefaction.

34 1. Introduction
Pr

35 The foundation constructed in the seismic-prone area is subjected to additional settlement due
36 to horizontal shaking; this type of settlement is called seismic settlement. In the routine design

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 of shallow and deep foundations, this seismic settlement must be rigorously calculated to
2 estimate the total settlement reasonably. The available analytical approaches (Tokimatsu and

ed
3 Seed 1987; Pradel 1998) do not consider the effect of the structural applied load on the seismic
4 settlement of the foundation in sand. Thus, these methods cannot calculate the seismic
5 settlement of the combined piled raft foundation (CPRF) of a nuclear power plant (NPP)
6 structure.

iew
7 In general, the design of pile group is based on the concept that piles carry the load of the
8 superstructure and the cap provides the rigid platform to group the piles and distribute the load
9 to piles uniformly. Whereas in CPRF, the load is shared by both pile group and raft”. The key
10 questions arise in the design of a CPRF of NPP structure i.e., the number of piles, load carried
11 by raft-piles and differential settlement due to the dynamic loading (earthquake loading).

v
12 Clancy and Randolph (1993) first reported the approximate analytical solution for the
13 combined effect of piles and raft. Poulos (2005) studied the design of CPRF on soft soil in

re
14 Mexico City. The conditions involved in the selection of CPRF is reported by Raut et al. (2014).
15 Poulos (2012) reported the design issues of piled-raft foundation. Xu et al. (2014)
16 recommended that ground improvement beneath the pile bottom reduced the settlement of
17 CPRF subjected to cyclic loading. Nguyen et al. (2013) studied the centrifuge modelling of the
er
18 piled raft foundation under vertical loading. Sinha and Hanna (2016) studied the settlement of
19 piled raft foundations in the static case. Various researchers (Kitiyodom and Matsumoto 2002;
20 Zhang et al. 2010; Osman and Randolph 2012; Tehrani et al. 2016; Salciarini et al. 2013) used
pe
21 the simplified technique to compute the behaviour of pile group and single pile without
22 considering the interaction between the raft-soil, pile-soil, raft-structure. Furthermore, these
23 studies are limited to static vertical and lateral loading. Gu et al. (2017) investigated the
24 settlement of the piled raft foundation due to the cyclic high-speed (300 km/h) train loading.
25 Luo et al. (2018) proposed the 3D boundary element method for the normalized vertical
ot

26 settlement of the CPRF. However, in the recent advancement of CPRF, some researchers
27 (Azizkandi et al. 2018; Bhaduri and Choudhury 2021; Zhu et al. 2022) carried out the dynamic
28 response of the CPRF foundation under the earthquake excitation. Further, seismic settlement
tn

29 is not examined in these studies, which is a major concern for the seismic design of CPRF. Lee
30 et al. (2021) carried out the soil-structure interaction analysis of NPP structure found on the
31 rigid base foundation; however, ignore the settlement of nuclear building. Li et al. (2022)
32 investigated the seismic response of a nuclear building resting on the CPRF using the centrifuge
rin

33 test. However, due to the limitation of the centrifuge test, seismic settlement of the NPP
34 structure rested on CPRF is ignored. Tang et al. (2022) provided the stone column to reduce
35 the seismic response of the pile-supported Wharves'. Jia et al. (2023) carried out the 1g-shaking
36 table test to reveal the failure mechanism of the pile group in liquefiable soil. The recent
ep

37 advancement is significantly carried out for the lateral seismic response of the pile group
38 foundation using the centrifuge test. In addition, a detailed review of the literature has indicated
39 that not much attention is given in earlier studies to comprehend or compute the seismic
40 settlement of CPRF of a NPP structure.
Pr

41 Many countries have successfully applied the CPRF under the tall building structures
42 (Katzenbach et al. 2005). Recently, researchers (Firoj and Maheshwari 2022a, b; Li et al. 2022)
43 used the CPRF foundation for the seismic design of NPP structures. A detailed literature review

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 by the authors (Maheshwari and Firoj 2021) has been carried out on the seismic behaviour of
2 the NPP structure, considering SSI effects. Further, Maheshwari and Firoj (2022) examined

ed
3 some major issues which were not discussed in the past literature, i.e.: (i) geometrical
4 nonlinearity at the interface; (ii) material nonlinearity using the advanced constitute soil model
5 for the liquefiable and non-liquefiable soils; However, the seismic settlement of the CPRF in
6 liquefiable and non-liquefiable soil was not studied, which has been considered in the present

iew
7 study.
8 Further, in past studies, the combined piled-raft foundation is reported mostly under vertical
9 static loading. However, for seismic loads, it is must to analyze the CPRF under the lateral
10 loading, which may lead to the differential settlement for a heavy structure such as NPPs in the
11 soft soil. As far author's knowledge, so far, no study related to the differential settlement of

v
12 piled raft foundations due to earthquake loading has been reported. In this study, emphasis is
13 made to examine the earthquake induced settlement of CPRF without and with considering the

re
14 NPP structure in both the non-liquefiable and liquefiable soils.

15 2. Finite Difference Modelling of CPRF er


16 A numerical model of CPRF is prepared where raft, piles, and soil are modeled using the finite
17 difference grid formulation, which allows rigorous handling of pile-soil and raft-soil
18 interaction. This is because the partial differential equation of motion is solved in the finite
pe
19 difference model. Thus, the geometrical nonlinearity can be better captured in finite difference
20 method (FDM) as compared to the finite element method (FEM) because in former high-order
21 approximations can be easily obtained. Moreover, FDM require less computational power. The
22 finite difference model is prepared in the FLAC 3D.
23 In the CPRF system, raft and soil are modeled using the 8 noded grid elements with linear
ot

24 shape function while the piles were modeled using the 2 noded embedded beam elements. Fig.
25 1 illustrates the schematic view of the numerical model for the CPRF corresponding to 6D
26 center-to-center spacing (where D is the pile diameter, which is considered 1m in Fig. 1). The
tn

27 number of piles is 4×4 (in a square pattern), i.e., total 16 piles. In order to account for the
28 interfaces between the raft, pile, and soil, master and slave surface contacts are used. Where
29 the raft and piles are treated as master surfaces and soil is assumed as slave surfaces.
30 Furthermore, at the pile-soil interface, surface-to-surface hard contact is assumed in the normal
rin

31 direction, while penalty constraint using the friction coefficient is applied in the tangential
32 direction due to its high accuracy (Hibbit et al. 2007). A detailed description to deal with the
33 interface nonlinearity is given in Firoj and Maheshwari (2022a). The inelastic behaviour of soil
34 is modeled using the well-known elasto-plastic model i.e., Drucker Prager cap model for non-
ep

35 liquefiable soil and using CycLiq for the liquefaction model. A detailed description of these
36 models can be found in Firoj and Maheshwari (2022a) and Maheshwari and Firoj (2022). For
37 the sesimic settlemnt
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
ed
iew
1
2 Figure 1: Schematic view of combined piled raft foundation having 4×4 piles @ 6D spacing

v
3. Absorbing Boundary

re
3

4 For the dynamic SSI problem, it is important to model the far-field soil medium. The reduction
5 in the size of soil domain makes the solution efficient, but if the absorbing boundary is not
6 selected accurately, then it seriously affects the accuracy of results and, thus the performance
er
7 of the structure (Kontoe et al. 2007). In the present analysis, the Kelvin element boundary is
8 used to model of soil at the far field, which absorbs the outgoing waves preventing these from
9 reflecting back to the structure. For this boundary, at the outer nodes, the spring-dashpot
pe
10 provided by Novak and Mitwally (1988) are attached and evaluated using Eq. 1.
𝐺
11 𝑘 = 2𝜋𝑅𝑜(𝑆1 + 𝑖𝑆2) (1)
12 Where, 𝑘 is complex stiffness, G = shear modulus of soil, 𝑅𝑜= radius of Kelvin element node
13 from the source of shaking; in equation (1) real part corresponds to stiffness, and the imaginary
ot

14 part represents the viscous dashpot. The values of dimensionless parameters S1 and S2 can be
15 evaluated as reported by Novak and Mitwally (1988); The parameters S1 and S2 depends on the
16 Poisson's ratio and dimensional frequency ao. The dimensionless frequency is defined as the
tn

17 ratio of ro/Vs. where, ro is the radial distance from the source of vibration to the truncated
18 boundary.  is the frequency of excitation. Vs is the shear wave velocity of soil.
rin

19 4. Damping of CPRF system

20 To capture the dynamic behaviour of CPRF model, both material and radiation damping are
21 considered. Thus, total damping of the system is defined as (Chopra, 2007)
ep

22 𝐶 = [𝐶𝑟] + [𝐶𝑚] (2)


23 Where, [𝐶𝑟] is the radiation damping matrix which has the nonzero terms only at diagonal
24 corresponding to the boundary where the Kelvin elements are attached. [𝐶𝑚] is the material
25 damping matrix, for which Rayleigh damping is considered and can be found using mass matrix
Pr

26 [M] and stiffness matrix [K] as follows:


27 [𝐶𝑚] = 𝛼1[𝑀] + 𝛼2[𝐾] (3)

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 Where, mass proportional (𝛼1) and stiffness proportional (𝛼2) parameters are calculated using
2 the following equation (Chopra 2007)

ed
2𝜁𝑓𝑖𝑓𝑗 2𝜁
3 𝛼1 = , 𝛼2 =
(𝑓𝑖 +𝑓𝑗) (𝑓𝑖 +𝑓𝑗)
4 Where, 𝑓𝑖 and 𝑓𝑗 are the first two frequencies in the direction of shaking. 𝜁 is damping ratio
5 and assumed 10% for soil and 5% for concrete.

iew
6 5. Validation of proposed model

7 For the validation of the proposed CPRF foundation of NPP structure, a 4×4 piled raft model
8 is modeled using the finite difference method as shown in Fig. 1. The same model was used as

v
9 benchmark problem by researchers (Ata et al. 2015; Patil et al. 2021). Table 1 shows the
10 material properties used in the present model (Sinha and Hanna, 2016). Fig. 2 represents the

re
11 comparison of present study with the published literature for the 6D spacing which shows good
12 agreement. However, the computation time used in the present study is significantly reduced
13 due to modelling of piles as a beam element instead of solid element. Firoj and Maheshwari
14 (2022a) validated NPP model for the seismic analysis, though not for the settlement. Therefore,
er
15 the model shown in Fig. 1 is further used for parametric and seismic analyses of CPRF model.

16 Table 1: Material and Geometrical properties of Numerical Model


pe
Parameter Soil Raft Pile
Size (m) 120×120×30 24×24×2 1(D)×15(L)
Unit Weight (kN/m3) 19 25 25
Modulus of Elasticity E (MPa) 54 34,000 25,000
Poisson's ratio (ν) 0.15 0.2 0.2
ot

Cohesion, c (kPa) 20 - -
Friction Angle of soil,  (degree) 20 - -
Load intensity (kPa)
tn

0 100 200 300 400 500


0
Settlement (m)
rin

0.1

0.2
ep

0.3
Present, S = 6D
Sinha and Hanna (2016), S = 6D
0.4
17
18 Figure 2: Validation of CPRF model
Pr

19 6. Static Settlement of CPRF without NPP

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 The static settlement of the CPRF foundation is an important concern in the design, which has
2 not been examined in literature. Therefore, this section examines the effects of various

ed
3 parameters on the static vertical settlement of CPRF. The effects of pile spacing, cohesion,
4 friction angle, and Poisson's ratio of soil are examined. All the results shown here assume
5 spacing between piles as 6D, except section 6.1. The unit weight and Young's modulus of soil,
6 raft, and pile are assumed to be the same as those presented in Table 1.

iew
7 6.1 Effect of pile spacing

8 Two pile spacings are used in the present study, i.e., 4D and 6D, which corresponds to the
9 medium and large pile spacing. Fig. 3 shows the effect of pile spacing on the load settlement

v
10 curve. It can be observed that with the increase in spacing of piles, settlement at the top of raft
11 is reduced by a significant amount due to the increase in group efficiency. In the 6D spacing

re
12 model, piles are behaving individually and uniformly distribute the load of superstructure over
13 the soil. However, large pile spacing may require a large size of raft, which may lead to an
14 uneconomical design. Nevertheless, in the present study size of the raft is kept fixed for both
15 the pile spacings. er Load intensity (kPa)

0 100 200 300 400 500


pe
0

0.1
Settlement (m)

0.2
ot

0.3

0.4 Present, S = 4D
tn

Present, S = 6D
0.5
16
17 Figure 3: Effect of pile spacing
rin

18 6.2 Effect of soil cohesion (c)

19 To check the effect of cohesion, analysis is performed on the piled raft foundation by varying
20 the soil cohesion as 20, 40, 60 and 100kPa. The plasticity of soil was considered using cap
ep

21 model. Fig. 4 shows the influence of variation in cohesion of soil on the average load settlement
22 curve at the top of raft. It can be observed that with the increase of cohesion, the load carrying
23 capacity of the piled raft foundation is increased that was as expected. The settlement of CPRF
24 is significantly decreased by increase in cohesion of soil from 20 kPa to 40 kPa at the high load
Pr

25 intensity but further increase in cohesion of soil is not much affecting the settlement of CPRF
26 due to enough densification of soil.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
Load (kPa)

0 100 200 300 400 500

ed
0

Settlement (m)
0.1

iew
0.2
c = 20 kPa

0.3 c = 35 kPa
c = 45 kPa
c = 55 kPa

v
0.4
1
2 Figure 4: Effect of cohesion,  = 30o, ν = 0.35

re
3 6.3 Effect of angle of friction of soil ()

4
5
er
Analyses are performed of the piled raft foundation having the angle of shearing resistance of
20, 30, 36 and 40. Fig. 5 shows the average load settlement curve at the top of raft with different
6 friction angles of soil. It can be prominent that settlement reduced promptly with the increase
7 in friction angle from 20 to 30. Further increase in the friction angle is not affecting the
pe
8 settlement of piled raft foundation. This is due to the enough rise in relative density (RD) of
9 soil. When the soil is densifying from the initial stage, then it decreases the settlement by the
10 significant amount, but further densification of soil is difficult to remove the remaining voids
11 from the soil.
ot

Load intensity (kPa)

0 250 500 750 1000


0
tn

0.1
Settlement (m)

0.2
rin

ɸ = 20
ɸ = 30
0.3 ɸ = 36
ɸ = 40
ep

12 0.4
13 Figure 5: Effect of friction angle, c = 40 kPa, ν = 0.35
Pr

14 6.4 Effect of Poisson's Ratio ()

15 Analyses are performed of the piled raft foundation having the Poisson's ratio of 0.35, 0.40,
16 and 0.45. Fig. 6 shows the effect of Poisson's ratio on the average load settlement curve at the

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 top of raft. It can be prominent that settlement is uniformly reducing with the rise in the
2 Poisson's ratio. Moreover, with the increase in Poisson's ratio, the load settlement curve

ed
3 becomes linear.
Load intensity (kPa)
0 200 400 600 800 1000
0

iew
0.1
Settlement (m)

0.2

v
  
  

re
  
0.3
4
5 Figure 6: Effect of Poisson’ ratio, ϕ = 30o, c = 40 kPa

6 7. Seismic Settlement of CPRF without NPP


er
7 For the dynamic analysis of CPRF, 4D and 6D spacing models were used to check the
pe
8 settlement under the real earthquake excitation. In the dynamic analysis, boundary and
9 damping play a vital role; Kelvin element boundary and Rayleigh damping are used to examine
10 the accurate behavior of piled raft foundation. The value of damping ratio is assumed as 10%
11 for soil and 5% for the raft and piles. The Rayleigh damping parameters are calculated using
12 the model analysis of piled raft foundation. The earthquake motion was applied at the base of
ot

13 the model as shown in Fig. 1.


tn

14 7.1 Input Time Histories

15 Fig. 7 shows the input earthquake motion for 4 earthquakes for the seismic analysis of piled
16 raft foundation. In Chamoli earthquake time history, point a1 is the first s-wave arrival at the
17 time period of 4.36 sec and having acceleration value of 0.102g while point a2 is corresponds
rin

18 to surface wave having PGA value of 0.374g at the 4.6 sec with low frequency. In Kobe
19 earthquake time history, points b1, b2 and b3 correspond to s-wave having the acceleration value
20 of 0.254g, 0.258g and 0.344g respectively. In Uttarkashi earthquake time history, two peaks
21 are dominant i.e., c1 and c2 having the PGA value of 0.286 and 0.295. In Northridge earthquake,
ep

22 points d1, d2 and d3 correspond to s-wave having the acceleration value of 0.409g, 0.568g and
23 0.257g respectively. Table 2 provides a summary of key parameters of the input earthquake
24 time histories.
25
Pr

26
27
28

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 Table 2: Key parameters of input earthquake motion
Magnitude Predominant

ed
Sl. No. Earthquake (year) PGA (g)
(Mw) Frequency (Hz)
1. Chamoli (1999) 6.8 0.374 1.66
2. Kobe (1995) 6.9 0.344 0.58
3. Uttarkashi (1991) 6.8 0.295 2.90

iew
4. Northridge (1994) 6.7 0.568 1.22
2

v
re
er
pe
ot
tn

3
4 Figure 7: Input time histories (a) Chamoli (1999), (b) Kobe (1995), (c) Uttarkashi (1991) and
rin

5 (d) Northridge (1994)

6 7.2 Settlement Time Histories of CPRF without NPP


ep

7 Fig. 8 shows the settlement time histories of the combined piled raft foundation under different
8 piles spacing. The maximum settlement observed at the top of raft under Chamoli (1999), Kobe
9 (1995), Uttarkashi (1991) and Northridge (1994) earthquake motion were 30.81 mm, 31.94
10 mm, 14.46 mm and 47.72 mm, respectively for 6D spacing while the maximum settlement was
Pr

11 36.01 mm, 35.93 mm, 18.46 mm and 50.57mm, respectively for 4D spacing which is listed in
12 Table 3. Thus, like static case, the settlement decreases significantly when the spacing between
13 piles is increased from 4D to 6D. It can be perceived that with the increase in spacing of piles,

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 settlement is reduced by 14-15% for different earthquake motions. This is attributed to the fact
2 that the contribution of piles in load sharing increases when the piles spacing is increased.

ed
3 Further, among the 4 earthquakes for both the spacings, the maximum settlement is for
4 Northridge (for which PGA was maximum) while the minimum settlement is for Uttarkashi
5 (for which PGA was minimum). Hence, the PGA is dominating the settlement. However, the
6 predominant frequencies of these earthquakes also play crucial role. For example, it can be

iew
7 observed that the with the increase in PGA of Kobe earthquake just by 17% as compared to
8 Uttarkashi earthquake, the seismic settlement increased significantly by 120%. This may be
9 because the predominant frequency (0.58Hz) of Kobe earthquake motion is near to the
10 fundamental frequency of system (0.753Hz). The fundamental frequency of the system is
11 obtained by carrying out the free vibration analysis of the system.

v
re
er
pe
ot
tn
rin
ep

12
13 Figure 8: Settlement time histories (a) Chamoli (1999) and (b) Kobe (1995) (c) Uttarkashi
14 (1991) and (d) Northridge (1994)
15
Pr

16
17
18

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 Table 3: Maximum Seismic Settlement of CPRF
Predominant Settlement (mm)

ed
Sl. No. Earthquake (year) PGA (g)
Frequency (Hz) 4D 6D
1. Chamoli (1999) 0.374 1.66 36.01 30.81
2. Kobe (1995) 0.344 0.58 35.93 31.94
3. Uttarkashi (1991) 0.295 2.90 18.46 14.46

iew
4. Northridge (1994) 0.568 1.22 50.57 47.72
2 Moreover, in Chamoli earthquake, e1 and e2 are the point when the s-wave and surface wave
3 occur respectively; and then settlement become constant. At point e2, surface wave
4 momentarily causes more settlement in both 4D and 6D spacing models but after that
5 settlement of CPRF is reduced due to low intensity of loading. In Kobe earthquake settlement

v
6 time history, f1, f2 and f3 are the point corresponding to s-wave of higher frequency. f1 is the
7 first arrival of s-wave and causes less settlement. Further s-wave at point f2 and f3 of higher

re
8 acceleration value causes more settlement. In case of Uttarkashi earthquake, settlement
9 occurred at point g1 and then more settlement at point g2 due to the second arrival of s-wave.
10 In Northridge earthquake settlement time history, h1, and h2 are the point corresponding to s-
11 wave of higher frequency. h1 is the first arrival of s-wave and causes less settlement. Further,
er
12 s-wave at point h2 and h3 of higher acceleration value causes more settlement like Kobe
13 earthquake time history. From the comparison of Fig. 8(a) and Fig. 8(b) it is also observed that
14 Kobe earthquake having lower PGA (0.344g) causes higher settlement as compares to Chamoli
pe
15 earthquake of higher PGA (0.374g) this may be due to proximity of predominant frequency of
16 Kobe earthquake near the fundamental frequency of the system. Furthermore, it observed that
17 for all 4 earthquakes, permanent deformation occurs due to the perfectly elasto-plastic
18 behaviour of soil.
ot

19 8. A Practical Application: Model Properties of NPP and CPRF

20 For the case study, a nuclear structure on the non-liquefiable and liquefiable soils are
tn

21 considered. The generalized form of the nuclear structure is shown in Fig. 9, which is similar
22 to that used by Firoj and Maheshwari (2022), however with smaller soil domain (where depth
23 is reduced to 30 m and horizontal extent to 200 m). The nuclear structure considered in the
24 present study is assumed to be a lumped model (for the sake of simplicity and computational
rin

25 cost) of a typical Indian nuclear reactor building as reported in Saxena and Paul (2012). These
26 authors considered raft foundation for NPP structure, however, CPRF is considered in the
27 present study.
28 Three quarters of finite difference model (to show the inner detail) of the nuclear structure, raft,
ep

29 pile and layered soil is shown in Fig. 10. Where the lumped model of structure-raft-piles is also
30 shown. The lumped mass system for the nuclear structure is widely used in dynamic analysis
31 (Lee et al. 2015; Bolisetti et al. 2018; Nguyen et al. 2020). The raft foundation is square having
32 45 m each side and 4 m thickness with 2 m embedment in the soil. In addition to the raft, a pile
Pr

33 group of circular patterns is considered in the present study to represent the CPRF. The
34 behaviour of the structural component is assumed elastic and the material properties are listed
35 in Table 4. The soil is assumed heterogeneous, having 3 m depth of each layer. Total depth and

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 width of the soil domain considered are 30 m and 200 m, respectively. The material properties
2 used for the soil are listed in Table 5 which are calibrated as per the procedure given in Wang

ed
3 et al. (2014). The analysis is performed for two soil models. The first model is non-liquefiable
4 soil in which the entire soil properties correspond to the Drucker Prager cap model. However,
5 in case of the liquefiable soil, CycLiq model is used in the upper liquefiable soil layers and
6 Drucker Prager model in the non-liquefiable soil layers.

v iew
re
er
pe
7
8 Figure 9: Generalized form of NPP structure
9
ot
tn
rin
ep
Pr

10

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 Figure 10: 3D equivalent finite difference model of NPP structure on CPRF (Three quarters
2 model is shown for soil the soil-foundation system to show inner details). It also shows

ed
3 Lumped model of structures-raft-piles
4 Table 4: Material Properties of Foundation and Nuclear Structure
Properties Structure Raft and Pile
Grade (as per Indian Standard) M35 M36

iew
Modulus of Elasticity (GPa) 28 30
Poisson's Ratio 0.2 0.2
Unit Weight (kN/m3) 25 24
Damping 5% 5%
5

v
6 Table 5: Material model parameters for the soil at the NPP site

re
Depth Density SPT G
RD κ h M lc eo ξ dre,1 dre,2 dir a dr np nd
(m) (kg/m3) N (MPa)
3 1820 12 80 33 0.007 0.3 1.14 0.0213 0.837 0.7 0.4 1200 0.3 30 0.05 1.1 7.8
6 1830 14 89 36 0.007 0.3 1.14 0.0213 0.837 0.7 0.4 1200 0.3 30 0.05 1.1 7.8
9 1820 20 113 43 0.006 0.6 1.35 0.0190 0.901 0.7 0.4 1200 0.25 30 0.05 1.1 8.0
12 1830 31 153 53 0.006 0.6
er 1.35 0.0190 0.901 0.7 0.4 1200 0.25 35 0.05 1.1 8.0
15 1850 25 134 48 0.006 0.6 1.35 0.0190 0.901 0.7 0.4 1200 0.25 30 0.05 1.1 8.0
18 1860 29 149 52 0.006 0.6 1.35 0.0190 0.901 0.7 0.4 1200 0.25 35 0.05 1.1 8.0
21 1890 23 129 46 0.006 0.6 1.35 0.0190 0.901 0.7 0.4 1200 0.25 30 0.05 1.1 8.0
24 1800 57 228 73 0.005 1.2 1.42 0.0120 0.923 0.7 0.4 1200 0.25 170 0.05 1.1 8.2
27 1820 57 230 73 0.005 1.2 1.42 0.0120 0.923 0.7 0.4 1200 0.25 170 0.05 1.1 8.2
pe
30 1790 72 266 81 0.005 1.2 1.42 0.0120 0.923 0.7 0.4 1200 0.25 170 0.05 1.1 8.2
7 Note: The parameters listed are for both elastic and nonlinear soil to deal with liquefaction. Notations
8 and model properties are the same as those reported by Wang et al. (2014)

9 9. Static Settlement Analysis of a NPP Structure


ot

10 In this section, static settlement analysis of a NPP rested on the raft, piles group and CPRF
11 (Figs. 9 and 10) are carried out for both the non-liquefiable and liquefiable soils. Here, for
tn

12 building the raft foundation model of NPP, piles are removed from the model. While for the
13 pile group model, raft is replaced by a cap with 1m thickness and there is no interaction between
14 the cap and soil.
15 The vertical load of the NPP structure is applied gradually in order to obtain the load settlement
rin

16 curve of the NPP structure on the various type of foundations. Fig. 11 shows the load settlement
17 curves for three types of foundation in the non-liquefiable and liquefiable soils. For non-
18 liquefiable soil, the settlement observed in raft, piles group and CPRF are 110 mm, 77 mm and
19 20 mm, respectively at load intensity of 75 kPa. It can be observed that initially upto a load
ep

20 intensity of 52 kPa, the raft has smaller settlement as compared to the pile group in both soil
21 types. With further increase in the load intensity, the raft has higher settlement compared to
22 pile groups. However, CPRF has remarkably low settlement among three at all load intensities.
23 Thus, indicating utility of CPRF in reducing the settlement. When liquefiable soil is considered,
Pr

24 it is observed that the settlement of all the foundations is increased by approximately by 10-
25 20% compared to that in the non-liquefiable soil. This may be attributed to the loss of stiffness
26 of soil because of increase in pore pressure due to liquefaction.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1
Load Intensity (kPa)

ed
0 30 60 90 120
0

20 CPR

Settlement (mm)

iew
40

60 Piles

80 Raf

100 Liquefiable

v
Non-Liquefiable
120
2

re
3 Figure 11: Load settlement curves of NPP structure in the non-liquefiable and liquefiable
4 soils

(a) Raft Dimension from Center (m)


-30 -20 -10
er 0 10 20 30
0
Raft
Piles Group
Settlement (mm)

5
CPRF
pe
10

15
ot

20

25
5
tn

(b) Raft Dimension from Center (m)


-30 -20 -10 0 10 20 30
0
Raft
rin

Piles Group
Settlement (mm)

5
CPRF

10
ep

15

20
Pr

25
6
7 Figure 12: Settlement along the length of raft in (a) non-liquefiable soil (b) liquefiable soil

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 Fig. 12(a) and (b) shows the static settlement at the raft level of the NPP structure in the non-
2 liquefiable and liquefiable soils, respectively. From Fig. 12(a), it is observed that the CPRF is

ed
3 in better position in reducing the static settlement of the nuclear building at all over the length
4 of raft. CPRF approximately reduced the static settlement of the NPP by an amount of 25%
5 and 15% as compared to the raft and piles group foundation, respectively in the non-liquefiable
6 soil. The corner of the raft/pile cap is less settled as compared to the central part of the raft/pile

iew
7 cap. This may be attributed to the difference in load carried by central and corner piles.
8 Nonetheless, there is not significant differential settlement observed for the static load for all
9 three types of foundation. Similar observation has been observed for the static settlement of
10 the NPP structure in the liquefiable soil with slightly increased value (Fig. 12b). Furthermore,
11 the behaviour of settling foundation is similar in both types of soil. It can be concluded that

v
12 CPRF foundation provide enough densification of soil which helps in reducing the static
13 settlement of NPP structure. CPRF serves in two ways (i) raft of CPRF uniformly distribute

re
14 the pressure over the soil and (ii) piles of CPRF reduced the settlement (Bhartiya et al. 2021).
15 Therefore, CPRF can be an advantageous foundation for the NPP structure even if it is
16 constructed on liquefiable soil.
17 NPPs are massive structures and consolidated settlement is an important parameter for the
er
18 design of NPP structure on the liquefiable soil. Therefore, consolidated settlement analysis of
19 NPP structure rested on the raft, piles group and CPRF is carried out. When the construction
20 of the foundation is done due to application of additional load, the pore water pressure in the
pe
21 water-logged soil increases as the hydraulic conductivity of the soil is very small. It takes some
22 time to dissipate this excess pore pressure and to increase the stress to be transmitted to the soil
23 layers. This steady state increase in the effective stress in the soil layer will cause settlement
24 over a period and is devoted to as consolidated settlement. In this analyses, steady state load of
25 NPP structure is applied over the various type of foundation for a one-year period and
ot

26 settlement is evaluated at one-month interval. The coefficient of consolidation Cv used in this


27 analysis is 1.6×104 cm2/sec. Fig. 13 shows the consolidated settlement of the NPP structure for
28 a one-year period in the liquefiable soil. It can be observed that at the end of the period, the
tn

29 CPRF reduced the consolidated settlement of the NPP structure by a margin of 30% and 20%
30 in contrast to the raft and piles group foundation, respectively. The consolidated settlement
31 with the CPRF is comparatively smaller due to the combining action of raft and piles.
32 Therefore, it can be concluded that the use of CPRF reduces the consolidated settlement
rin

33 remarkably.
ep
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
ed
v iew
1
2 Figure 13: Consolidated settlement of NPP in the liquefiable soil

re
3 10. Seismic Settlement Analysis of a NPP Structure

4 In this section, seismic settlement analyses of a NPP structure rested on the raft, piles group
er
5 and CPRF (Figs. 9 and 10) are carried out for both the non-liquefiable and liquefiable soils. In
6 order to absorb seismic wave reflection from the boundary, free field boundary condition is
7 used at the outer nodes of the model while the base of the model is assumed fixed. The seismic
pe
8 input motion (Fig. 7a) was applied at the base of the soil and seismic settlement and lateral
9 displacement are investigated. The effects of presence of different foundations (raft, pile and
10 CPRF) and structure on the free field response are also examined (Fig. 10).

11 10.1 Influence of Interface Nonlinearity


ot

12 To examine the effect of geometrical nonlinearity with the liquefiable soil, the rigid interface
13 behaviour of the raft-soil and pile-soil (i.e., without geometrical nonlinearity) is compared with
tn

14 slip and separation behaviour of raft-soil and pile-soil (i.e., with geometrical nonlinearity). For
15 this determination, spectrum ratio is calculated to obtain the seismic response of the NPP and
16 raft which is the ratio of output spectral acceleration divided by input spectral acceleration. Fig.
17 14(a) and Fig. 14(b) show the spectrum ratio at the top of NPP and at raft foundation level,
rin

18 respectively. The peak spectrum ratio of the NPP structure and raft due to the interface
19 nonlinearity is increased by a margin of 25.47% and 26.43%, respectively. Thus, the peak
20 spectrum ratios are significantly increased due to the raft-soil and pile-soil interface
21 nonlinearity in the liquefiable soil. However, Firoj and Maheshwari (2022a) observed that due
ep

22 to the geometrical nonlinearity seismic response is increased by 7-10% in non-liquefiable soil.


23 Therefore, it can be inferred that the interface nonlinearity has greater significance in the
24 liquefiable soil as compared to the non-liquefiable soil. This may be attributed to the loss of
25 stiffness due to slip and separation at the interface of raft-soil and pile-soil. Therefore, interface
Pr

26 nonlinearity is used in the further seismic settlement analysis of NPP structure.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
ed
iew
1

v
2 Figure 14: Influence of interface nonlinearity on the response of (a) NPP Structure (b) Raft
3 both in liquefiable soil

re
4 10.2 Influence of Type of Foundation in Non-Liquefiable and Liquefiable Soils

5 In this section, the effects of presence of a NPP structure (rested on raft and CPRF) are
er
6 examined over the free field response. In Fig. 15(a) and (b), the NPP structure is considered to
7 be founded on the CPRF and raft foundation, where the excess pore pressure at the depths of
8 2m and 4m, respectively are shown in liquefiable soil. It can be observed that due to the
pe
9 presence of structure and foundation, the excess pore pressure is reduced significantly at both
10 the depths. However, this reduction is higher when the CPRF is considered. It is also noticeable
11 that at 2m depth, the excess pore pressure is fluctuating due to disturbance caused by the
12 presence of piles and raft. Due to the presence of NPP-CPRF system, the excess pore pressure
13 ratio (ru) is never reaching 1 at both the depths which was the case with the free field. However,
ot

14 in case of NPP-Raft system the pore pressure is comparatively higher at both depths which
15 indicates that raft foundation of NPP may not be enough to reduce pore pressure. The presence
16 of CPRF reduced the chances of liquefaction occurrence; however, liquefaction is observed in
tn

17 the free field shaking of the NPP site. Therefore, it can be concluded that the CPRF may help
18 to reduce the excess pore pressure beneath the NPP structure founded in the liquefiable soil.

19 Fig. 16(a) shows the settlement time history of NPP-CPRF system in comparison with the free
rin

20 field and NPP-raft system in liquefiable soil. Liquefaction induced settlement of the NPP-
21 CPRF, and NPP-Raft system is increased significantly as compared to the free field. This may
22 be attributed to the excessive weight of the NPP-structure founded on the CPRF. Therefore, a
23 check is required for the settlement demand of the NPP structure. Furthermore, it can be
ep

24 observed that NPP-CPRF system reduced the settlement of the NPP structure by significant
25 amount as compared to the NPP-Raft system which attributes that combined action of piles and
26 raft uniformly distributed that load and reduced the settlement of NPP structure in the
27 liquefiable soil. Fig. 16(b) shows the horizontal displacement response of the ground surface
Pr

28 with NPP-Raft, NPP-CPRF and free field in liquefiable soil. It can be depicted that lateral
29 response of the soil in presence of CPRF is reduced by 12% and 19.17% as compared to the
30 NPP-Raft and to the free field because piles provide the lateral support to the soil and increase

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 the total stiffness of the NPP-CPRF system. Therefore, due to the NPP-CPRF system, the
2 increase in settlement (Fig. 16a) is acceptable as it is expected moreover, it helps in reducing

ed
3 the horizontal displacement significantly (Fig. 16b).

v iew
re
er
4
pe
5 Figure 15: Effect of foundation type of NPP structure on pore pressure time history below
6 the raft in liquefiable soil at (a) 2m depth (b) 4m depth
ot
tn
rin
ep

7
8 Figure 16: Influence of foundation type of NPP structure in liquefiable soil on (a) settlement
9 time history (b) horizontal displacement at the ground surface level
Pr

10 Fig. 17 shows the seismic settlement of NPP structure rested on the raft, piles group and CPRF
11 considering nonlinearity of soil in non-liquefiable soil. The maximum settlement observed in

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 raft, piles group and CPRF is 205mm, 158mm and 150mm, respectively. CPRF reduced the
2 peak settlement of structure by 5% and 27.2% as compared to piles group and raft foundation,

ed
3 respectively. It can be noted that the reduction in peak settlement of NPP structure as compared
4 to raft is remarkable however, as compared to piles group, it is marginal.
Time (s)
0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25

iew
0
Settlement (mm)

10

15
Raft

v
20 Piles Group
CPRF

re
25
5
6 Figure 17: Seismic settlement of NPP structure founded on different foundation in non-
7 liquefiable soil
Raft Dimension from Center (m)
(a)
0
-30 -20
er
-10 0 10 20 30

Raft
5 Piles Group
pe
Max. Settlement (mm)

CPRF
10

15

20
ot

25

30
tn

35
8
Raft Dimension from Center (m)
(b)
-30 -20 -10 0 10 20 30
0
Raft
rin

5 Piles Group
Max. Settlement (mm)

10 CPRF

15

20
ep

25

30

35
Pr

40
9
10 Figure 18: Influence of foundation type on seismic settlement of NPP in (a) non-liquefiable
11 (b) liquefiable soil

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 Fig. 18(a) shows the seismic settlement of the NPP structure on three different foundations
2 along the diameter of the raft. The center of the raft is marked with zero and the maximum

ed
3 settlement of the raft is evaluated due to the Chamoli earthquake in the non-liquefiable soil. It
4 can be observed that the raft foundation undergoes maximum settlement compared to the piles
5 group and CPRF at all the locations along the diameter of the raft. The reduction in maximum
6 settlement of the CPRF is 63.98% and 19.44% as compared to the raft and pile group

iew
7 foundations, respectively. It is also observed that there is large differential settlement in the
8 pile group foundation compared to the raft and CPRF foundation. The differential settlement
9 observed in raft, pile group and CPRF is 6.59mm, 13.89 mm and 6.48mm, respectively.
10 Therefore, it can be inferred that the contribution of raft compared to piles in CPRF foundation
11 for uniformly distributing the load over the soil is greater. Similar observations can be seen for

v
12 the seismic settlement of raft, pile group and CPRF in the liquefiable soil (Fig. 18b). However,
13 the maximum and differential settlement for all type of foundations in the liquefiable soil is

re
14 relatively greater. It can be noticed that behaviour of the seismic settlement is quite different
15 from the static settlement. In case of static settlement (Fig. 12), central displacement is higher
16 as compared to that at the ends. However, in case of seismic settlement (Fig. 18), one side of
17 the raft is more tilted as compared to the other side (i.e. more differential settlement). This may
er
18 be due the nature of the applied loading.
19 Fig. 19 shows the seismic settlement of CPRF in the non-liquefiable and liquefiable soils under
20 the Chamoli earthquake excitation. It can be observed that the settlement of liquefiable soil is
pe
21 significantly higher in the left side of the raft; however, in the left side it is more or less same
22 in both the soils. The maximum differential settlement of the CPRF in the liquefiable and non-
23 liquefiable soil is observed as 12.84 mm and 9.14 mm respectively. Higher differential
24 settlement in the liquefiable soil is attributed to the reduction in the stiffness of the soil.
ot

Raft Dimension from Center (m)


-30 -20 -10 0 10 20 30
0
Nonliquefiable Soil
tn

5 Liquefiable Soil
Max. Settlement (mm)

10
rin

15

20

25
ep

30
25
26 Figure 19: Effect of liquefaction model on seismic settlement of CPRF of NPP
27
Pr

28

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 11. Conclusions

ed
2 A 3D model of a NPP structure founded on raft, piles and CPRF was developed. The effects of
3 various parameters on the static and seismic settlement of CPRF (alone) as well as NPP
4 structure (with different foundation) were examined in both non-liquefiable and liquefiable
5 soils. The Drucker Prager with cap plasticity and CycLiq models were used to define the

iew
6 constitutive relationships of non-liquefiable and liquefiable soils, respectively. Effects of three
7 different types of foundations i.e. raft, piles and CPRF on the seismic settlement are
8 investigated. Also, seismic differential settlement of the CPRF of NPP structure is investigated
9 considering the geometrical and material nonlinearity. The following conclusions can be drawn
10 from this study:

v
11 1. Static settlement of CPRF decreases with the increase in spacing of piles. The average
12 settlement is decreased with the increase in friction angle, cohesion and the Poisson's

re
13 ratio of soil.
14 2. In case of seismic loading too, with the increase in spacing of piles, settlement of CPRF
15 is reduced by 14-15% due to the larger contribution of piles in load sharing.
16 3. In case of static settlement of NPP structure, the corner of the raft/ pile cap is less
er
17 settled as compared to the central part of the raft/pile cap. The consolidated settlement
18 of the NPP on CPRF is comparatively smaller compared to that on piles and raft
19 foundations.
20 4. Geometrical nonlinearity significantly increased the seismic response of the NPP
pe
21 structure in the liquefiable soil.
22 5. In the presence of piles, the excess pore pressure is reduced significantly at NPP site
23 compared to the free field. The lateral response of the soil in presence of CPRF is
24 reduced by a margin of 20% as compared to the free field due to the lateral resistance
25 provided by piles.
ot

26 6. CPRF reduced the maximum settlement of structure by 5% and 25% as compared to


27 piles group and raft foundation, respectively. The differential settlement of the CPRF
28 and raft foundation is low as compared to the pile group foundation.
tn

29 7. For the soil properties considered, the differential settlement of the CPRF in the
30 liquefiable soil is increased significantly as compared to the non-liquefiable soil due to
31 the reduction in stiffness.
32 Thus, it can be concluded that the CPRF provides a more efficient supporting system where a
rin

33 raft and piles group foundation may not meet the required design strength criteria mainly
34 governed by the seismic differential settlement. The results presented in this paper can be used
35 for the seismic design of NPP structure on CPRF.
ep

36 Data Availability Statement

37 Some or all data, models, or code that support the findings of this study are available from the
Pr

38 corresponding author upon reasonable request.

39 Declaration of competing interest

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 The authors declare that they have no known competing financial interests or personal
2 relationships that could have appeared to influence the work reported in this paper.

ed
3 Acknowledgment

4 The research presented in this paper is supported by a fellowship to the second author by the

iew
5 Ministry of Education, Government of India. This support is gratefully acknowledged.

6 References

7 1. Ata, A., Badrawi, E., Nabil, M. 2015. Numerical analysis of unconnected piled raft with

v
8 cushion. Ain Shams Engineering Journal, 6(2): 421-428.

re
9 2. Azizkandi, A. S., Baziar, M. H., Yeznabad, A. F. 2018. 3D dynamic finite element
10 analyses and 1 g shaking table tests on seismic performance of connected and
11 nonconnected piled raft foundations. KSCE Journal of Civil Engineering, 22(5): 1750-
12 1762.
13 3. Bhaduri, A., Choudhury, D. 2021. Steady-state response of flexible combined pile-raft
14
er
foundation under dynamic loading. Soil Dynamics and Earthquake Engineering, 145:
15 106664.
16 4. Bhartiya, P., Basu, D., Chakraborty, T. 2021. Nonlinear settlement of piled rafts in
pe
17 sandy soil. International Journal of Geomechanics, 21(11): 04021214.
18 5. Bolisetti, C., Whittaker, A. S. and Coleman, J. L. 2018. "Linear and nonlinear soil-
19 structure interaction analysis of buildings and safety-related nuclear structures." Soil
20 Dynamics and Earthquake Engineering, 107: 218-233.
21 6. Chopra, A. K. 2007. Dynamics of structures. Pearson Education India.
ot

22 7. Clancy, P., Randolph, M.F. 1993. An approximate analysis procedure for piled raft
23 foundations. International Journal for Numerical and Analytical Methods in
24 Geomechanics, 17(12): 849-869.
tn

25 8. Firoj, M., Maheshwari, B. K. 2022a. Effect of CPRF on nonlinear seismic response of


26 an NPP structure considering raft-pile-soil-structure-interaction. Soil Dynamics and
27 Earthquake Engineering, 158: 107295.
28 9. Firoj, M., Maheshwari, B. K. 2022b. A new nonlinear spring‐dashpot model of CPRF
rin

29 of NPP structure based on coupled BEM‐FEM approach. Earthquake Engineering &


30 Structural Dynamics. https://doi.org/10.1002/eqe.3794
31 10. Gu, L., Ye, G., Wang, Z., Ling, X., Zhang, F. 2017. Settlement mechanism of piled-
32 raft foundation due to cyclic train loads and its countermeasure. Earthquake
ep

33 Engineering and Engineering Vibration, 16, 499-511.


34 11. Hibbit, H. D., Karlsson, B. L., Sorrensen, P. 2007. Abaqus theory manual, SIMULIA,
35 Providence, RI.
36 12. Jia, K., Xu, C., Dou, P., Zhang, X., Song, J. 2023. Seismic Behavior and Failure
Pr

37 Mechanism of Pile-Group in Mildly Sloping Liquefiable Soils with Crusts: Large-Scale


38 Shaking Table Experiment. Journal of Earthquake Engineering, 1-28.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 13. Katzenbach, R., Bachmann, G., Boled-Mekasha, G., Ramm, H. 2005. Combined Pile
2 Raft Foundation (CPRF): an appropriate solution for foundation of high-rise building.

ed
3 Sloak Journal of Civil Engineering (SJCE), 19-29.
4 14. Kitiyodom, P., Matsumoto, T. 2002. A simplified analysis method for piled raft and
5 pile group foundations with batter piles. International Journal for Numerical and
6 Analytical Methods in Geomechanics, 26(13): 1349-1369.

iew
7 15. Kontoe, S., Zdravkovic, L., Potts, D. M., Salandy, N. E. 2007. The use of absorbing
8 boundaries in dynamic analyses of soil-structure interaction problems. 4th Int. Conf. in
9 Earthquake Geotechnical Engineering, Aristotle University of Thessaloniki,
10 Thessaloniki, Greece.
11 16. Lee, E.H., Jung, D.R., Rhee, I., Kim, J.M., 2021. A nonlinear soil-structure interaction

v
12 analysis technique based on seismic isolation design response spectrum for seismically
13 isolated nuclear structures with rigid basemat. Nuclear Engineering and Design, 381,

re
14 p.111334.
15 17. Lee, H., Ou, Y. C., Roh, H., Lee, J. S. 2015. Simplified model and seismic response of
16 integrated nuclear containment system based on frequency adaptive lumped-mass stick
17 modeling approach. KSCE Journal of Civil Engineering, 19(6): 1757-1766.
er
18 18. Li, Z., Yang, Y., Gong, W., Yi, P. C., Dai, G. Liang, F. 2022. Investigation on the
19 seismic response of nuclear power stations with a pile-raft foundation using centrifuge
20 tests. Earthquake Engineering and Engineering Vibration, 21(3): 753-768.
21 19. Luo, R., Yang, M., Li, W. 2018. Normalized settlement of piled raft in homogeneous
pe
22 clay. Computers and Geotechnics, 103, 165-178.
23 20. Maheshwari, B. K., Firoj, M. 2021. A State of Art: Seismic Soil–Structure Interaction
24 for Nuclear Power Plants. Latest Developments in Geotechnical Earthquake
25 Engineering and Soil Dynamics, 393-409.
26 21. Maheshwari, B. K., Firoj, M. 2022. Seismic Response of Combined Piled Raft
ot

27 Foundation using Advanced Liquefaction Model. In review for Geotechnique.


28 22. Maheshwari, B. K., Firoj, M. 2020. Equivalent linear spring-dashpot model for
29 embedded foundations of NPP. In 17th World Conference on Earthquake Engineering,
tn

30 17WCEE Sendai, Japan. 4c-0014.


31 23. Nguyen, D. D. C., Jo, S. B., Kim, D. S. 2013. Design method of piled-raft foundations
32 under vertical load considering interaction effects. Computers and Geotechnics, 47:
33 pp.16-27.
rin

34 24. Nguyen, D. V., Kim, D., Nguyen, D. D. 2020. Nonlinear seismic soil-structure
35 interaction analysis of nuclear reactor building considering the effect of earthquake
36 frequency content. Structures, 26: 901-914.
37 25. Novak, M., Mitwally, H. 1988. Transmitting boundary for axisymmetrical dilation
ep

38 problems. Journal of engineering mechanics, 114(1): pp.181-187.


39 26. Osman, A. S. and Randolph, M. F. 2012. An analytical solution for the consolidation
40 around a laterally loaded pile. International journal of geomechanics., 12(3): pp.199-
41 208.
Pr

42 27. Patil, G., Choudhury, D., Mondal, A. 2021. Three-Dimensional soil–foundation–


43 superstructure interaction analysis of nuclear building supported by combined piled–
44 raft system. International Journal of Geomechanics, 21(4): 04021029.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752
1 28. Poulos, H. G. 2005. Piled raft and compensated piled raft foundations for soft soil sites.
2 Advances in Designing and Testing in Deep Foundations Engineering. Geotechnical

ed
3 special publication 129, ASCE, Reston, VA, 214–234.
4 29. Poulos, H. G. 2012. Foundation design for tall building. Geotechnical Engineering,
5 State of the Art and Practice. Geotechnical special publication 226, ASCE, Reston, VA,
6 786–809.

iew
7 30. Pradel, D. 1998. Procedure to evaluate earthquake-induced settlements in dry sandy
8 soils. Journal of Geotechnical and Geoenvironmental Engineering, 124(4), 364-368.
9 31. Raut, J. M., Khadeshwar, S. R., Bajad, S. P., Kadu, M. S. 2014. Simplified design
10 method for piled raft foundation. Advances in Soil Dynamics and Foundation
11 Engineering, Geotechnical special publication 240, ASCE, Reston, VA, 462–471.

v
12 32. Salciarini, D., Ronchi, F., Cattoni, E., Tamagnini, C. 2013. Thermomechanical effects
13 induced by energy piles operation in a small piled raft. International journal of

re
14 Geomechanics, 15(2): 04014042.
15 33. Saxena, N., Paul, D. K. 2012. Effects of embedment including slip and separation on
16 seismic SSI response of a nuclear reactor building. Nuclear engineering and design,
17 247: 23-33. er
18 34. Sinha, A., Hanna, A. M. 2016. 3D numerical model for piled raft foundation.
19 International Journal of Geomechanics, 17(2): pp.04016055-9.
20 35. Tang, L., Liu, S., Ling, X., Wan, Y., Li, X., Cong, S., Su, L. 2022. Seismic soil
21 liquefaction mitigation using stone columns for pile-supported wharves. Journal of
pe
22 Earthquake Engineering, 26(16), 8229-8256.
23 36. Tehrani, F. S., Salgado, R., Prezzi, M. 2016. Analysis of axial loading of pile groups in
24 multilayered elastic soil. International Journal of Geomechanics, 16(2): 04015063.
25 37. Tokimatsu, K., Seed, H. B. 1987. Evaluation of settlements in sands due to earthquake
26 shaking. Journal of geotechnical engineering, 113(8), 861-878.
ot

27 38. Wang, R., Zhang, J. M., Wang, G. 2014. A unified plasticity model for large post-
28 liquefaction shear deformation of sand. Computers and Geotechnics, 59: 54-66.
29 39. Xu, Q., Zhang, J., Tang, Z. 2014. Field Grouting Experiment on Reinforcing the Pile-
tn

30 Raft Composite Foundation of High-Speed Railway. Transportation Research Board


31 93rd Annual Meeting Washington DC.
32 40. Zhang, Q. Q., Zhang, Z. M., He, J. Y. 2010. A simplified approach for settlement
33 analysis of single pile and pile groups considering interaction between identical piles
rin

34 in multilayered soils. Computers and Geotechnics, 37(7-8): 969-976.


35 41. Zhu, W., Ye, G., Gu, L., Zhang, F. 2022. 1g model test of piled-raft foundation
36 subjected to vibration load and its simulation considering small confining stress. Soil
37 Dynamics and Earthquake Engineering, 156: 107212.
ep
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4385752

You might also like