A

PROJECT REPORT

On

FORMULATION OF A RELATION FOR

PULLOUT LOAD OF INLCLINED SQUARE
ANCHOR PLATES FOR UNREINFORCED SOFT
CLAY UNDER CYCLIC LOADING
Submitted to

JADAVPUR UNIVERSITY
In partial fulfillment of requirements for the award of the degree of

Bachelor of Engineering
In
Civil Engineering

By
GAGAN DHANUKA
(ROLL NO: 001910401180 CLASS: BCE IV-B3)

Under the Guidance of

SUMIT KUMAR BISWAS
Professor of
CIVIL ENGINEERING DEPARTMENT

JADAVPUR UNIVERSITY
KOLKATA-32

DEPARTMENT OF CIVIL ENGINEERING, JADAVPUR UNIVERSITY

188, Raja S.C. Mallick Rd, Kolkata- 700032

1
 CERTIFICATE OF RECOMMENDATION

This is to certify that the Fourth Year First Semester Interim Major Project Report titled
“FORMULATION OF A RELATION FOR PULLOUT LOAD OF INLCLINED
SQUARE ANCHOR PLATES FOR UNREINFORCED SOFT CLAY UNDER
CYCLIC LOADING” is being submitted by GAGAN DHANUKA(Class Roll
No.001910401180) to the Civil Engineering Department, Jadavpur University for the partial
fulfillment of the requirement for awarding the undergraduate degree is a record of bonafide
project work carried out by him under my direct supervision and guidance. The project work
contained in this interim report has not been submitted in part or full to any other university
or institution or professional body for the award of any degree or diploma.

-------------------------------------------------

Thesis supervisor
Prof. Dr.Sumit Kumar Biswas
Professor
Department of Civil Engineering, Jadavpur University Kolkata, India.

2
 ACKNOWLEDGEMENT

I owe my most sincere thanks and profound gratitude for the indispensable advice and
inspiration rendered by the supervisor, Prof. Sumit Kumar Biswas at each phase of the
project work. Thank you for providing me with such an opportunity to work on such a topic
and directing me with your knowledge and experience in many long discussions on various
topics related to my project.

I would also like to take this opportunity to express my gratefulness to all the respected
teachers of the Civil Engineering Department of Jadavpur University for their valuable
suggestions and their continuous support and to all my beloved classmates for their
continuous inspiration, assistance, and cooperation to pursue this project work. Again, I
would like to thank, Prof. Sumit Kumar Biswas for providing various references and
resources to refer to which helped in the completion of my project.

I would like to express my gratitude to my family, for their unconditional support and
prayers at all times.

Last but not the least; I am thankful to the Almighty and also to all those, whose efforts
either directly or indirectly have contributed well during the course of this project work.

Date: 27.4.2023
Place: Jadavpur University, Kolkata

Signature of candidate

3
 CONTENTS

SL NO. DESCRIPTION PAGE NO.

1 INTRODUCTION 6

2 LITERATURE REVIEW 7 TO 16

3 MOTIVATION FOR WORK 16

4 OBJECTIVE 16

5 STATISTICAL ANALYSIS AND RESULTS 17 TO 37

6 SUMMARY, CONCLUDING REMARKS AND FUTURE SCOPE OF WORK 38

7 REFERENCE 38 TO 40

4
 LIST OF SYMBOLS

SYMBOL DESCRIPTION OF SYMBOL

° = Angle in Degree

Φ = Angle of Shearing Resistance

δ = Angle of Wall Friction

Cu = Undrained Cohesion of Soil

ϒ = Bulk Unit Weight of Soil

ϒd = Dry Unit Weight of Soil
Kp = Coefficient of Passive Pressure
W = Weight of Soil
Wa = Weight of Anchor Plate
Fc = Breakout Factor

β = Inclination of Anchor
Qg = Gross Ultimate Load

Qu = Net Ultimate Load

H = Depth of embedment of Anchor

B = Width of Anchor Plate
H/B = Embedment Ratio
H’ = Distance of Geotextile from bottom of Anchor Plate

Bg = Width of Geotextile

WL = Liquid Limit
PL = Plastic Limit

E = Young’s Modulus
μ = Poisson’s Ratio
UR = Unreinforced

5
 INTRODUCTION
 The foundations of many civil engineering structures are subjected to vertical or inclined tensile loads.
To withstand such loads, horizontal plate anchors are widely used both in onshore and offshore
structures. Different types of anchorages are used in the field depending on the size and type of loading,
the type of structure to be supported, the importance of the structures and the conditions of the subsoil.
 Over the almost four past decades, researchers (like Rowe and Davis (1982), Dickin (2002), Merifield
et al. (2001) and many other renowned researchers) have presented theoretical and experimental studies
on the behaviour of plate anchors in different types of soils under different conditions.
 An excellent description and use of different type of anchors in field are reported by Dutta and Singh
(1984). Anchorage by horizontal plate anchor, of different shapes like square, circular, rectangular is one
of the most common types of anchors used in civil engineering constructions. The ultimate strength of
these anchors depends on the shape and size of the anchor, the depth of anchorage, the characteristics of
the coating soil, the inclination of the tensile loads, etc. However, when the depth of anchor is shallow,
excavation costs less to accommodate the anchor, and control of pit placement is easier and safer.
 However, in order to withstand the tensile load, the anchor plate size or depth of penetration, or both,
must be increased depending on the size of the excavation area and the depth of the excavation. This not
only leads to an increase in the area and cost of excavation work, but also problem of excavation below
possible existing water table and compacting fill material below water table at great depths. In such
condition it will be worthwhile to search alternate cost-effective method to improve the resistance
capacity of a shallow anchor by adopting suitable method as suggested by Khatun and Chattopadhyay
(2010).
 It is known that the capacity of the anchor plate can be increased by grouping the anchors, increasing the
unit weight of the embedded soil, increasing the embedment depth and the size of the anchor plate. But,
in present days one of the possible alternatives for such kind of problem is use of geosynthetics. Subba
Rao et al. (1988), Krishnaswamy and Parashar (1994), Bhattacharya et al. (2008), Ravichandran et
al. (2004), Ravichandran et al. (2008) studied the pull-out behaviour of the anchors in various types of
soils with reinforcement and obtained the capacity considering the effects of embedment, size and shape
of the anchor, reinforcement and density of soil under different loading conditions. Without changing the
embedment depth and the size of the anchor plate, the pulling capacity of these anchors can be increased
by introducing a coaxial geotextile sheet onto the anchor plate and then compacting the fill over it. The
increase in the tensile load-carrying capacity of such a combination will depend on the relative size of the
coaxial geotextile laid, relative to that of the anchor plate, the depth of anchorage, the characteristics of
the geotextile layer and the properties of the filling material.

The present study is aimed to understand the relation of pull-out behaviour as well as breakout factor
of square anchor plates under unreinforced soft clay by using numerical values of unreinforced soft
clays for different inclination and embedment ratios under cyclic loading. A detailed review of
available literature is mentioned later.

6
 LITERATURE REVIEW
GENERAL:-
In this chapter, a review of available literature relevant to this research are to be furnished, overview
of previous research works done in the concerned areas of vertical and inclined anchors embedded
in different types of soil, are presented. The review has been presented for different methodologies
and in chronological order under each methodology.
In this context it is to be mentioned that approaches to analyze the response of vertical and inclined
anchors embedded in different types of soil may be classified in two broad spectrum : -
 Vertical and inclined plate anchors in sand.
 Vertical and inclined plate anchors in clay.

A)Vertical and inclined plate anchors in sand:-

 Balla (1961) conducted laboratory and field tests in dense sand, and found that for circular
shallow anchors, the failure surface extended to the ground surface as circular arcs. The ultimate
pull-out capacity was shown to comprise of two components: weight of soil in failure zone,
and the shearing resistance developed along the curved failure surface. The ultimate load
capacity was expressed as:

3
( Hh )+ F ( ϕ , Hh )]
Qu=H γ [ F1 ϕ , 2 ……….(1)

Where h = diameter of plate. F1 and F2 are functions dependent on the angle of internal friction Ø
and embedment ratio H/h.
 Meyerhof & Adams (1968) noticed that there is lack of agreement on uplift-capacity theories of
foundations based on slip surface mainly due to the difficulty of predicting the geometry of the failure
zone. Based on model tests, they proposed a semi- theoretical relationship for strip, circular &
rectangular footings in sand and clay soils. They observed very distinct failure pattern in sandy soils,
whereas the failure pattern was complicated in clayey soils due to the formation of tensile cracks. The
theory is derived for a strip footing and is then modified for circular and rectangular footings and for
group action. The same theory can be applied to plate anchors. The proposed relationship for ultimate
pull-out capacity of strip footings is:-
2
Qu=2 cH +ϒ H K p tanδ+W ..……….(2)

Where δ = Ø/2 to 2Ø/3, KP= coefficient of passive earth pressure, and W = weight of soil above the
footing.
 For inclined anchors, failure surface was observed to have radial and elliptical surface cracks as reported
by Harvey and Burley (1973).
7
 Chattopadhyay et al. (1986) proposed a theoretical model assuming a failure surface to predict the
ultimate break out capacity of horizontal circular plate anchors embedded in sand .The proposed theory
indicates that the existence of a characteristic relative depth, beyond which breakout factor approaches a
constant value.
 Ilamparuthi et al. (2002) proposed that the uplift capacity of circular anchors is governed by their
diameter, embedment ratio, and sand density. Two modes of failure develop within the soil mass
depending on the anchor embedment ratio. The surface anchoring behaviour is characterized by a raised
trunk of a soil cone extending from the anchor to the sand surface, with sloping sides approximately
φ / 2 to the vertical, regardless of the density of the sand.
The behaviour of the deep anchorage is characterized by a rupture zone in the form of a balloon in the
mass of the ground on the anchor. The flat part of this rupture surface emerges from the upper edge of
the anchor and is inclined at 0.8φ with respect to the vertical, it is also independent of the density of the
sand. A three-phase behaviour that characterizes the superficial case and a behaviour of two phases the
deep case.
 Ravichandran et al. (2004) studied the behaviour of rectangular plate anchors in the bed of
unreinforced and reinforced sand (horizontal and vertical). The vertical reinforcement showed a greater
increase in the lifting capacity of the anchors than the horizontal reinforcement. According to them, this
difference was due to the better interlocking arrangement attributed by vertical reinforcement than that
of horizontal reinforcement.
 Singh et al. (2011) investigated experimental, theoretical and numerical approaches for estimation of
load capacity of horizontal and inclined plate anchors. A parametric analysis of load capacity was then
carried out for plate anchors embedded in sandy soils by varying embedment ratio for horizontal anchors
and by varying inclination angle for inclined plate anchors. They found from the experimental and
numerical analysis that the anchors be installed deeply so that inaccuracy in the embedded depth should
not substantially affect the designed ultimate capacity.
 Makarchian et al. (2012) performed experimental and numerical investigation on behavior of uplift
capacity of circular plate anchors of different diameter embedded in sand of different relative density in
unreinforced and reinforced condition with multi- layers of geonet. Based on laboratory test and
numerical study they concluded that:-
1) With increase in embedment ratio pullout capacity increases for both unreinforced as well as for
reinforced condition.
2) The ultimate uplift capacity in dense sand condition is more than in medium dense condition.
3) In experimental tests, the axial movement required to reach ultimate pullout load is larger than FEM
approximately five times.
 Das et al. (2013) reported that with the use of a geotextile sheet of adequate diameter, the detachment
capacity of the shallow anchors could be increased in many folds, according to the increase requirement.
They presented a theoretical model to predict the breaking capacity of circular plate anchors covered by
a coaxial geotextile sheet. The detachment capacity of said combination depends on the diameter of the
anchor, the relationship between the diameter of the coaxial sheet and that of the anchor, the angle of
friction between the geotextile sheet and the surrounding soil, the depth of the embedment and the
properties of the soil surrounding.
 In 2016, Paramita Bhattacharya and Jyant Kumar studied the vertical pullout capacity of strip and
circular plate anchors embedded horizontally in a layered sandy medium. The results were computed by
using the plane strain and axi-symmetric lower-bound limit analyses in combination with finite elements
and linear optimization. The soil medium below the anchor plate was assumed to be comprised of loose
sand. Two different sand layers were considered above the anchor with different combinations of their
internal friction angles. For several embedment ratios (H/B), the variations of the pullout factors Fγ and
Fq due to the components of soil unit weight and surcharge, respectively, were computed as a function
8
 of Hdense/H for different cases (where, H and Hdense imply (A) the depth of the anchor plate from
ground surface and (B) the thickness of the dense sand layer, respectively). It were noted that the pullout
resistance increased continuously with an increase in Hdense/H. For a given H/B, with the same value of
Hdense/H, the uplift resistance became greater for a case when the dense sand layer was kept just above
the plate rather than placing it close to ground surface. Compared with strip anchors, the uplift factors
became considerably greater for circular anchors. The effect of the variation of the unit weights of the
two layers above the anchor plate on its pullout resistance was also revealed.
B)Vertical and inclined plate anchors in clay:-
 Vesic (1971) proposed an analytical approach for determining the pullout capacity of horizontal
plate anchors based on the solutions for the problem of an expanding cavity close to the surface
of a semi-infinite rigid plastic solid. These solutions gave the ultimate radial pressure needed for
the breakout of a cylindrical or a spherical cavity embedded at a depth below the surface of the
solid. The pullout capacities for strip and circular anchors were then assessed by assuming that
the pullout load was equivalent to the ultimate cylindrical or spherical cavity pressure, plus the
weight of soil acting directly above the anchors. The ultimate pullout capacity is given by:-
Qu= A (ϒH F q + c F C ) ………..(3)
Where A = Area of plate, H = embedment depth, c = cohesion of soil, Fq & FC= breakout factors.
 Based on experimental results of plate anchors embedded in clays, Meyerhof (1973) proposed the
following equation for ultimate pullout capacity:
Qu= A (ϒH +c F C ) …………(4)
For circular and square anchors,
H
F C =1.2∗( )≤ 9 …………(4.1)
h
For strip anchors,
H
F C =0.6∗( )≤ 8 …………(4.2)
h

 Based on various model test results, das (1980) suggested an empirical procedure in a non-dimensional
form for obtaining the breakout factor, Fc (β = 0˚). According to this procedure:-

………….(5)
For square anchors, Fc* ≈ 9 …………(5.1)
D
=2.5+0.107 Cu ≤ 7 …………(5.2)
B
Where Cu is in KN/m².
Based on above equations the variation of Fc (β=0˚) was obtained by compiling the laboratory model
test results of various authors (Table 1).

9
 (After Das (1980))
 Wang and Wu (1980) concluded that for constant relative depth ratio, the anchor resistance increases
with increase in anchor inclination. A probable explanation offered for this behaviour was the difference
in the extent of the ruptured soil zone. They stated that extent of ruptured soil zone increases with anchor
inclination and the rate of this extension depends on the relative depth ratio. Fig.2.1 shows the nature of
the failure surface in soil for horizontal, inclined, and vertical plate anchors in clay subjected to ultimate
pull out load.

Fig. 2.1.Nature of failure surface in soil at ultimate load (After Das (1980)).
(a) horizontal anchor ;(b) inclined anchor ;(c) vertical anchor.
The width of the anchor in each case is equal to B, and the average depth of embedment is D. the net
ultimate holding capacity for an inclined anchor can be given by:-
QU =Q g – W α cos β ……………(6)
Where QU = net ultimate load; Qg= gross ultimate load; Wα= self-weight of the anchor ; β= inclination
of the anchor with respect to the horizontal.
The net ultimate holding capacity can be expressed as:-
QU = A C u F c – W cos β …………(6.1)
Where A= area of the anchor plate, cu= undrained cohesion the clay, Fc = break out factor
and W= weight of the soil immediately above the anchor (Fig. 2.2).

Fig. 2.2 .Geometric parameters of an inclined square anchor in clay. (After Das (1980)).

For square anchors measuring B x B:-

………….(6.3)

10
 If the square anchor is horizontal (fig 2.1 (a)), β = 0.

…………..(6.4)
So similarly for vertical square anchors (fig-2.1(c)), β=90˚.

…………….(6.5)
The break out factor, Fc, for a given anchor inclination increase with the average embedded ratio, D/B up
to a maximum Fc= Fc* and remains constant thereafter. The average embedded ratio at which Fc=Fc* is
attained is referred to as the critical embedded ratio (D/B)cr, this is shown in figure 2.3.

Fig 2.3 Nature of variation of Fc with (D/B) for a given anchor inclination, β(after Das (1980)).

For D/B ≤ (D/B)cr, the anchor is referred to as a shallow anchor since the failure surface at ultimate load
extends to ground surface. At D/B > (D/B)cr, local shear failure around the anchor take place and it is
called as a deep anchor.

Fig. 2.4. Variation of Fc v/s {(D/B)/(D/B)cr} for square anchor in clay (ϕ = 0 condition) (after Das
(1980)).

 Das, Tarquin and Moreno (1985) suggested an empirical relationship for obtaining the breakout factor
for vertical anchor in clay (β=90˚); for square anchors, their results can be summarized as follows:-
(D/B) cr = 4.2 + 6.06x10²‫ ־‬Cu ≤ 6.5 ……..(7)
Where Cu is in KN/m².

11
 …..(7.1)
The empirical relationship given by above equation were based on model test with a plate anchor having
dimensions of 50.8mm x 50.8mm embedded in clays having average Cu’s of 20.31KN/m² and 42.45
KN/m² and with an anchor measuring 38.1 mm x 38.1 mm embedded in clays with average Cu’s of 16.0
KN/m² and 52.04 KN/m².
 Rowe & Davis (1982) carried out an elasto-plastic finite element analysis of the undrained behaviour of
anchor plates in homogeneous, isotropic saturated clay, so as to predict the behaviour of strip anchor
plates buried to various depths, with both vertical and horizontal loading. Consideration was also given
to the effect of anchor thickness and shape. The numerical solutions were obtained using a soil-structure
interaction theory with the following considerations: plastic failure within the soil, anchor break away
from the soil behind the anchor, and shear failure at a frictional dilatant soil-structure interface without
the introduction of special joint or interface elements.
The results were compared with the authors’ model tests and other available experimental data. The
results were presented in the form of charts which can be used in hand calculations for determining
design failure loads.
 Saran et al. (1986) proposed an analytical procedure to predict the load-displacement characteristics of
plate anchors in C-Ø soils using a non-linear constitutive relationship. Expressions were presented to
obtain critical loads and breakout loads for strip, square, and circular anchors. The validity of the
proposed theory was established by comparing it with large field and laboratory data, and it was found to
be valid for shallow anchors only. The change in behaviour between shallow and deep anchors is
explained with the help of critical depth ratio, which varies with size and shape of anchor as well as soil
parameters. In soft clays, the critical depth ratio for strip anchors is about 1.5 to 2 times that for
circular/square anchors. The values are 3 for strip anchors and 1.75 for circular anchors. They proposed
the following expression for ultimate pullout capacity:
Qu= A (ϒH F γ +c F C ) ………..(8)

Where Ø = angle of internal friction of soil, and Fc &Fγ= breakout factors dependent on depth to width
ratio.
For strip anchors:
Fc = H/B = 2λ ………(8.1)
Fγ= 1 + λ tanØ ………(8.2)
For square and circular anchors:
FC = 4 λ (1 + λ tan Ø) ………(8.3)
Fγ = 1 + 2 λ tan Ø + (4λ2/3) tan Ø (8.4)
 Das and Puri (1989) presented laboratory model test results for the holding capacity of inclined shallow
square plate anchors in clay (φ=0) concept. The inclination of the anchor with respect to the horizontal
was varied from zero to 90°. Based on the model study, an empirical relationship for estimating the
ultimate holding capacity of the anchors has been presented by them.
12
 Ghaly et al (1991) presented an experimental and theoretical analysis on the behaviour of spiral screw
anchors in non-cohesion soils. A mathematical model was developed using the limit equilibrium method
based on the failure mechanism observed in the experimental work. It was found that the theoretical
model was in agreement with the results of the experimental and field tests, in which the pullout capacity
depended on the depth of the anchor and the angle of shearing resistance of the sand.
 Nene and Garg (1991) investigated the behaviour of shallow plate anchors in reinforced cohesive soil
using both woven and non-woven geotextile. The breakout loads were computed by limit equilibrium
method. In order to validate the analytical method, model laboratory tests were conducted in cohesive
soil with circular and square anchors with geo-synthetic placed at a distance of B/2 and B, B being the
width of anchor, from the top of the plate anchor and width four times the width of model footing. Based
on limited number of model tests at two different embedment depths they concluded that the uplift
capacity increased both in woven and non woven geo-textile used although greater increase was
observed for woven geo-textile. The experimental results were found to be in good agreement with the
analytical method suggested.
 Rao & Kumar (1994) used the method of characteristics coupled with a log-spiral failure surface to
develop a theory for vertical uplift capacity of shallow horizontal strip anchors in a general C-Ø soil.
They adopted a methodology similar to that used in finding the bearing capacity of foundations under
compression, and separated the effects of cohesion, surcharge, and density on the uplift capacity. The
theory was shown to be capable of predicting accurately anchor pullout behaviour in clays and also in
loose and medium-dense sands.
The proposed equation for net ultimate pullout capacity per unit length of the strip anchor in clays is:
QU , net=C F C +q F q +0.5 ϒB Fϒ ………………(9)
Where C= cohesion of soil, γ = unit weight of soil, B = width of plate anchor, and F C, Fq, Fϒ= uplift
capacity factors which are functions of embedment ratio and soil friction angle.
 Merifield et al. (2003) applied three-dimensional numerical limit analysis to evaluate the effect of
anchor shape on the pullout capacity of horizontal anchors in undrained clay. The anchor was idealized
as either square, circular, or rectangular in shape. Estimates of the ultimate pullout load were obtained by
using a newly developed three-dimensional numerical procedure based on a finite-element formulation
of the lower bound theorem of limit analysis. This formulation assumed a perfectly plastic soil model
with a Tresca yield criterion.
They presented results in the familiar form of breakout factors based on various anchor shapes and
embedment depths, and also compared them with existing numerical and empirical solutions. They
expressed the ultimate pullout of plate anchors as follows:
QU = A CU NC ……………….(10)
Where, NC = NC0 + (ϒH/CU) (24) ……………….(10.1)
NC0 = S [2.56 loge(2H/B)] ……………..(10.2)
And S = shape factor.
 Thorne et al. (2004) studied the uplift behaviour of horizontal strip anchors in clay under fast loading.
The possible failure mechanisms were reviewed, including failure due to shear and traction in the soil
and the development of suction in the porous fluid. Based on their findings the following conclusions
were drawn :
 The behaviour of the strip anchors in the pullout capacity are functions of the non-dimensional
parameters H/B, ϒH/C, Uc/C ,where H is the embedment depth, C is the cut resistance without
drainage, Uc is the magnitude of the maximum tensile stress of the pore water in the soil and ϒ
and B are the unit weight and width of plate.
 b) Shallow anchors in relatively strong soil tend to fail due to the development of a tensile
failure in the soil that is above the anchor and the ultimate capacity is a function of the undrained
shear strength of the soil, its own weight and the tensile capacity of the porous fluid

13
  c) The failure mechanism of the deep anchors where the initial vertical total stress in the plate
exceeded seven times the resistance without draining involved only one cut fault located around
the anchor. The ultimate capacity in such a case becomes a function only of the resistance
without draining the soil.
 Merifield et al. (2005) studied the ultimate capacity of inclined strip anchors. Consideration was given
to the effect of embedment depth and anchor inclination. The results were presented as breakout factors
in chart form to facilitate their use in solving practical design problems. They found that using the lower
and upper bound limit theorems, small error bounds of less than ±7% were achieved on the true value of
the breakout factor for anchors inclined at 22.5, 45, and 67.5° to the vertical in a weightless soil and also
the ultimate anchor capacity increases linearly with overburden pressure up to a limiting value that
reflects the transition from a no localized to localized or “deep” failure mechanism.
 Randolph et al. (2005) investigated the behaviour of inclined strip plate anchors using finite element
analysis of high deformation. The rotation behaviour of the anchor plates during continuous extraction in
clay was introduced by the finite element method. They observed that the extraction capacity increased
with increasing insertion depth and inclination with a final extraction capacity factor Nc = 12 at an
embedment ratio of 3 for the anchors attached to the ground. But for the anchors that were allowed to
separate from the ground, a limiting value of the uplift capacity factor was not obtained until an
embedment ratio of 8 in soil without weight. They also concluded that for the analysis of the pullout
capacity of small deformation, the factor increased with increase in the depth of embedment and before
the limiting capacity factor for a given embedment was reached, the vertical plate anchor showed pullout
capacity smaller than that of the horizontal anchor.
 Bhattacharjee et al. (2008) studied behavior of square plate anchors under uplift load in reinforced clay
using a three dimensional finite element displacement model with ANSYS software. Soil anchor System
was discretized with eight nodediso-parametric brick elements for the soil and four nodediso-parametric
shell elements for the plate. Geotextile used as reinforcing material was modeled with two noded spar
element activating tension only. Nonlinear soil behaviour was considered with Drucker-Prager model
and the geometrical nonlinearity of geotextile was also addressed in the analysis. They found that,
compared to unreinforced clay, the ultimate uplift capacity was more in reinforced clay with less
ultimate displacement. The uplift capacity was found to be dependent on embedment ratio and the
position of the geotextile with respect to the embedment depth. The pullout capacity was found to
increase with increase in embedment ratio but it decreased with the increase in height of placement of
geotextile above the plate.
 Song et al. (2008) studied the behaviour of circular and strip plate anchors during vertical pullout, with
fully attached and vented rear faces of anchors, in uniform and normally consolidated clays by means of
small strain and large deformation finite element analyses. They proposed the following relation for
vented strip plate anchors:-
QU = A C NC ………….(11)
Where, Nc= 8.6 + 2.2(H/B) for H/B ≤ 1.4 (11.1)
Nc= 11.7 for H/B ≥ 1.4 (11.2)
 Khatri & Kumar (2009) employed an axi-symmetric static limit analysis formulation in combination
with finite elements to obtain the vertical uplift resistance of circular plate anchors, embedded
horizontally in a clayey stratum whose cohesion increases linearly with depth. The variation of the uplift
factor with changes in the embedment ratio was computed for several rates of increases of soil cohesion
with depth. It was noted that in all cases, the magnitude of the uplift factor increases continuously with
depth up to a certain value of critical embedment ratio, beyond which it becomes essentially constant.
 Wang et al. (2010) performed three-dimensional large deformation finite-element analyses to
investigate plate anchor capacity during vertical pullout. Continuous pullout of plate anchors was

14
 simulated, and the large deformation results for strip, circular, and rectangular anchors were compared
with model test data, small strain FE results, and plastic limit solutions. The effects of anchor roughness,
aspect ratio, soil properties, and soil overburden pressure were investigated. It was found that the anchor
roughness had minimal effect on anchor performance. The soil beneath the anchor base separates from
the anchor at a certain embedment depth near the mud line, once tensile stresses were generated. The
ratio of separation depth to anchor width was found to increase linearly with the ratio of soil undrained
shear strength to the product of soil effective unit weight and anchor width, and was independent of the
initial anchor embedment depth. They expressed the maximum uplift capacity of rectangular plate
anchors as follows:
QU = A CU NC ………….(12)
NC = NC0 + (ϒH/CU) ……….(12.1)
Where, Nco= Anchor capacity factor in weightless soil.
 Mistri et al. (2011) presented the analysis of finite elements for the anchoring of plates in homogeneous
and non-homogeneous soils using the PLAXIS 3D. In the initial stages, the final uplift capacity in
homogeneous clay shows a rapid increase and can become almost constant at great depth. They proposed
that such a change in the rate of increase occurs at the depth of transition where the behavior of the
surface anchor changes to the deep anchor. However, such transitional behavior is not observed
markedly in the clay when increasing the shear strength. As the consistency of the soil increases, the
variation in the final uplift capacity also increases.
 Ardebili et al. (2016) carried out a study on the pullout capacity of anchor by using Mohr- Coulomb
model, Modified Cam-Clay and Soft Soil models. A series of finite- element analyses were performed
using the three constitutive models. Undrained effective stress analyses were conducted to study the
response of both strip and circular plate anchors in saturated soils. The capacities of plate anchors were
assessed through the application of the displacement control approach. The effective stress parameters
were correlated with the undrained shear strength, and Nc (dimensionless breakout factor) values from
the three constitutive models were presented and compared with the lower bound solution. Effective
stress analyses were performed while using MC, MCC, and SS constitutive models, for both plane-strain
and axisymmetric conditions.
 Beirne et al. (2017) investigated field data from reduced scale anchor tests at two sites to validate a new
release-to-rest model for dynamically installed anchors. This model considered the motion of the anchor
from the point of release in the water column, modeling the drag resistance acting on the anchor and its
mooring line. They stated that although dynamically installed anchors were an attractive and often a
cost- effective anchoring solution, their global acceptance had been somewhat hampered by uncertainties
on their striking the seabed within an acceptable spatial variation, and on achieving the targeted
embedment depth in the seabed. The latter was addressed in their research paper through a new release-
to-rest model for anchor installation. The model simulates the motion history of the anchor during free-
fall in water and dynamic embedment in soil, providing as output the final anchor embedment depth as
output to calculate anchor capacity. The importance of considering the motion response in water was
demonstrated through model simulations that highlighted the role of drag resistance acting on the trailing
mooring line. The observations from these simulations were also reflected in measurements made in
field tests. Those field test results were used to validate the model.

Based on the review of the literature in relation to the researches subjected to the approaches to analyze
pullout behavior of anchor in clay, it is evident that so far different experimental, numerical and
analytical approaches have been adopted to assess the pullout behavior of anchor in clay. On the basis of
the outcome of these past works an attempt has been made in the present study to assess the pullout
behavior of inclined square plate anchor in clay. To complete such study, experimental investigations
have been carried out as well as numerical analyses have been executed. Consequently the results

15
 obtained from experimental studies and numerical analyses have been observed to be in a good
agreement when compared with each other.
 Biradar. J et al. (2019) carried out numerical simulations on three different sizes of square anchor
plates. A single layer geosynthetic was used as reinforcement in the analysis and placed at three different
positions from the plate. The effects of various parameters like embedment ratio, position of
reinforcement, width of reinforcement, frequency and loading amplitude on the pull out capacity were
presented in the study. The load-displacement behaviour of anchors for various embedment ratios with
and without reinforcement was also noted. The pull out load, corresponding to a displacement equal to
each of the considered maximum amplitudes of a given frequency, was presented in terms of a
dimensionless breakout factor. The pull out load for all anchors was noticed to increase by more than
100% with embedment ratio varying from 1 to 6. Finally a semi empirical formulation for breakout
factor for square anchors in reinforced soil was proposed by carrying out regression analysis on the data
obtained from numerical simulations.
In the above section the studies related with analyzing the response of horizontal and inclined anchors
embedded in different types of soil, have been covered. It has been noted form the above sections that
researchers have been studied the response of horizontal and inclined anchors embedded in different
types of soil from both experimental as well as numerical approach.
 Majumder et al. (2019) investigated the pullout behaviour of plate anchors in geotextile reinforced soft
clay by using experimental model set up and numerical model analysis in ABAQUS software. They
concluded that the pullout capacity increases with the increase in plate size, embedded depth and
inclusion of geotextile as reinforced material.

16
 MOTIVATION FOR WORK
If the considerations evolved from the critical appraisal of the literature review (concerned with the
investigations subjected to the approaches to analyse response of Vertical and inclined anchors embedded in
different types of soil) are to be shaped judiciously then it is observed that there is still some gap in relation
with evaluating the response of geotextile unreinforced Vertical and inclined anchors embedded in soft clay.
So, on the basis of the outcome of these past works an attempt has been made in the present study to
formulate an empirical relation via statistical analysis of the response of geotextile unreinforced Vertical and
inclined anchors embedded in soft clay.
Based on this, the objective and scope of the work have been decided for the present research.

OBJECTIVE
The aim of present study has been concentrated on finding out an empirical relationship between the pullout
behavior of inclined square plate anchors of different width which have been embedded in soft clay with
different inclination, with different amplitude and frequency.
Hence the determination of pullout capacity for different sizes of square anchor plates has been envisaged
with different embedment depths with different angle of inclination in unreinforced soil. With the above in
view, the following objectives have been identified for the present study:
i.Study the relation of pullout behaviour of inclined plate anchors with and without reinforcement in
soft clay by statistical analysis of experimental and theoretical investigation under the effect of shape,
angle of inclination, size of anchors and embedment depth, with different amplitude and frequency on
such pullout of plate anchors.
ii. Study the relation of breakout factor of inclined plate anchors with and without reinforcement in
soft clay by statistical analysis of experimental and theoretical investigation under the effect of shape,
angle of inclination, size of anchors and embedment depth, with different amplitude and frequency on
such pullout of plate anchors.
iii.To compare the results obtained from experimental and Numerical Analyses.

17
 STATISTICAL ANALYSIS AND RESULTS
GENERAL:-
Statistical analysis is the collection and interpretation of data in order to uncover patterns and trends.
Statistical analysis can be used in situations like gathering research interpretations, statistical modelling or
designing surveys and studies. The goal of statistical analysis is to identify trends. Statistical analytic tools
can help with predictive modelling.
THEORY REVIEW:-
LINEAR REGRESSION:-
It is a statistical technique used to establish a relationship between a dependent variable and one or more
independent variables. It assumes that there is a linear relationship between the variables, which means that
a change in the independent variable will result in a proportional change in the dependent variable.
The basic equation for a linear regression model with one independent variable (also called simple linear
regression) is:
y = β0 + β1x + ε
Where:
• y is the dependent variable (the one you are trying to predict)
• x is the independent variable (the one you are using to make the prediction)
• β0 is the intercept (the point where the line intersects the y-axis when x is equal to 0)
• β1 is the slope (the rate at which y changes with a one-unit increase in x)
• ε is the error term (the difference between the predicted value of y and the actual value of y)
To estimate the parameters β0 and β1, the method of least squares is commonly used. This involves
minimizing the sum of the squared differences between the predicted values and the actual values of the
dependent variable.

STANDARD DEVIATION:-
Often denoted as σ (sigma) for a population or s for a sample, is a measure of the variability or dispersion of
a set of data values. It is one of the most commonly used measures of variability in statistics.
The standard deviation is calculated as the square root of the variance, which is the average of the squared
differences between each data point and the mean of the data set. The formula for calculating the standard
deviation for a sample is:
s=√ (Σ ((xi−x̄ )¿¿ 2)/(n−1))¿
where s is the standard deviation, xi represents each individual data point in the sample, x̄ is the sample
mean, n is the sample size, and Σ represents the sum of the values.

18
The standard deviation provides information about the amount of variation or spread in a set of data values.
A larger standard deviation indicates a greater amount of variation or spread, while a smaller standard
deviation indicates less variation or spread. It is commonly used to assess the variability of a data set,
describe the dispersion of data points around the mean, and quantify the degree of uncertainty or risk
associated with a set of data values.
The standard deviation is a fundamental concept in statistics and is used in various statistical analyses, such
as hypothesis testing, confidence interval calculations, and modeling. It is important to understand the
standard deviation when interpreting and analyzing data to make meaningful statistical inferences and draw
accurate conclusions.

STANDARDIZED RESIDUALS:-
It is a measure of the residuals (i.e., the differences between observed and predicted values) in a linear
regression model that have been scaled by their standard deviation. Standardized residuals are useful for
identifying outliers and assessing the goodness of fit of a linear regression model.
The formula for calculating standardized residuals in a linear regression model is:
Standardized Residual (r*) = (Residual (r)) / (Standard Deviation of Residuals (s))
where:
• Residual (r): The difference between the observed value of the dependent variable and the
predicted value from the linear regression model for a particular data point.
• Standard Deviation of Residuals (s): The standard deviation of the residuals in the linear
regression model.
Standardized residuals are typically calculated for each data point in the dataset and can be positive or
negative values. A standardized residual close to 0 indicates that the data point's residual is close to the
expected value based on the linear regression model. A standardized residual with an absolute value greater
than 2 or 3 may indicate an outlier or influential data point that has a substantial impact on the model's fit.

R-SQUARED (R2):-
It is a common metric used to assess the goodness of fit of a regression model. It represents the proportion of
the total variation in the dependent variable that is explained by the regression model. R2 ranges from 0 to 1,
where a higher value indicates a better fit of the model to the data.
A high R2 value (close to 1) indicates that the regression model explains a large proportion of the variability
in the dependent variable, suggesting a good fit. On the other hand, a low R2 value (close to 0) indicates that
the regression model explains a small proportion of the variability in the dependent variable, suggesting a
poor fit.

CORRELATION MATRIX:-
It is a square matrix that displays the correlations between pairs of variables in a dataset. Each element in the
matrix represents the correlation coefficient between two variables, which measures the strength and
direction of the linear relationship between them. The correlation coefficient can range from -1 to 1, where -
1 indicates a perfect negative linear relationship, 0 indicates no linear relationship, and 1 indicates a perfect

19
positive linear relationship. A correlation matrix is often used in statistics and data analysis to explore the
relationships between multiple variables simultaneously. It can be used to identify patterns, dependencies,
and associations between variables. Correlation matrices are commonly used in fields such as finance,
economics, social sciences, and machine learning for tasks such as feature selection, risk assessment, and
predictive modeling.

P-VALUE (PROBABILITY VALUE):-

It is a measure that quantifies the strength of evidence against a null hypothesis in a hypothesis test. It is
commonly used in hypothesis testing to determine whether the results of a statistical analysis are statistically
significant or not.
The null hypothesis (H0) is a statement that assumes no effect or no difference between groups or variables,
while the alternative hypothesis (Ha) is a statement that posits an effect or difference. The p-value represents
the probability of obtaining a test statistic as extreme or more extreme than the one actually observed,
assuming that the null hypothesis is true. In other words, it is the probability of observing the observed test
statistic, or a more extreme value, purely by chance under the assumption that the null hypothesis is true.
If the p-value is lower than a predetermined significance level (often denoted as α), typically set at 0.05 or
0.01, it is considered statistically significant. This implies that the evidence against the null hypothesis is
strong, and the null hypothesis is rejected in favor of the alternative hypothesis. Conversely, if the p-value is
higher than the significance level, the evidence is not strong enough to reject the null hypothesis, and no
statistically significant conclusion can be drawn.
It is important to note that a p-value does not provide information about the magnitude or importance of an
effect or the probability of the null hypothesis being true. It only indicates the strength of evidence against
the null hypothesis based on the observed data and the chosen significance level. Therefore, interpretation of
p-values should be done cautiously, and other factors such as effect size, sample size, study design, and
practical significance should also be considered in making inferential statistical conclusions.

F VALUE (F-STATISTICS):-
It is a measure used in analysis of variance (ANOVA) and regression analysis to assess the overall
significance of a model or the differences among group means. It is named after Sir Ronald A. Fisher, a
prominent statistician.
The F value is calculated by comparing the variance between group means (often referred to as the
"explained variance") to the variance within groups (often referred to as the "unexplained variance"). The F
value is calculated as the ratio of the explained variance to the unexplained variance, and it follows the F-
distribution under the assumption of the null hypothesis being true.
In regression analysis, the F value is used to assess the overall significance of a regression model, which
includes multiple predictor variables. It measures the goodness of fit of the entire regression model by
comparing the variance explained by the model (often referred to as the "regression sum of squares") to the
variance unexplained by the model (often referred to as the "residual sum of squares"). The F value is
calculated as the ratio of the regression sum of squares to the residual sum of squares, and it follows the F-
distribution under the assumption of the null hypothesis being true.
A higher F value indicates a greater degree of difference between group means or a better fit of the
regression model. The F value is typically compared to a critical value or evaluated against a pre-determined
significance level (often denoted as α) to determine whether the model or group differences are statistically
20
significant. If the calculated F value exceeds the critical value or if the p-value associated with the F value is
lower than the significance level, it indicates that the model or group differences are statistically significant,
and the null hypothesis is rejected in favor of the alternative hypothesis.
CONFIDENCE INTERVAL (CI):-
It is a range of values that is used to estimate an unknown population parameter, such as a mean, proportion,
or regression coefficient, with a specified level of confidence. It is an interval estimate that provides a
measure of the uncertainty associated with the estimated parameter.
Confidence intervals are commonly used in inferential statistics to provide a range of plausible values for a
population parameter based on a sample statistic. The confidence level, often denoted as (1 - α), where α is
the significance level, represents the probability that the confidence interval contains the true population
parameter. For example, a 95% confidence interval implies that there is a 95% probability that the interval
contains the true parameter, and only a 5% probability that it does not.
A confidence interval is typically calculated using a point estimate, such as a sample mean or sample
proportion, and a margin of error. The margin of error is calculated based on the variability of the sample
statistic and the desired level of confidence. The most common method for calculating confidence intervals
is using the normal distribution for large sample sizes and the t-distribution for small sample sizes.
Confidence intervals provide a range of plausible values for a population parameter, taking into account the
uncertainty associated with sample estimates. They are useful for making inferential conclusions and
interpreting the precision of sample estimates in statistical analyses.

PREDICTED VALUE VS ACTUAL VALUE GRAPH:-

In Statistics also known as a predicted vs observed plot, it is a graphical representation that allows for visual
comparison between the predicted values and the actual values of a set of data points. It is commonly used to
assess the accuracy and performance of a prediction or estimation model, such as a regression model or a
machine learning model.
The predicted value vs actual value graph is typically created by plotting the predicted values on the x-axis
and the actual values on the y-axis. Each data point represents a predicted value and its corresponding actual
value. The graph allows for a visual comparison of how well the predicted values align with the actual
values.
The significance of a predicted value vs actual value graph in statistics can be summarized as follows:
 Model accuracy assessment: The graph provides a visual assessment of the accuracy of the
prediction or estimation model. If the predicted values are close to the actual values, the points on the
graph will be closely clustered around a diagonal line, indicating that the model is accurately
predicting the actual values. On the other hand, if there are significant deviations between the
predicted and actual values, the points on the graph may be scattered or biased, indicating that the
model may have limitations or inaccuracies.
 Model validation: The graph can be used as a tool for model validation, to check whether the
predicted values are consistent with the actual values. It can help to identify potential outliers, trends,
or patterns in the discrepancies between the predicted and actual values, which may indicate issues
with the model or data quality.
 Model improvement: The graph can provide insights into areas where the model can be improved.
For example, if there are systematic deviations between the predicted and actual values, the graph

21
 can help identify areas where the model may be underestimating or overestimating the values, and
adjustments to the model can be made accordingly.
 Communication and interpretation: The graph can be used as a visual tool for communicating the
performance of a prediction or estimation model to stakeholders or audiences who may not be
familiar with statistical concepts. It can help to visually convey the accuracy or limitations of the
model in a clear and understandable manner.
In summary, a predicted value vs actual value graph is a useful graphical tool in statistics for assessing
model accuracy, validating models, identifying areas for improvement, and communicating model
performance. It can provide insights and visual representations of the relationship between predicted and
actual values, aiding in the interpretation and analysis of prediction or estimation models.

IMPORTANT ASPECTS OF RESULTS:-

The Statistical analysis encompasses :-
a)Pullout behaviour of square anchor plates is formulated under cyclic loading for unreinforced soft clay
under different inclination, plate area, embedment ratios, with different amplitude and frequency in
terms of breakout factor.

A)PULLOUT LOAD FORMULATION IN UNREINFORCED SOFT

CLAY
UNDER CYCLIC LOADING:-
In this section formulation of pullout load with variation of different important design parameters (i.e. plate
size, inclination angle and embedment ratio, amplitude and frequency) under the effect of cyclic loading in
unreinforced condition, have been presented through linear regression with the help of different table and
graphs with concluding observations.
NOTE:- 1) case code: P = plate size ; E = embedment ratio ; A = Inclination angle wrt. Horizontal.
2) cases of plate size above 50mm square plate is separated due to analysis procedure.

TABLE:- PULLOUT LOAD FOR SQUARE ANCHOR PLATES UNDER CYCLIC LOADING

Em
be Pull Out Load Breakout
Plate Inclinati dm (in factor(in
Case theta (A)*(θ/90)*(H/
Case No. Size on Angle ent Corresponding correspond
Code /90 B)*(BF)(α/N)
(mm) (degree) Ra to 2mm ing to 2mm
tio amplitude) amplitude)
s
25P-
1 IE- 0.000625 90 1.00 1 6.10 4.79 0.002994
90A
Case 1 25P- 1 30.1 2.00
IE-
30 0.33
30A-
0.2Hz 0.000625 0.004125
Case 2 25P- 0.000625 30 0.33 2 35.91 3.0 0.012375
22
 2E-
30A-
0.2Hz
Case 3 25P- 3 39.13 3.00
3E-
30 0.33
30A-
0.2Hz 0.000625 0.0185625
Case 4 25P- 1 32.2 2.00
1E-
45 0.50
45A-
0.2Hz 0.000625 0.00625
Case 5 25P- 2 35.97 2.62
2E-
45 0.50
45A-
0.2Hz 0.000625 0.01635
Case 6 25P- 3 39.9 3.00
3E-
45 0.50
45A-
0.2Hz 0.000625 0.028125
Case 7 25P- 1 34.88 3.00
1E-
60 0.66
60A-
0.2Hz 0.000625 0.012375
Case 8 25P- 2 37.66 3.00
2E-
60 0.66
60A-
0.2Hz 0.000625 0.02475
Case 9 25P- 3 41.98 3.00
3E-
60 0.66
60A-
0.2Hz 0.000625 0.037125
Case 10 50P- 1 37.85 0.7
1E-
30 0.33
30A-
0.2Hz 0.002500 0.00567765
Case 11 50P- 2 41 0.75
2E-
30 0.33
30A-
0.2Hz 0.002500 0.01230075
Case 12 50P- 3 43.15 0.78
3E-
30 0.33
30A-
0.2Hz 0.002500 0.019416375
Case 13 50P- 1 39.25 0.7
1E-
45 0.50
45A-
0.2Hz 0.002500 0.00892
Case 14 50P- 0.002500 45 0.50 2 42.3 1 0.0192275
23
 2E-
45A-
0.2Hz
Case 15 50P- 3 45.94 0.84
3E-
45 0.50
45A-
0.2Hz 0.002500 0.03132375
Case 16 50P- 1 49.46 0.90
1E-
60 0.66
60A-
0.2Hz 0.002500 0.01483845
Case 17 50P- 2 52.16 0.9
2E-
60 0.66
60A-
0.2Hz 0.002500 0.0312972
Case 18 50P- 3 55.88 1.0
3E-
60 0.66
60A-
0.2Hz 0.002500 0.050292
Case 19 75P- 1 48.58 0.39
1E-
30 0.33
30A-
0.2Hz 0.005625 0.007287638
Case 20 75P- 2 55.24 0.45
2E-
30 0.33
30A-
0.2Hz 0.005625 0.0165726
Case 21 75P- 3 62.19 0.5
3E-
30 0.33
30A-
0.2Hz 0.005625 0.027982969
Case 22 75P- 1 51.65 0.42
IE-
45 0.50
45A-
0.2Hz 0.005625 0.011739375
Case 23 75P- 2 57.66 0.5
2E-
45 0.50
45A-
0.2Hz 0.005625 0.026206875
Case 24 75P- 3 66.16 0.53
3E- 45 0.50
45A-
0.2Hz
75P-
Case 25 IE- 54.32
60A- 0.005625 60 0.66 1 0.44 0.045106875
0.2Hz 0.005625 0.016294163
Case 26 75P- 0.005625 60 0.66 2 62.86 0.51 0.037719
24
 2E-
60A-
0.2Hz
Case 27 75P- 3 68.95 0.56
3E-
60 0.66
60A-
0.2Hz 0.005625 0.06205815

STATISTICAL ANALYSIS AND FORMULATION OF PULLOUT LOAD (UNREINFORCED)

FOR SQUARE PLATE SIZE (2Hz 2mm)
:Summary statistics:
Obs.
Obs. with
Observation without Minimu Maximu Std.
Variable missing Mean
s missing m m deviation
data
data
46.78
PULL OUT LOAD (F) 26 0 26 30.100 68.950 4 11.024
(A)*(θ/90)*(H/B)*(BF)
(α/N) 26 0 26 0.004 0.062 0.022 0.015
Summary statistics (Quantitative data / Validation):
Obs.
Obs. with
Observation without Minimu Maximu Std.
Variable missing Mean
s missing m m deviation
data
data
45.94
PULL OUT LOAD (F) 1 0 1 45.940 45.940 0  
(A)*(θ/90)*(H/B)*(BF)
(α/N) 1 0 1 0.031 0.031 0.031  

Correlation matrix:
  (A)*(θ/90)*(H/B)*(BF)(α/N) PULL OUT LOAD (F)
(A)*(θ/90)*(H/B)*(BF)(α/N) 1 0.709
PULL OUT LOAD (F) 0.709 1

Regression of variable PULL OUT LOAD (F) (UR):

Goodness of fit statistics (PULL OUT LOAD (F) (UR)):

25
 Statistic Training set Validation set
Observations 26 1
Sum of weights 26 1
DF 24 -1
R² 0.503
Adjusted R² 0.482
MSE 62.949
RMSE 7.934

:Analysis of variance (PULL OUT LOAD (F) (UR)):

p-values
Sum of Mean
Source DF F Pr > F signification
squares squares
codes
Model 1.000 1527.635 1527.635 24.268 <0.0001 ***
Error 24.000 1510.778 62.949
Corrected Total 25.000 3038.413        
Computed against model Y=Mean(Y) Signification codes: 0 < *** < 0.001 < ** < 0.01 < * < 0.05 < . < 0.1 < ° < 1

:Model parameters (PULL OUT LOAD (F) (UR)):

Lower Upper p-values
Standard
Source Value t Pr > |t| bound bound signification
error
(95%) (95%) codes
Intercept 35.038 2.847 12.306 <0.0001 29.162 40.915 ***
(A)*(θ/90)*(H/B)*(BF)
(α/N) 532.995 108.195 4.926 <0.0001 309.691 756.298 ***
Signification codes: 0 < *** < 0.001 < ** < 0.01 < * < 0.05 < . < 0.1 < ° < 1

:Equation of the model (PULL OUT LOAD (F) (UR)):

Pull Out Load (in Corresponding to 2mm amplitude) = 35 +532 *(A)*(θ/90)*(H/B)*(BF)(α/N)

:Standardized coefficients (PULL OUT LOAD (F) (UR)):

Upper p-values
Valu Standar Lower bound
Source t Pr > |t| bound signification
e d error (95%)
(95%) codes
(A)*(θ/90)*(H/B)*(BF)
(α/N) 0.709 0.144 4.926 <0.0001 0.412 1.006 ***
Signification codes: 0 < *** < 0.001 < ** < 0.01 < * < 0.05 < . < 0.1 < ° < 1

26
Pull Out Load (in Corresponding to 2mm amplitude)
Regression of Pull Out Load (in Corresponding to 2mm amplitude) by (A)*(θ/
90)*(H/B)*(BF)(α/N) (R²=0.503)
100
90
80
70
60
50
40
30
20
10
0
-0.01 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07

(A)*(θ/90)*(H/B)*(BF)(α/N)

Training set Validation set

Model Conf. interval (Mean 95%)
Conf. interval (Obs 95%)

Standardized residuals / (A)*(θ/90)*(H/B)*(BF)(α/N)

2

1.5
Standardized residuals

0.5

0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
-0.5

-1

-1.5

-2

(A)*(θ/90)*(H/B)*(BF)(α/N)

Training set Validation set

27
 Pull Out Load (in Corresponding to 2mm amplitude) / Standardized residuals
Pred(Pull Out Load (in Corresponding to 2mm amplitude)) / Standardized residuals
2
2

1.5
1.5

1
1

0.5
0.5
Standardiz ed residuals

Standardized residuals
0
0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70
-0.5 -0.5

-1 -1

-1.5 -1.5

-2 -2

Pull Out Load (in Corresponding to 2mm amplitude) Pred(Pull Out Load (in Corresponding to 2mm amplitude))

Training set Validation set Training set Validation set

Pred(PULL OUT LOAD (F) Pred(Pull Out Load (in Corresponding to 2mm amplitude)) - Pull Out Load (in Cor-
(UR)) / Standardized residuals responding to 2mm amplitude)
120
2.5

2
100
1.5

1 80
Pull Out Load (in Corresponding to 2mm amplitude)

0.5
60
Standardized residuals

0
0 5 10 15 20 25 30 35 40
-0.5

-1
20
-1.5

-2 0
0 20 40 60 80 100 120
Pred(PULL OUT LOAD (F)
(UR)) Pred(Pull Out Load (in Corresponding to 2mm amplitude))

Training set Validation set Training set Validation set

28
 Standardized residuals / Pull Out Load (in Corresponding to 2mm amplitude) (Train- Standardized residuals / Pull Out Load (in Corresponding to 2mm amplitude) (Valida-
ing set) tion set)

Obs29
Obs28
Obs27
Obs26
Obs25
Obs24
Obs23
Obs22
Obs21
Obs20
Obs19
Obs18
Obs16 Obs17
Obs15

Observations
Observations

Obs14
Obs13
Obs12
Obs11
Obs10
Obs9
Obs8
Obs7
Obs6
Obs5
Obs4
Obs3
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8

Standardized residuals Standardized residuals

STATISTICAL ANALYSIS AND FORMULATION OF PULLOUT LOAD(UNREINFORCED)

FOR SQUARE PLATE SIZE (AMPLITUDE,FREQUENCY)
:Summary statistics:
Obs.
Obs. with
Observation without Std.
Variable missing Minimum Maximum Mean
s missing deviation
data
data
116.38
PULL OUT LOAD (F) 26 0 26 75.150 172.390 8 27.490
(A)*(θ/90)*(H/B)*(BF)
(α/N) 26 0 26 0.026 0.371 0.131 0.080

:Summary statistics (Quantitative data / Validation):

Obs.
Obs. with
Observation without Std.
Variable missing Minimum Maximum Mean
s missing deviation
data
data
129.12
PULL OUT LOAD (F) 1 0 1 129.120 129.120 0  
(A)*(θ/90)*(H/B)*(BF)
(α/N) 1 0 1 0.070 0.070 0.070  

Correlation matrix:
  (A)*(θ/90)*(H/B)*(BF)(α/N) PULL OUT LOAD (F)
(A)*(θ/90)*(H/B)*(BF)(α/N) 1 0.598
PULL OUT LOAD (F) 0.598 1

:Regression of variable PULL OUT LOAD (F) (UR):

Goodness of fit statistics (PULL OUT LOAD (F) (UR)):
Statistic Training set Validation set
Observations 26 1
Sum of weights 26 1
DF 24 -1
R² 0.357
Adjusted R² 0.330
29
MSE 506.063
RMSE 22.496

:Analysis of variance (PULL OUT LOAD (F) (UR)):

Sum of Mean p-values
Source DF F Pr > F
squares squares signification codes
Model 1.000 6746.297 6746.297 13.331 0.001 **
Error 24.000 12145.522 506.063
Corrected Total 25.000 18891.819        
Computed against model Y=Mean(Y) Signification codes: 0 < *** < 0.001 < ** < 0.01 < * < 0.05 < . < 0.1 < ° < 1

:Model parameters (PULL OUT LOAD (F) (UR)):

Lower Upper p-values
Standar
Source Value t Pr > |t| bound bound signification
d error
(95%) (95%) codes
10.42
Intercept 89.507 8.583 8 <0.0001 71.792 107.221 ***
(A)*(θ/90)*(H/B)*(BF) 205.33
(α/N) 3 56.238 3.651 0.001 89.264 321.403 **
Signification codes: 0 < *** < 0.001 < ** < 0.01 < * < 0.05 < . < 0.1 < ° < 1

:Equation of the model (PULL OUT LOAD (F) (UR)):

Pull Out Load (in Corresponding to 5mm amplitude) = 89.5+205.3*(A)*(θ/90)*(H/B)*(BF)(α/N)

:Standardized coefficients (PULL OUT LOAD (F) (UR)):

Lower Upper p-values
Standar
Source Value t Pr > |t| bound bound signification
d error
(95%) (95%) codes
(A)*(θ/90)*(H/B)*(BF)
(α/N) 0.164 3.651 0.001 0.260 0.935 *** 0.164
Signification codes: 0 < *** < 0.001 < ** < 0.01 < * < 0.05 < . < 0.1 < ° < 1

30
Pull Out Load (in Corresponding to 5mm amplitude)
Regression of Pull Out Load (in Corresponding to 5mm amplitude) by (A)*(θ/
90)*(H/B)*(BF)(α/N) (R²=0.357)
250

200

150

100

50

0
-0.1 0 0.1 0.2 0.3 0.4 0.5

(A)*(θ/90)*(H/B)*(BF)(α/N)

Training set Validation set

Model Conf. interval (Mean 95%)
Conf. interval (Obs 95%)

Standardized residuals / (A)*(θ/90)*(H/B)*(BF)(α/N)

2

1.5
Standardized residuals

0.5

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
-0.5

-1

-1.5

(A)*(θ/90)*(H/B)*(BF)(α/N)

Training set Validation set

31
 Pull Out Load (in Corresponding to 5mm amplitude) / Standardized residuals Pred(Pull Out Load (in Corresponding to 5mm amplitude)) / Standardized residuals
2 2

1.5 1.5

1 1

0.5 0.5

Standardized residuals
Standardized residuals

0 0
0 20 40 60 80 100 120 140 160 180 200 0 20 40 60 80 100 120 140 160 180 200
-0.5 -0.5

-1 -1

-1.5 -1.5

Pull Out Load (in Corresponding to 5mm amplitude) Pred(Pull Out Load (in Corresponding to 5mm amplitude))

Training set Validation set Training set Validation set

Pull Out Load (in Corresponding to 5mm amplitude)

Pred(Pull Out Load (in Corresponding to 5mm amplitude)) - Pull Out Load (in Cor-
responding to 5mm amplitude)
300

250

200

150

100

50

0
0 50 100 150 200 250 300

Pred(Pull Out Load (in Corresponding to 5mm amplitude))

Training set Validation set

Standardized residuals / Pull Out Load (in Corresponding to 5mm amplitude) (Train- Standardized residuals / Pull Out Load (in Corresponding to 5mm amplitude) (Valida-
ing set) tion set)
Obs29
Obs28
Obs27
Obs26
Obs25
Obs23
Obs22
Obs21
Obs20
Obs19
Obs18
Obs17
Obs16
Obs15 Obs24
Observations

Observations

Obs14
Obs13
Obs12
Obs11
Obs10
Obs9
Obs8
Obs7
Obs6
Obs5
Obs4
Obs3
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2 -1.2 -0.7 -0.2 0.3 0.8

Standardized residuals Standardized residuals

32
 STATISTICAL ANALYSIS AND FORMULATION OF PULLOUT LOAD (UNREINFORCED)
FOR SQUARE PLATE SIZE (AMPLITUDE=2MM FREQUENCY= 0.5 Hz)

:Summary statistics:
Obs.
Obs. with Std.
Observation without Minimu Maximu
Variable missing Mean deviatio
s missing m m
data n
data
41.66
PULL OUT LOAD (F) 26 0 26 26.100 59.990 0 9.594
(A)*(θ/90)*(H/B)*(BF)
(α/N) 26 0 26 0.002 0.022 0.008 0.005
Summary statistics (Quantitative data / Validation):
Obs.
Obs. with Std.
Observation without Minimu Maximu
Variable missing Mean deviatio
s missing m m
data n
data
31.24
PULL OUT LOAD (F) 1 0 1 31.240 31.240 0  
(A)*(θ/90)*(H/B)*(BF)
(α/N) 1 0 1 0.004 0.004 0.004  

Correlation matrix:
  (A)*(θ/90)*(H/B)*(BF)(α/N) PULL OUT LOAD (F)
(A)*(θ/90)*(H/B)*(BF)(α/N) 1 0.714
PULL OUT LOAD (F) 0.714 1

Regression of variable PULL OUT LOAD (F) (UR):

Goodness of fit statistics (PULL OUT LOAD (F) (UR)):
Statistics Training set Validation set
Observations 26 1
Sum of weights 26 1
DF 24 -1
R² 0.510
Adjusted R² 0.490
MSE 46.979
RMSE 6.854

:Analysis of variance (PULL OUT LOAD (F) (UR)):

p-values
Sum of Mean
Source DF F Pr > F signification
squares squares
codes
Model 1.000 1173.634 1173.634 24.982 <0.0001 ***
Error 24.000 1127.500 46.979
Corrected Total 25.000 2301.134        
Computed against model Y=Mean(Y) Signification codes: 0 < *** < 0.001 < ** < 0.01 < * < 0.05 < . < 0.1 < ° < 1

:Model parameters (PULL OUT LOAD (F) (UR)):

33
 Lower Upper p-values
Standard
Source Value t Pr > |t| bound bound signification
error
(95%) (95%) codes
12.64
Intercept 31.282 2.473 <0.0001 26.177 36.387 7 ***
(A)*(θ/90)*(H/B)*(BF) 1305.79 1844.99
(α/N) 6 261.253 4.998 <0.0001 766.596 5 ***
Signification codes: 0 < *** < 0.001 < ** < 0.01 < * < 0.05 < . < 0.1 < ° < 1
:Equation of the model (PULL OUT LOAD (F) (R)):
Pull Out Load (in Corresponding to 2mm amplitude) = 31.2+1305.3*(A)*(θ/90)*(H/B)*(BF)(α/N)

Standardized coefficients (PULL OUT LOAD (F) (UR)):

Upper p-values
Valu Standar Lower bound
Source t Pr > |t| bound signification
e d error (95%)
(95%) codes
(A)*(θ/90)*(H/B)*(BF)
(α/N) 0.714 0.143 4.998 <0.0001 0.419 1.009 ***
Signification codes: 0 < *** < 0.001 < ** < 0.01 < * < 0.05 < . < 0.1 < ° < 1
Pull Out Load (in Corresponding to 2mm amplitude)

Regression of Pull Out Load (in Corresponding to 2mm amplitude) by (A)*(θ/

90)*(H/B)*(BF)(α/N) (R²=0.510)
80

70

60

50

40

30

20

10

0
-0.005 0 0.005 0.01 0.015 0.02 0.025

(A)*(θ/90)*(H/B)*(BF)(α/N)

Training set Validation set

Model Conf. interval (Mean 95%)
Conf. interval (Obs 95%)

34
 Standardized residuals / (A)*(θ/90)*(H/B)*(BF)(α/N)
1.5

1
Standardized residuals

0.5

0
0 0.005 0.01 0.015 0.02 0.025
-0.5

-1

-1.5

-2

(A)*(θ/90)*(H/B)*(BF)(α/N)

Training set Validation set

Pull Out Load (in Corresponding to 2mm amplitude) / Standardized residuals Pred(Pull Out Load (in Corresponding to 2mm amplitude)) / Standardized residuals
1.5 1.5

1 1

0.5 0.5

0 0
Standardized residuals

0 10 20 30 40 50 60 0 10 20 30 40 50 60
Standardiz ed residuals

-0.5 -0.5

-1 -1

-1.5 -1.5

-2 -2

Pull Out Load (in Corresponding to 2mm amplitude) Pred(Pull Out Load (in Corresponding to 2mm amplitude))

Training set Validation set Training set Validation set

35
 Pull Out Load (in Corresponding to 2mm amplitude)
Pred(Pull Out Load (in Corresponding to 2mm amplitude)) - Pull Out Load (in Cor-
responding to 2mm amplitude)
100

90

80

70

60

50

40

30

20

10

0
0 10 20 30 40 50 60 70 80 90 100

Pred(Pull Out Load (in Corresponding to 2mm amplitude))

Training set Validation set

Standardized residuals / Pull Out Load (in Corresponding to 2mm amplitude) (Train- Standardized residuals / Pull Out Load (in Corresponding to 2mm amplitude) (Valida-
ing set) tion set)
Obs29
Obs28
Obs27
Obs26
Obs25
Obs24
Obs23
Obs22
Obs21
Obs20
Obs19
Obs18
Obs17 Obs4
Obs16
Observations
Observations

Obs15
Obs14
Obs13
Obs12
Obs11
Obs10
Obs9
Obs8
Obs7
Obs6
Obs5
Obs3
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8

Standardized residuals Standardized residuals

STATISTICAL ANALYSIS AND FORMULATION OF PULLOUT LOAD(UNREINFORCED)

FOR SQUARE PLATE SIZE (AMPLITUDE 5MM, FREQUENCY =0.5 Hz)
Summary statistics:
Obs.
Obs. with
Observation without Std.
Variable missing Minimum Maximum Mean
s missing deviation
data
data
102.90
PULL OUT LOAD (F) 26 0 26 65.250 149.980 6 23.278
(A)*(θ/90)*(H/B)*(BF)
(α/N) 26 0 26 0.010 0.135 0.049 0.030

Summary statistics (Quantitative data / Validation):

Obs.
Obs. with
Observation without Std.
Variable missing Minimum Maximum Mean
s missing deviation
data
data
PULL OUT LOAD (F) 1 0 1 78.120 78.120 78.120  
(A)*(θ/90)*(H/B)*(BF) 1 0 1 0.025 0.025 0.025  
36
(α/N)

Correlation matrix:
  (A)*(θ/90)*(H/B)*(BF)(α/N) PULL OUT LOAD (F)
(A)*(θ/90)*(H/B)*(BF)(α/N) 1 0.656
PULL OUT LOAD (F) 0.656 1

Regression of variable PULL OUT LOAD (F) (UR):

Goodness of fit statistics (PULL OUT LOAD (F) (UR)):
Statistic Training set Validation set
Observations 26 1
Sum of weights 26 1
DF 24 -1
R² 0.430
Adjusted R² 0.406
MSE 321.712
RMSE 17.936

Analysis of variance (PULL OUT LOAD (F) (UR)):

Sum of p-values
Source DF Mean squares F Pr > F
squares signification codes
Model 1.000 5825.568 5825.568 18.108 0.000 ***
Error 24.000 7721.090 321.712
Corrected Total 25.000 13546.657        
Computed against model Y=Mean(Y) Signification codes: 0 < *** < 0.001 < ** < 0.01 < * < 0.05 < . < 0.1 < ° < 1

Model parameters (PULL OUT LOAD (F) (UR)):

Lower Upper
Standar p-values
Source Value t Pr > |t| bound bound
d error signification codes
(95%) (95%)
11.15
Intercept 77.431 6.944 1 <0.0001 63.100 91.762 ***
(A)*(θ/90)*(H/B)*(BF) 514.70
(α/N) 2 120.954 4.255 0.000 265.065 764.340 ***
Signification codes: 0 < *** < 0.001 < ** < 0.01 < * < 0.05 < . < 0.1 < ° < 1
Equation of the model (PULL OUT LOAD (F) (R)):
ut Load (in Corresponding to 5mm amplitude) = 77+514*(A)*(θ/90)*(H/B)*(BF)(α/N)

Standardized coefficients (PULL OUT LOAD (F) (UR)):

Lower Upper
Standar p-values
Source Value t Pr > |t| bound bound
d error signification codes
(95%) (95%)
(A)*(θ/90)*(H/B)*(BF)
(α/N) 0.656 0.154 4.255 0.000 0.338 0.974 ***
Signification codes: 0 < *** < 0.001 < ** < 0.01 < * < 0.05 < . < 0.1 < ° < 1

37
 Regression of Pull Out Load (in Corresponding to 5mm amplitude) by (A)*(θ/
Pull Out Load (in Corresponding to 5mm amplitude)
90)*(H/B)*(BF)(α/N) (R²=0.430)
200

180

160

140

120

100

80

60

40

20

0
-0.02 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16

(A)*(θ/90)*(H/B)*(BF)(α/N)

Standardized residuals / (A)*(θ/90)*(H/B)*(BF)(α/N)

1.5

1
Standardized residuals

0.5

0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
-0.5

-1

-1.5

-2

(A)*(θ/90)*(H/B)*(BF)(α/N)

Training set Validation set

38
 Pull Out Load (in Corresponding to 5mm amplitude) / Standardized residuals Pred(Pull Out Load (in Corresponding to 5mm amplitude)) / Standardized residuals
1.5 1.5

1 1

0.5 0.5

0 0

Standardized residuals
0 20 40 60 80 100 120 140 160
Standardized residuals

0 20 40 60 80 100 120 140 160

-0.5 -0.5

-1 -1

-1.5 -1.5

-2 -2

Pull Out Load (in Corresponding to 5mm amplitude) Pred(Pull Out Load (in Corresponding to 5mm amplitude))

Training set Validation set Training set Validation set

Standardized residuals / Pull Out Load (in Corresponding to 5mm amplitude) (Train- Standardized residuals / Pull Out Load (in Corresponding to 5mm amplitude) (Valida-
ing set) tion set)
Obs29
Obs28
Obs27
Obs26
Obs25
Obs24
Obs23
Obs22
Obs21
Obs20
Obs19
Obs18
Obs17 Obs4
Obs16
Observations
Observations

Obs15
Obs14
Obs13
Obs12
Obs11
Obs10
Obs9
Obs8
Obs7
Obs6
Obs5
Obs3
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8

Standardized residuals Standardized residuals

Pred(Pull Out Load (in Corresponding to 5mm amplitude)) - Pull Out Load (in Cor -
Pull Out Load (in Corresponding to 5mm amplitude)

responding to 5mm amplitude)

250

200

150

100

50

0
0 50 100 150 200 250

Pred(Pull Out Load (in Corresponding to 5mm amplitude))

39
 SUMMARY,CONCLUDING REMARKS
AND FUTURE SCOPE OF STUDY
Summary:-

The aim of the present study is to statistically analyse the behaviour of square plate anchor in soft clay soil,
the model tests have been performed with three different sizes of plate of 25mm, 50mm and 75mm at three
different embedment ratios of 1, 2 and 3 at three different inclined angles of 30, 45 and 60 degree with
horizon for unreinforced soil condition. The numerical models also have been run according to its model
anchor specification. A thin geotextile layer, having width of 4 times the size of anchor plate is placed at
distance of 0.25times of the embedment depth form anchor plate parallel to the plate to study the pullout
capacity in unreinforced soft clay soil. Validation of the numerical model has been carried out with respect
to the studies from a paper by Das and Puri (1989).
 FINAL FORMULATION OF PULLOUT LOAD FOR UNREINFORCED CONDITION
UNDER CYCLIC LOADING:-
F(UR) = 532*(A)*(θ/90)*(H/B)*(BF)(α/N) +35 (2mm-0.2Hz)
= 205*(A)*(θ/90)*(H/B)*(BF)(α/N) +89.5 (5mm-0.2Hz)
= 1305*(A)*(θ/90)*(H/B)*(BF)(α/N) +31 (2mm-0.5Hz)
= 514*(A)*(θ/90)*(H/B)*(BF)(α/N) + 77 (5mm-0.5Hz)

Concluding remarks:-
The following concluding remarks may be drawn from the present study:-
1) The prediction of pullout load for unreinforced cases of square plate anchors when the other
parameters are varying can be predicted with sufficient accuracy.

2) The formulation is valid for inclined square plate anchors and for vertical square plate anchors a
extra factor has to be considered which is left for future study as more intermediate data is necessary.

FUTURE SCOPE OF WORK:-

The further research in this area may be carried out in the following directions:
a)The relation is valid for square footing unreinforced clay under cyclic loading which can be extended to
strip and rectangular footing.
b) A relation can be also be drawn for square footing under cyclic loading which can be extended to strip
and rectangular footing for reinforced pullout load and breakout factor.

40
REFERENCES :-
 Ardebili A, Z., Gabr, M. A., & Rahman, M. S. (2016). “Uplift Capacity of Plate Anchors in
Saturated Clays: Analyses with Different Constitutive Models”. ASCE International
Journal of Geomechanics, Volume 16, No 2

 Balla, A. (1961). “The resistance to breaking out of mushroom foundations for pylons”.
Proc. 5th Intl. Conf. on Soil Mechanics and Foundation Engineering, Paris, France: pp.569-
576.

 Beirne,O., C., Loughlin,O., and Gaudin, C. (2017). “Assessing the penetration resistance
acting on a dynamically installed anchor in normally consolidated and overconsolidated
clay”. Canadian Geotechnical Journal, Volume 54 No-(1), pp 1–17.

 Bhattacharya, P., Debjit, B., Mukherjee, S.P., Chattopadhyay, B.C., (2008). “Pullout
Behaviour of Square Anchors in Reinforced Clay’’, The 12th International Conference of
International Association for Computer Methods and Advances in Geomechanics
(IACMAG), Goa, India, pp.3441-3447.

 Bhattacharya, P., Kumar, J. (2016). “Uplift Capacity of Anchors in Layered Sand Using
Finite-Element Limit Analysis: Formulation and Results”. International Journal of
Geomechanics / Volume 16 Issue 3 - June 2016

 Biradar, J., Banerjee, S., Shankar, R., Ghosh, P., Mukherjee, S., Fatahi, B. (2019)
“Response of square anchor plates embedded in reinforced soft clay subjected to cyclic
loading”. Geomechanics and Engineering, Volume 17, Issue 2, Pages.165-173

 Chattopadhyay, B.C. and Pise, P.J., (1986). “Breakout resistance of horizontal anchors in
sand, Soils and Foundations”, Japanese Society of Soil Mechanics and Foundation
Engineering, Volume 26, No.4, pp.16-22

 Das, B. M. (1980). “A procedure for estimation of ultimate uplift capacity of foundations

in clay”. Soils and Foundations, 20 (1): 77-82.

 Das, B. M., Puri, V. K. (1989).“ Holding capacity of Inclined Square plate anchors in Clay,
Soils and Foundations”, Japanese Society of Soil Mechanics and Foundation Engineering,
Volume 29, No.3, pp.138-144.

 Das, B. M., Tarquin A. J., and Moreno R. (1985) “Model tests for pullout resistance of
vertical anchors in clay.” Civil Eng. for Practicing and Design Engineers., Pergamon Press,
New York, 4(2):191-209.

 Das, T. K., Chattopadhyay, B.C., Roy, S., (2013). “Pull-out Capacity of Plate Anchors with
Coaxial Geotextile Reinforcement”, Annals of Pure and Applied Mathematics, Vol. 5,
No.1, pp.53-63.

 Ghaly. A, Hanna. A, Hanna. M, (1991), “Uplift Behavior of Screw Anchors in Sand”. I: Dry
Sand, International Journal of Rock Mechanics & Mining Sciences & Geomechanics

41
 Abstracts 117(5).

 Ilamparuthi, K., Dickin, E.A., Muthukrisnaiah, (2002).“Experimental investigation of the

uplift behaviour of circular plate anchors embedded in sand”, Canadian Geotechnical
Journal 39, pp.648-664.

 Datta, M. and Sing, P.K. (1984). “Anchors for offshore structure – geotechnical aspects”.
Indian Geotechnical Journal, Vol.-14, 255-281.

 Krishnaswamy, N.R. and Parashar, S.P., (1994). “Uplift behavior of plate anchors with
geosynthetics, Geotextiles and Geomembranes 13”, pp.67-89.

 Kumar, J., and Khatri VN (2009) “Bearing capacity factor Nc under ϕ=0 condition for piles
in clays”. Numer and Analytical Methods Geomech Volume33, Issue9 25 June 2009,
Pages 1203-1225.

 Chottopadhyay, B.C., Khatun, S. (2010) “Uplift Capacity of Plate Anchors with

Reinforcement”. Indian Geotechnical Conference – 2010, GEOtrendz December 16–18,
2010

 Majumder, A., Roy, R., Banerjee. S., Mukherjee, S., Biswas, S., (2019) “Pullout behaviour
of plate anchors in geotextile reinforced soft clay”.

 Makarchian, M., Gheitasi, M., Badkhasan, E.,(2012), “Experimental & Numerical Study
of Uplift Behavior of Anchorsembedded in Reinforced Sand”, 5th Asian Regional Conference
on Geosynthetics, Geosynthetics Asia.

 Merifield, R.S., Sloan, S.W., Yu, H.S., (2001).”Stability of plate anchors in undrained clay”,
Geotechnique 51, No.2, pp.141-153

 Susovan Tunga (2020), “PULLOUT BEHAVIOR OF INCLINED SQUARE PLATE ANCHORS

IN GEOTEXTILE REINFORCED SOFT CLAY”

42