Journal of Petroleum Science and Engineering 128 (2015) 4452

Contents lists available at ScienceDirect

Journal of Petroleum Science and Engineering

journal homepage: www.elsevier.com/locate/petrol

A simple and accurate numeric solution procedure for nonlinear

buckling model of drill string with frictional effect
Youhong Sun a, Yongping Yu a,b,n, Baochang Liu a
a
b

College of Construction Engineering, Jilin University, Changchun 130026, PR China

Department of Mechanical Engineering, University of California, Berkeley, CA 94720, USA

art ic l e i nf o

a b s t r a c t

Article history:
Received 9 April 2014
Accepted 17 February 2015
Available online 25 February 2015

The nonlinear buckling analysis of drill string in a rigid well is presented in this paper. Considering
effects of friction and boundary constraints, this problem could be taken as a model of a rod laterally
constrained in a rigid cylinder (horizontal, oblique and vertical rigid cylinder could be included). After
introducing a new variable, the resulting coupled nonlinear integraldifferential equations are successfully solved by employing the extended system shooting method. Examples with various friction
coefcients and combinations of boundary conditions are proposed. It is found that the axial frictional
force plays a more signicant part on buckling load for horizontal well than vertical one. Compared to
experimental data or results obtained by using the discrete singular convolution algorithm (DSC) and the
nite element method (FEM), the accuracy of the formulations and solution procedures, is veried.
Whats more, the nonlinear buckling behaviors of two instances of vertical scientic wells are analyzed.
The present results are useful for practical design applications related to calculation of buckling loads
and selection of bottom hole assembly (BHA) elements.
& 2015 Elsevier B.V. All rights reserved.

Keywords:
Buckling
Horizontal well
Friction
Shooting method
Drill string

1. Introduction
This study of the structural buckling behavior of drill pipes is
motivated by interest in many aspects of petroleum engineering
(such as the functions of pipes, safety and surveying accuracy of
down-hole instruments, etc.). Because of the high frequency of
drill string failure, drill string lock-up, and casing wear, the
stability of drill string has been a serious problem in oil/gas eld
operations for many years (Gulyayev et al., 2009; Tan and Gan,
2009). On the other hand, with the development of drilling
technology, oil/gas and deep continental scientic drilling wells
become very long currently, even more than ten kilometers.
Furthermore, some oil/gas and deep continental scientic drilling
wells may have very complex geometrical congurations, such
that parts of wells may be inclined, vertical, horizontal, just plane
curved, and even 3-D curved. Therefore, it is important and
meaningful to investigate the buckling behavior of drill string for
the science and technologies in petroleum engineering, deep
continental scientic drilling and other related elds.

n
Corresponding author at: College of Construction Engineering, Jilin University,
Changchun 130026, PR China.
E-mail addresses: yuyongping@jlu.edu.cn, yuypjlu@berkeley.edu (Y. Yu).

http://dx.doi.org/10.1016/j.petrol.2015.02.026
0920-4105/& 2015 Elsevier B.V. All rights reserved.

In despite of the complexity of the problem, many results are

still reported. Paslay and Bogy (1964) rst studied the problem of
sinusoidal buckling of the tube. Based on the principle of minimum potential energy, the problem of helical buckling of a vertical
tube was rst analyzed by Lubinski et al. (1962). Since then,
Cheatham and Pattillo (1984), He and Kyllingstad (1993), Miska
and Cunha (1995) have studied helical buckling of tubes in vertical,
horizontal or inclined wellbores, based on the energy method.
Experimental study of helical buckling of a horizontal rod in a tube
was performed by McCann and Suryanarayana (1994). While
Wicks et al. (2008) reviewed available analytical and experimental
results on the structural behavior of constrained horizontal cylinders subjected to axial compression, torsion, and gravity.
Frictional interactions with the constraining wellbore could
cause buckling lock-up of drill strings, in which the drill strings are
unable to progress, especially in large displacement horizontal
wells. Understanding the post-buckling behavior of drill string,
therefore, is important to avoid lock-up condition. Mitchell (1986),
Wu and Juvkam-Wold (1993), Gao and Miska (2009) have took the
friction effects into consideration in the theoretical analysis, while
McCann and Suryanarayana (1994) study the frictional effects on
buckling behavior by experimental method. Recently, Gan et al.
(2009) use the differential quadrature element method to investigate effect of the gravitational and friction loads on buckling
behavior of drill string, in order to overcome the shortage of the

Y. Sun et al. / Journal of Petroleum Science and Engineering 128 (2015) 4452

local methods (such as the nite element method), and the global
method (such as the differential quadrature method). By using the
discrete singular convolution (DSC) algorithm, Wang and Yuan
(2012) investigate the effects of friction and boundary constraints
on the nonlinear buckling behavior of a relatively short rod
constrained in a rigid horizontal cylinder and subjected to axial
compression, gravitational and frictional loads. A modied version
of DSC-based iterative scheme is presented to solve the coupled
nonlinear integraldifferential equations. They conclude that effects of friction on the buckling behavior are strongly depended on
the boundary conditions for short drill pipes.
The purpose of this paper is to investigate the nonlinear buckling analysis of drill string in a rigid well by using the extended
system shooting method. The accuracy of the formulations and
solution procedures is veried, compared to experimental data or
results obtained by using the DSC and FEM. Through analyzing the
buckling deformation of drill strings in two specic wells, we nd
that the axial frictional force plays a less signicant part on
buckling loads for vertical well than horizontal one. The present
results are useful for practical design applications related to
calculation of buckling loads and selection of bottom hole assembly (BHA) elements.

45

drag friction force is taken into considerations, here W n is the contact

force per unit length, and f W n is the drag friction force per unit
length acting opposite to the loading direction. Note that initially the
rod laterally contacts with the cylinder to simplify the problem,
although real boundary conditions would involve rod centralized in
packers. A right-handed Cartesian coordinate system is set in Fig. 1.
The drill string is assumed inextensible, which always contacts with
the outer well during deformation. The buckling governing equation
of a drill string with a wellbore can be expressed as in the following
form (Mitchell, 1988; Gao et al., 2002; Wicks et al., 2008; Wang and
Yuan, 2012)
 2 2
4
2
d
d
d
d d q  sin sin
EI 4 F 2 F 0  6EI
0
1

ds
ds ds2
r
ds
ds

W n  Fr

d
ds

2


 q  cos sin EIr

d
ds

4
0

where is the axial force (Wicks et al., 2008),

Z s
W n ds
F P qs cos 
0

The different boundary conditions are as follows:

0 L 0 L 0

S S

2. Nonlinear buckling equations

0 L 0 0 L 0

SC

For the title problem, a model of a drill string laterally constrained

by a horizontal or vertical rigid well is schematically displayed in
Fig. 1. Assume that the rod can rotate freely with respect to its axis,
thus only axial friction force is encountered (Wicks et al., 2008). The
rod is subjected to a compressive force P at its left end and a resulting
RL
compressive force F b P qL cos  0 W n ds at its right end if the

0 L 0 0 L 0

C S

0 L 0 0 0 L 0

CC

where is the deviation angle of the drill string axis in the X; Y

plane, and r is the radial clearance between the pipe and the
wellbore. The material of the rod is isotropic with an elasticity
modulus of E. The length, weight per unit length and bending
rigidity of the rod are L, q and EI, respectively. The path of the
deformed drill string centerline is s A 0; L, and is the frictional
coefcient. The inclined angle of the drill string is ; the horizontal
well is corresponding to 901, while the vertical one is 01.
The helix angle shown in Fig. 1 is small, therefore, could be
approximated by
 sin r

d
ds

For details of the model derivation, readers are referred to

Wang and Yuan (2012).
The path of the deformed drill pipe centerline could be
calculated by the following formulae, once are obtained, (Yuan
and Wang, 2012)
9
8
8 9 8
9
 r d
sin >
ds
>
>
>
Z s>
=
<
<u >
= >
< r cos >
=
cos
r d
v
ds
9
r sin

ds
>
>
>
>
>
>
0 >
: ; :
;
;
: 1  1r 2 d2 >
w
0
s0
2

ds

where u; v; andw are components of the position vector of the

deformed drill string centerline. For simply expressing, Eqs. (1), (2)
(4)(7) and (9) can be expressed in the following dimensionless
form:
!
Z S
4 qS cos 
Nd q cos  N 0  60 2
0

Q sin sin 0
Z
N  qS cos 
Fig. 1. Sketch of a drill string in a rigid well (including horizontal and vertical
wells).

10
S

!
Nd 0 2 0 4  Q cos sin 0

0 2 0 2 0

S S

11
12

46

Y. Sun et al. / Journal of Petroleum Science and Engineering 128 (2015) 4452

0 2 0 0 2 0

SC

13

0 2 0 0 2 0

CS

14

0 2 0 0 0 2 0

CC

15

9
8 9 8 cos
>
>
>
>
=
<
<u >
= >
v
Rsinh
i
>
> S
>
> >
2
2
: ;
;
: 0 1  12J 0 dJ >
w

16

where
S
N

2s
;
L

W n L4
4

2 EIr

2r
;
L

; q

qL3
2 3 EI

P
PL2
2 ;
P cr
4 EI

u
;u ;
r

v
v ;
r

qL4
2 4 EIr

w
;
r

2r
L

17

Without the friction effect, the above governing differential

equations are the same as the ones in Mitchell (1988). Due to the
friction effect, Eqs. (10) and (11) coupled each other, are the
nonlinear integraldifferential equations. The exact solutions of
Eqs. (10) and (11), therefore, are difcultly obtained.

It is well known that ordinary differential equation (ODE)

boundary-value problems (BVPs) can be solved by shooting
methods (Seydel, 1994; Yu, et al., 2012). Shooting methods
integrate initial-value problems in a systematic way in order to
nd the initial values of the particular trajectory that satisfy the
boundary conditions. The initial and nal values of an arbitrary
trajectory of the differential equation do not generally satisfy the
RS
boundary conditions. For the sake of the integral term 0 Nd, the
classical shooting method could not directly be employed to
obtained the solution. So a new variable is rst introduced to
the system:
Z S

Nd:
18
0

Eqs. (10) and (11), then, are transformed into a system of

differential equations with the corresponding transformed boundary conditions, without the integral term.

0 2

qS cos  q cos   6 Q sin sin 0

0  qS cos  0 2 0 4  Q cos sin 0

u0  0 sin
v0 0 cos
1 1
w0  J 0 2
J 2
0 0;

0 0 ;

u0 1;
0 0;

u0 1;
0 0;

u0 1;
0 0;

u0 1;

0 0;

"

2 ; 2

00 2 ; 00 2

26

here and hereafter }mk  1 ; } is omitted for variables evaluated at

m; mk  1 ; .
The elements of Eq. (26) can be calculated
as follows. For
 
n
dn
convenience, we denote
,
etc.
Differentiating
;

n

dS
Eqs. (19) and (20) with respect to ; and , respectively, the
following IVPs for derivatives s; s, s; s could be
obtained:
h
i
2
4 qS cos   60 00
q cos  120 00  0 0 Q cos sin 

0
 0  0
h
i
4 qS cos   60 2 00 q cos  120 00  0 0

0
Q cos sin   0  0
h
i
0  20 0 qS cos   20 2
0 2 Q sin sin  0
h
i
0  20 0 qS cos  20 2
2

0 ; 0 v0 w0 0;

0 0;

0 ;

22

0 ; 0 v0 w0 0;

C  C2 2 0

00 0 1;

000
0 0 0

27

Solving Eqs. (19), (20) and (27) simultaneously, the elements

needed in Eq. (26) can be got. By using Eq. (25), the iteration could
be implemented and then the solution to buckling behavior of the
drill string could be obtained in terms of (i.e. P), for given values
of dimensionless parameters , q, Q , , and J.

21

0 ; 0 v0 w0 0;

C  S2 2 0
0 ;

20

0 ; 0 v0 w0 0;

S  C2 2 0

0 0 0;

where

25

0 Q sin sin  0
0 0; 0 0 1; 00 0 000
0 0 0
19

0 0 0;

mk mk  1  Fm mk  1 ;   1 Fmk  1 ; ;

0 0 0 0;

S  S2 2 0

0 0 ;

Eq. (24) is a system of nonlinear equations. A combination of

Newtons method with an integrator serves as the simplest
shooting device. For each , m0 is assumed to be the initial
iteration vector, mk in the kth step can be calculated from that
of the k  1th step using

Fm mk  1 ;

3. Solution method

advised to Yu et al., 2012) would be used to solve Eqs. (19)(23).

Here, the solving procedures of the extended system shooting
method are presented only to Eqs. (19) and (20), the system
including the SS boundary condition, as follows.
The trajectory that starts from an initial vector m ; T , and
, is denoted by S; m; ; 0 r S r 2. Note that the dimensionless
parameters , q, Q , , and J are constants when real physical and
geometric parameters of the system are given. The vector m in
terms of would be solved, such that the following boundary
conditions at S 2 are satised:
!  
2; m;
0
Fm;

:
24
2; m;
0

23

Note that solutions of the system are all depending on the

dimensionless parameter (i.e. P). The extended system shooting
method (for details of the procedures of this method, readers are

4. Results and discussion

4.1. Example I
To verify the derivations of formulae and the solution procedures, some present results are compared to the data available in
literature (McCann and Suryanarayana, 1994; Huang and Pattillo,
2000) or recalculated by using the FEM and the DSC method
(Wang and Yuan, 2012). For comparisons, the geometrical and
material parameters are =2, r 0:015875 m, L 3:683 m,

Y. Sun et al. / Journal of Petroleum Science and Engineering 128 (2015) 4452

q 0:86828 N=m,EI 5:9966 Nm2 (Huang and Pattillo, 2000;

McCann and Suryanarayana, 1994). Four combinations of boundary
conditions, namely, SS, SC, CS, and CC, are considered. The
coefcient of friction varies from 0 to 0:6.
The numerical solutions of critical linear buckling pressure
obtained by the present extended system shooting method P LY
cr
and the DSC one P LW
(Wang and Yuan, 2012) are displaced in Fig. 2
cr
for various friction coefcients and four different boundary conditions (i.e. SS, SC, CS, and CC). From Fig. 2, it is clearly seen that
numerical solutions of critical linear buckling pressure obtained by
the present extended system shooting method P LY
and the DSC one
cr
P LW
(Wang and Yuan, 2012) are very close to each other. Note that
cr
the present numerical result P Lcr 36:245 for the SS drill string and
0, is exactly the same as the theoretical datum (Dawson and
Paslay, 1984), both of them are close to the experimental buckling
load P Lcr 35:72 determined under unloading (McCann and
Suryanarayana, 1994). The present buckling critical load for CC drill
pipe is close to the experimental buckling load P Lcr 53:02 obtained
under loading (McCann and Suryanarayana, 1994).
The comparisons of Y obtained by the present extended
system shooting method, W by the DSC (Wang and Yuan, 2012)
and F by FEM for applied load of P 70 N and P 85 N, 0:5
and SS condition are shown in Fig. 3. As seen in the graph, the
present results agree very well with the DSC and FEM data.
HY
The numerical results of critical helix buckling pressure P cr
obtained by the present extended system shooting method and
P HW
DSC one are displayed in Fig. 4. It is clearly seen that the
cr
present data are very close to that of DSC obtained by Wang and
Yuan (2012). Both the present numerical result P HY
85:7715
cr
and DSC data P HW
85:77 for the SS drill string and 0, are
cr
closest to the experimental buckling load P HM
84:52 detercr
mined under loading (McCann and Suryanarayana, 1994), compared with P HC
79:21 from Chen et al. (1990), and P HH
71:60
cr
cr
obtained by using the RayleighRitz technique (Huang andPattillo,
2000). It is different from the linear buckling case, all results from
different methods agree with the experimental one obtained
during loading (McCann and Suryanarayana, 1994).
The numeric results of , the contact force per unit length W n ,
and the frictional force f obtained by the present extended system
shooting method, for applied load of P 80 N, 0:4 and all four
boundary conditions are shown in Figs. 57, respectively. As seen in
the graphs, the effect of boundary condition on the buckling

47

Fig. 3. Comparisons of obtained by the present method, the DSC and FEM.

Fig. 4. Comparisons of critical helix buckling loads P H

cr obtained by the present and
the DSC method for various friction coefcients.

Fig. 5. Comparisons of for different boundary conditions (P 80 N, 0:4).

Fig. 2. Comparisons of critical linear buckling loads P Lcr obtained by the present and
the DSC method for various friction coefcients.

deformation of drill string is very obvious. Furthermore, for the

existence of the frictional force, the symmetric feature of both drill
string buckling deformation and the contact force per unit length W n
is broken in Figs. 5 and 6, compared with the absence of friction.
The 3-D buckling congurations of drill string with four boundary
conditions (i.e. SS, SC, CS, CC) at the load P 80 N and 0:4
are shown in Fig. 8(ad), respectively. As could be seen in Fig. 8(ad),
different units, i.e., centimeters in the u- and v-axis and meters in the
w-axis, are used for clarity. Note that the deformed drill string axis is

48

Y. Sun et al. / Journal of Petroleum Science and Engineering 128 (2015) 4452

obtained by using Eq. (9) in Fig. 8(ad). When simple support is

applied, the buckling deformation of the drill string is very large; while
the clamped condition is supposed, the buckling behavior of the drill

Fig. 6. Comparisons of W n for different boundary conditions (P 80 N, 0:4).

Fig. 9. The designing sketch of the SW well depth structure.

Table 1
The parameters of the BHA of the SW well.
Element name of the BHA

Drill bit 150=70

Core barrel LSC133/71
Heavy weight drill pipe
5:5  3:5=8
Heavy weight drill pipe 5  3
Heavy weight drill pipe
5:5  3:5=8
Fig. 7. Comparisons of f for different boundary conditions (P 80 N, 0:4).

Weight Length Out

diameter
L=m
M=kg
d=mm

=mm

30
1020
3100

0.5
9.5
40.0

150
133.0
139.7

40
31
23.8

750
3100

10
40.0

127
139.7

25.4
23.8

Note that the weight of the BHA is 8t, the length is 100 m.

Fig. 8. Buckling congurations of drill string at P 80 N, 0:4: (a), SS; (b), SC; (c), CS; (d), CC.

Thickness

Y. Sun et al. / Journal of Petroleum Science and Engineering 128 (2015) 4452

49

string is restricted. It is obviously found that the boundary condition

signicantly affects on the buckling deformation of drill string,
especially, for the short drill string. When the length of the drill string
is increased, the effect of boundary condition becomes weaker.
4.2. Example II

Fig. 10. Comparisons of corresponding with the rst three post buckling
equilibrium paths for SS (P 0, 0:4).

Fig. 11. Comparisons of W n corresponding with the rst three post buckling
equilibrium paths for SS (P 0, 0:4).

In this section, the buckling behavior of certain vertical scientic

drilling well named as SW well is analyzed by employing the above
model, for 0 and only the SS boundary condition. The designing
sketch of the SW well structure is displayed in Fig. 9. Because we are
focusing on the buckling deformation of the BHA (i.e. the drill string
assembly under the neutral point), therefore, only the parameters of
the BHA are shown in Table 1. Out diameter of the drill string
d 139:7 mm (i.e. r 5:15 mm), Youngs modulus of steel material
E 2:0  1011 Pa, q 784 N=m, and 0:4 are given for model
simplication. And then, the dimensionless parameters could be
obtained as: 0:00012943, J 0:000323584, q 1:04213. Furthermore, the rst three critical linear buckling loads are P 1cr
 4:473963, P 3cr  2:251563, P 3cr  0:683373.
Note that the top end of the BHA is the position of the neutral
point (i.e. P 0). The higher the position of the neutral point is, the
longer the BHA is. Therefore, buckling of the drill string is more
serious. In order to suppress buckling, or at least minimize it, one
can decrease the length of BHA via proper selection of drillstring
combination, or adding stabilizers.
For P 0, corresponding with the rst three post buckling
equilibrium paths, the post buckling deformation , the contact
force per unit length of the drill string W n , and the frictional force
f , and the 3-D post buckling congurations of a drill string are
displayed in Figs. 1013, respectively. Note that origin point of the
drill string centerline coordinate s 0 is the neutral point, and
s L locates in the drill bit.
From Figs. 1013, it is not hard to observe that the maximum
deection of the rst buckling mode of the BHA is very large,
because of the very large dimensional length. It is bad for drilling
engineering; therefore, some strategies should be adopted to
solve this problem. One could choose to add stabilizers to
decrease the buckling deection of the drill string. Moreover,
the locations of the added stabilizers could be at the equilibrium
points (i.e. 0), or the maximum point. The advantages of
adding stabilizers at the equilibrium points of second or higher
order buckling modes are that the features of the original system
are not changed, such as deformation shapes, 3-D post buckling
congurations. However, the loading capacity is improved (the
length is equivalently shorter). Note that the function of adding
stabilizers is that the model is equivalently added to simple
support, corresponding length of the drill string being shorter,
the loading capacity is greatly improved. In the other hand, when
the locations of the added stabilizers are at the maximum point,
the features of the original system are changed, such as deformation shapes, 3-D post buckling congurations, and the critical
loads. By the way, the friction force in the vertical well is smaller
than the horizontal one for the different direction between
contact force and gravity.
4.3. Example III

Fig. 12. Comparisons of f corresponding with the rst three post buckling
equilibrium paths for SS (P 0, 0:4).

In this section, the buckling behavior of the BHA in certain

vertical well adopting impregnated diamond core bit is analyzed
by employing the above model, for 0, 0:4 and only the SS
boundary condition. The aim of the example is to analyze the
effect of drill string wall thickness on the stability of BHA, given
the bit size (i.e. given diameter of the well D 150 mm, and drill

50

Y. Sun et al. / Journal of Petroleum Science and Engineering 128 (2015) 4452

Fig. 13. Buckling congurations of drill string corresponding with the rst three post buckling equilibrium paths for SS (P 0, 0:4): (a), the rst equilibrium path;
(b), the second equilibrium path; (c), the third equilibrium path.

Table 2
The parameters of the drill string for impregnated diamond core bit.
Values of the parameters of the drill string

Thickness =mm
25.4

34.9

Out diameter d=mm

The length L=m
Specic gravity q=N=m

127
95
524

127
75
653

Bending stiffness EI=Nm2

The space gap r=mm
The upper end pressure P

2:33412  106
11:5
0

2:57131  106
11:5
0

Fig. 15. Comparisons of W n of the drill strings with two different wall thickness for
SS (P 0, 0:4).

Fig. 14. Comparisons of of the drill strings with two different wall thickness for
SS (P 0, 0:4).

pressure 50 kN), and the same outside diameter of the drill string
d 127 mm. Via calculating, the values of the parameters the two
different wall thickness drill strings for impregnated diamond core
bit are listed in Table 2.
For P 0 (i.e. the top end of the BHA is the position of the
neutral point), the post buckling deformation , the contact force
per unit length of the drill string W n , and the frictional force f , and
the 3-D post buckling congurations of two different wall thickness ( 25:4 mmand 34:9 mm) drill strings are displayed in
Figs. 1417, respectively.
The maximum deformation max of the drill string with wall
thickness 25:4 mm is beyond 1.5 times of the drill string with

Y. Sun et al. / Journal of Petroleum Science and Engineering 128 (2015) 4452

Fig. 16. Comparisons of f of the drill strings with two different wall thickness for
SS (P 0, 0:4).

51

Fig. 18. Post-buckling mode of the oblique drill strings for SS (P 800 N).

4.4. Example IV
Consider the buckling of an oblique drill string, the geometrical and material parameters are =4, r 0:01727 m, L
21:34 m, q 13:2 N=m, EI 343:3498 Nm2 (Munteanu and
Barraco, 2008). The value of compressive load is given as
P 800 N. In this section, the buckling behavior of an oblique drill
string is analyzed by employing the above model, for only the SS
boundary condition. For various values of the coefcient of friction, the
angles , and congures of the drill string are shown in Figs. 18 and 19,
respectively.
As could be seen from Fig. 18, the friction load has a
signicantly effect on the post-buckling mode . The maximum
value j max j decreases with the increase of , its value is about
0:4363 rad for 0:0, while the value is 0:2223 rad for 0:4.
The relative error of j max j , with respect to the one for 0:0,
reaches about 49:0%. Similarly, one could nd that the friction
load plays important roles on the post-buckling congure of the
oblique drill string, from Fig. 19. With the increase of , the level
of post-buckling congure becomes less severe. Specically, the
maximum values j vmax j are 0:7298 cm, 0:5844 cm, 0:3808 cm,
for 0:0, 0:2, 0:4, respectively. While the relative errors
of j max j , with respect to the one for 0:0, are about 19:9%,
47:8%, respectively.

5. Conclusions

Fig. 17. Buckling congurations of the drill strings with two different wall
thickness for SS (P 0, 0:4): (a), for 25:4 mm; (b), for 34:9 mm.

wall thickness 34:9 mm; while the 3-D post buckling conguration of the wall thickness 25:4 mm drill strings is more
serious than the wall thickness 34:9 mm one. From these
gures, given drill pressure, one could conclude that adopting
thicker wall drill string could make smaller buckling deformation
for the same out diameter of the drill string, even for including the
effect of friction load.

The nonlinear buckling of drill string in a well, subjected to end

compression, gravity, and friction, is successfully analyzed by using the
extended system shooting method. The accuracy of the present
method is veried by comparisons the present data to results obtained
by the nite element method and existing data in literatures, through
numerical examples with various values of friction coefcients and
four different boundary conditions. The effect of the friction force on
lateral and helical buckling loads is revealed. Via two scientic drilling
well instances, effect of friction load, and wall thickness of the drill
string on the buckling deformation are presented. The results are
useful for practical design applications related to calculation of
buckling loads and selection of the parameters of drilling well, such
as the bottom hole assembly (BHA) elements.

52

Y. Sun et al. / Journal of Petroleum Science and Engineering 128 (2015) 4452

Fig. 19. Post-buckling congurations of the oblique drill strings for SS (P 800 N): (a) for 0:0; (b) for 0:2; (c) for 0:4.

Acknowledgments
The work described in this paper is supported by the National
Natural Science Foundation of China (Grant no. 11402095), International S&T Cooperation Program of China (Grant no. 2013D
FR70490), and Innovative Project of Scientic Forefront and Interdisciplinary of Jilin University (Grant no. 2013ZY14).
References
Cheatham, J.B., Pattillo, P.D., 1984. Helical postbuckling conguration of a weightless column under the action of an axial load. Soc. Pet. Eng. J. AIME 36, 467472.
Chen, Y.C., Lin, Y.H., Cheatham, J.B., 1990. Tubing and casing buckling in horizontal
wells. J. Pet. Technol. 42 (191), 140141.
Dawson, R., Paslay, P.R., 1984. Drillpipe buckling in inclined holes. J. Pet. Technol. 36,
17341738.
Gan, L., Wang, X., Tan, M., 2009. Nonlinear buckling analysis of tubular in a deviated
well with differential quadrature element incremental iterative method. Chin.
J. Comput. Phys. 26, 129134.
Gao, D.L., Liu, F.W., Xu, B.Y., 2002. Buckling behavior of pipes in oil and gas wells.
Prog. Nat. Sci. 12 (2), 126130.
Gao, G.H., Miska, S., 2009. Effects of boundary conditions and friction on static
buckling of pipe in a horizontal well. SPE, 782796.
Gulyayev, V.I., Gaidaichuk, V.V., Solovjov, I.L., Gorbunovich, I.V., 2009. The buckling
of elongated rotating drill strings. J. Pet. Sci. Eng. 67, 140148.
He, X., Kyllingstad, A., 1993. Helical buckling and lock-up conditions for coiled
tubing in curved wells. SPE, 25370.
Huang, N.C., Pattillo, P.D., 2000. Helical buckling of a tube in an inclined wellbore.
Int. J. Non Linear Mech. 35, 911923.
Lubinski, A., Althouse, W.S., Logan, J.L., 1962. Helical buckling of tubing sealed in
packers. J. Pet. Technol. 14, 655670.

McCann, R.C., Suryanarayana, P.V.R., 1994. Experimental study of curvature and

frictional effects on buckling. OTC 7568. In: Presented at the 26th Annual
Offshore Technology Conference, Houston, TX, 25 May.
Miska, S., Cunha, J.C., 1995. An analysis of helical buckling of tubulars subjected to
axial and torsional loading in inclined wellbores. SPE 29460. In: Presented at
the SPE Production Operations Symposium, Oklahoma City, OK, 24 April.
Mitchell, R.F., 1986. Simple frictional analysis of helical buckling of tubing. SPE
13064, 457465.
Mitchell, R.F., 1988. New concepts for helical buckling. SPE Drill. Eng. 3, 303310.
Munteanu, M.G., Barraco, A., 2008. Post buckling behavior of a slender beam
constrained to a cylindrical tube. In: Fifth ECCOMAS, Venice, Italy.
Paslay, P.R., Bogy, D.B., 1964. The stability of a circular rod laterally constrained to
be in contact with an inclined circular cylinder. J. Appl. Mech. Trans.-ASME 31,
605610.
Seydel, R., 1994. Practical Bifurcation and Stability Analysis-From Equilibrium to
Chaos, second ed. Springer Verlag.
Tan, M.L., Gan, L.F., 2009. Equilibrium equations for nonlinear buckling analysis of
drill-strings in 3D curved well-bores. Sci. China, Ser. E: Technol. Sci. 52 (3),
590595.
Wang, X., Yuan, Z., 2012. Investigation of frictional effects on the nonlinear buckling
behavior of a circular rod laterally constrained in a horizontal rigid cylinder.
J. Pet. Sci. Eng. 90-91, 7078.
Wicks, N., Wardle, B.L., Patis, D., 2008. Horizontal cylinder-in-cylinder buckling
under compression and torsion: review and application to composite drill pipe.
Int. J. Mech. Sci. 50 (3), 538549.
Wu, J., Juvkam-Wold, H.C., 1993. Helical buckling of pipes in extended reach and
horizontal wellsPart 2: Frictional drag analysis. J. Energy Resour. Technol. 115,
196201.
Yu, Y.P., Wu, B.S., Lim, C.W., 2012. Numerical and analytical approximations to large
post-buckling deformation of MEMS. Int. J. Mech. Sci. 55, 95103.
Yuan, Z., Wang, X., 2012. Nonlinear buckling analysis of inclined circular cylinderin-cylinder by the discrete singular convolution. Int. J. Non Linear Mech. 47,
699711.