0% found this document useful (0 votes)
36 views12 pages

Longitudinal Condition Assessment of Exposed Pipelines Encased in Concrete Box in A Riverbed

Uploaded by

LUCKSON MUZINDA
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
36 views12 pages

Longitudinal Condition Assessment of Exposed Pipelines Encased in Concrete Box in A Riverbed

Uploaded by

LUCKSON MUZINDA
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 12

ICPTT 2009 © 2009 ASCE 102

Longitudinal Condition Assessment of Exposed Pipelines Encased in Concrete Box


in a Riverbed

Jong Hwa Won1, Moon Kyum Kim2, Young-Ho Woo3, Tae Min Kim4, Won Joon Lee 5
1
Ph.D. Candidate, Dept. of Civil and Environmental Engineering, Yonsei University,
Downloaded from ascelibrary.org by Carleton University on 11/01/20. Copyright ASCE. For personal use only; all rights reserved.

Seoul, Korea; Fax: +82-2-364-5300; Phone: +82-2-2123-2803; e-mail:


[email protected]
2
Professor, Dept. of Civil and Environmental Engineering, Yonsei University, Seoul,
Korea; Fax: +82-2-364-5300; Phone: +82-2-2123-2803; e-mail: [email protected]
3
Professor, Dept. of Civil Engineering, Myongji College, Seoul, Korea; Fax:
+82-2-303-1132; Phone: +82-2-300-1138; e-mail: [email protected]
4
MS. Candidate, Dept. of Civil and Environmental Engineering, Yonsei University,
Seoul, Korea; Fax: +82-2-364-5300; Tel: +82-2-2123-2803; e-mail:
[email protected]
5
MS. Candidate, Dept. of Civil and Environmental Engineering, Yonsei University,
Seoul, Korea; Fax: +82-2-364-5300; Phone: +82-2-2123-2803; e-mail:
[email protected]

ABSTRACT

Scour around a river crossing pipeline creates a free span of pipeline and leads to
excessive flexural and deformation stresses. Generally, generated stresses on pipelines
under high internal pressure are dominated by circumferential stress. Longitudinal stress,
however, is a main contributor in the procedure of stress development because a
free-span pipeline exhibits unstable behavior as a simple supported beam and results in
bending stress. For this study, a specific section of river crossing pipeline was defined as
follows: API 5L Gr. X65, Do=762mm (30-inch). Based on a plastic design concept, the
interface stress for the steel-concrete composite pipeline and its performance under
specific flow velocity were defined using results from FEM analysis. In order to
describe the loading condition of the exposed encased pipelines, Prandtl-von Karman
universal velocity law is adopted with rough river bed surface, fixed water depth and
variable scouring depth. In the general condition of a river, although 20m free-span
pipelines have been developed, encased pipelines perform well according to the
von-Mises failure criterion.

INTRODUCTION

In the case of a river crossing pipeline, there are many risk factors which can create
differences in the burial environment before and after pipeline construction. Among
them, scouring is one of the most critical factors threatening a pipeline’s integrity, and
cover depth over the pipeline thus should be monitored. Scour around a river crossing

ICPTT 2009
ICPTT 2009 © 2009 ASCE 103

pipeline creates a free span of pipeline and leads to excessive stresses due to
hydrodynamic forces in all directions in terms of self weight, lifting and drag forces.
Moreover, in Korea, rivers have very large coefficients of river regime; as seasons
change, the level of river beds fluctuate through a large range. Approximately a half of
the annual precipitation falls during the summer rainy season, as a result the coefficient
of the river regime in Korea usually range from 100 up to 700. So riverbed erosion is
easy to occur during the rainy season. The large variation in the flow discharge causes
Downloaded from ascelibrary.org by Carleton University on 11/01/20. Copyright ASCE. For personal use only; all rights reserved.

problems in pipeline management concerning scouring. Actually, National


Transportation Safety Board (NTSB) reported due to the flooding, eight pipelines
ruptured and many others were undermied in the San Jacinto River (NTSB 1996). So
until now, many studies have been investigated to define scouring and to to prevent
damage from this phemomenon. Samchek and others conducted case studies of scour
and erosion at water crossings (Samchek, et.al. 2004). Another study investigated a case
history of scour impacted pipelines in river crossings (Rajah, et. al. 2005). Joyce and
Chantler proposed a method of estimating the potential bed scour and a design method
for protecting pipelines(Joyce and Chantler 2004).
Lately many pipeline designers have become aware of erosion of the riverbed at the
pipeline crossing. Therefore, to protect pipeline from potential damage, the KOGAS
(Korea Gas Corporation) design criteria holds that buried pipes should be encased in
concrete in the case of river crossing pipelines due to unexpected and sudden impacts
such as from anchors.
Generally, generated stresses on pipelines under high internal pressure are dominated by
hoop stress. Longitudinal stress, however, is a main contributor in the procedure of
stress development because a free-span pipeline exhibits unstable behavior as a simple
supported beam and results in bending stress. However, no research has been done on
longitudinal (longitudinal direction) behavior and the relationship of stress distribution
between the inner steel pipe and outer protective structure for pipelines encased in
rectangular concrete.

KOGAS PIPELINE DESIGN CRITERIA

KOGAS established a pipeline design specification for the safe and efficient operation
of pipelines system. The guideline defines material, diameter, internal pressure and
encasement for each burial condition of long distance transportation pipelines.
Generally, the API 5L Grade X65 is used for gas transportation pipeline which is
762mm (30") in outside diameter for the purpose of transportation amount of gas from
the storage to supply base. In the case of urban areas or regions which have poor
conditions for buried pipelines, they have a 17.5 mm wall thickness and its operation
internal pressure of 7.85 N/mm2. Moreover, KOGAS recommends that offshore or
river crossing pipelines should be protected by a concrete encasement. Figure 1 (a)
shows the model for this research and Figure 1 (b) is the example of pipelines encased in
rectangular concrete box for the case of Lower Snake River crossing gas pipeline.
Properties of materials are summarized in table 1.

ICPTT 2009
ICPTT 2009 © 2009 ASCE 104
Downloaded from ascelibrary.org by Carleton University on 11/01/20. Copyright ASCE. For personal use only; all rights reserved.

(a) KOGAS Concrete Encasement (b) Typical Gas Pipeline trench Section for
Standard for X65 Steel pipe in Outside Lower Snake River Crossing Gas Pipeline
Diameter 762mm (30 inch) (US Army Corps of Engineers, 1999)
Figure 1. Protection encasement for river crossing pipeline

TABLE 1. Steel Pipe and Concrete Encasement Characteristics


2 2
E (N/mm ) γ (N/mm ) υ remarks
API 5L σy 448 N/mm2
2.1 ⅹ105 7.6 ⅹ 10-4 0.3
Gr. X65 σu 530 N/mm2
4 -4
Concrete 2.0 ⅹ10 2.4 ⅹ 10 0.2

THEORETICAL APPROACH FOR DOUBLE-WALLED PIPELINES

In this study, it is necessary to approach with longitudinal integrity assessment of


unsupported pipelines river crossing by current-induced scouring effect. The
longitudinal analysis in pipe mechanics divided into two basic components of
longitudinal and flexural analysis. For considering longitudinal bending effect of flow
velocity variation as a beam, only flexural analysis is conducted based on the result of
numerical analysis. Also cylinder sectional analysis is essential step to extract stress
induced pure-bending except longitudinal effect (Poisson’s effect).

LONGITUDINAL EFFECT - FLEXURAL ANALYSIS

Beam bending (longitudinal deflection), which causes flexural stresses is typically


occurred by a)non-uniform settlement or bedding, b)permanent ground deformation such
as side-hill soil creep or land slide and massive soil movement or settlement and
c)current flowing on exposed free-spanning pipeline. In order to estimate longitudinal
bending stress, geometry of deflected pipeline should be defined as following figure 2.
(b). The flexural stress could be obtained from mean radius of the bend, RL . It can be

ICPTT 2009
ICPTT 2009 © 2009 ASCE 105

measured by a certain length, s, along the inside of the bend and by measuring the offset,
e, at the middle of the bend as shown in figure 2. Equation (1) is the exact value for the
mean radius of the bend.
Downloaded from ascelibrary.org by Carleton University on 11/01/20. Copyright ASCE. For personal use only; all rights reserved.

(a) Schematic of Exposure (b)Geometry for Evaluating Deformed Radious, rL


Pipeline
Figure 2. Longitudinal Behavior of In-stream Pipelines

s2 e
R = + + r = RL + r (1)
8e 2
where, R = mean bent raduis of the center of pipe cross section, RL = mean raduis of
the bend, s = straightedge of known length and e = offset at the middle of the bend. In
the case of initially straight pipeline bent into a circular arc of radius R , longitudinal
strain and stress due to beam bending could be calculated following (2) based on
elastic theory.
r r
εL = and σL = E
R R
(2)
where, ε L = longitudinal strain, σ L = longitudinal stress, E = Young’s modulus, and
r = outside radius of the pipe cross section.

SECTIONAL ANALYSIS OF PIPES ENCASED IN RECTANGULAR


CONCRETE

In order to estimate the stress distribution between the inner and the outer cylinder, the
interface pressure should be defined in advance. The magnitude of the resulting interface

ICPTT 2009
ICPTT 2009 © 2009 ASCE 106

pressure between members may be calculated based on same radial displacement at the
interface between inner and outer cylinder. Equation (3) can be adopted in the process to
determine interface pressure between the inner pipe and the outer rectangular
encasement. Referring to figure 3, the thickness of outer encasement is changeable from
place to place such as crown (top of pipe, 0°), shoulder (45°), springing (90°) and
bottom (180°). Therefore, thickness of encasement should be considered for each part of
the pipe. In this study, every point of pipe section is examined. In order to express in
Downloaded from ascelibrary.org by Carleton University on 11/01/20. Copyright ASCE. For personal use only; all rights reserved.

terms of interface pressure, evaluation of radial displacement and Hooke’s law result in
the interface pressure on the concrete encasement inner surface.

Pi Pi
P= = (3)
⎛ E ⎞ ⎧ 2(b − a ) ⎫ ⎛ b + c 2
⎞ 2
⎛ E ⎞ ⎛ 2t ⎞ ⎛ b 2 + c 2 ⎞
1+ ⎜ s ⎟ ⎨ ⎬ ⎜ 2 2 + υc ⎟ 1+ ⎜ s ⎟ ⎜ ⎟ ⎜ 2 2 + υc ⎟
⎝ Ec ⎠ ⎩ b + a ⎭ ⎝ c − b ⎠ ⎝ Ec ⎠ ⎝ 2b − t ⎠ ⎝ c − b ⎠

where, a = inner radius and b = outer radius of inner pipes and c =distance from center
to outside of encasement, Ec = elastic modulus of concrete and υc = Poisson’s ratio of
concrete Es = elastic modulus of steel and t = thickness of pipe cross section = b − a .
The concrete encasement thickness ti is changeable according to its location and it can
be expressed in terms of θ . Expressing the distance from the center to the encasement
outside surface, c ' in the function of θ ;
c
c ' = ti + b = (0D < θ ≤ 45D ) (4)
cos θ

c
c ' = ti + b = (45D < θ ≤ 90D ) (5)
cos(90 − θ )

where, ti (i = Cr , Sh, Sp, bt ) = encasement thickness at i point, c = the distance from the
center to the outside surface of concrete encasement at θ = 0 . In order to make suitable
equation for cylinder in rectangular encasement, substituting (4) and (5) into (3) ( c ' into
c ), results in (6) and (7).

Pi
P= (0D < θ ≤ 45D ) (6)
⎛ Es ⎞ ⎛ 2t ⎞ ⎧ b 2 cos 2 θ + c 2 ⎫
1+ ⎜ ⎟⎜ ⎟⎨ 2 2 + υc ⎬
⎝ Ec ⎠ ⎝ 2b − t ⎠ ⎩ c − b cos θ
2

Pi
P= (45D < θ ≤ 90D ) (7)
⎛ E ⎞ ⎛ 2t ⎞ ⎧ b 2 sin 2 θ + c 2 ⎫
1+ ⎜ s ⎟⎜ ⎟ ⎨ 2 2 2 + υc ⎬
⎝ Ec ⎠ ⎝ 2b − t ⎠ ⎩ c − b sin θ ⎭

ICPTT 2009
ICPTT 2009 © 2009 ASCE 107
Downloaded from ascelibrary.org by Carleton University on 11/01/20. Copyright ASCE. For personal use only; all rights reserved.

Figure 3. Terminology of pipe and concrete encasement cross section

OPEN-CHANNEL VELOCITY DISTRIBUTION MODEL

The depth of the flow and the boundary friction between the channel surface and the
fluid are two main factors which have an effect on the velocity of the flow in an open
channel. The various velocity distribution induces different drag forces and lift forces.
Therefore, when an object is exposed to running water in an open channel, it is affected
by various drag and lift forces. The exposed pipe in an open channel flow is shown in
FIGURE 4. The parameter L denotes the axial length of the exposed pipe and it is taken
as 10m. There is friction between a running fluid and the boundary of the channel in the
real condition. Therefore, the velocity of flow has parabolic distribution as following
(13).
2x
v1 = v0 (1 − ( ) 2 )
L
(13)
where, v1 stands for the velocity of flow. It is 0 at the boundary wall and becomes the
maximum velocity, v0 in the middle of the channel as below figure 4.
There are two kinds of velocity distributions in turbulent flow. One is called potential
flow where the distribution is uniform along the depth of the channel. The other is
boundary layer flow where the velocity keeps decreasing parabolically as the depth
increases as above FIGURE 5. In the boundary layer flow, the velocity distribution
suggested by Prandtl-von Karman may be taken as:
30 z
v = 5.75V f log (14)
k
where, v = velocity of the flow according to the distance z from the channel bed, k =
the roughness height (=0.6069m, Han River, Korea) and V f = friction velocity; that is,

V f = gHs (15)

ICPTT 2009
ICPTT 2009 © 2009 ASCE 108

where, g = gravity acceleration and s = degree of slope. The exposed pipe is affected by
the drag force, that acts in the direction of flow on the channel bed; that is,
V2
f D = CD AP ρ (16)
2
In this equation, ρ = density of water, AP = wetted area, C D = drag coefficient (=2.0
Downloaded from ascelibrary.org by Carleton University on 11/01/20. Copyright ASCE. For personal use only; all rights reserved.

for regular square section) and V = velocity of the flow. The lift force, f L also has an
effect on the exposed pipe. It may be represented by the well known Bernoulli energy
equation.
γ
+P = − (v12 − v22 ) − γ ( z1 − z2 ) (17)
2g
where, ΔP = pressure difference, γ = unit weight of water and vi and zi =
velocity of the flow and the elevation respectively at point i measured from the
channel bottom( i =1 : upper bound of encasement, i =2 : bottom). Therefore, z1 − z2
indicates the height of the concrete protection.

Figure 4. The overview of the exposed Figure 5. The front view of the exposed
pipe in an open chanel pipe in an open chanel

ANALYSIS CONDITION

A realistic characterization of the behavior of the exposed pipeline can be based on the
following assumptions:

y the cross section of the channel remains a rectangle;


y there is no separation between the pipe and the concrete protection;
y the application of this document is limited to elastic response;
y all the points of the scoured channel bed have same elevation.

ICPTT 2009
ICPTT 2009 © 2009 ASCE 109

The front view of the exposed pipe in an open channel is shown in the above FIGURE 5.
H denotes the depth of the channel. Considering the maximum depth when the flood
occurs, 20m is selected as a standard on the basis of the information from the Han River
flood control office. The parameter h is the length from the scoured channel bed to the
bottom of the concrete protection. For numerical analysis, 1m and 2m are used as the
dimension of h respectively. The parameter t is the thickness of the concrete protection
Downloaded from ascelibrary.org by Carleton University on 11/01/20. Copyright ASCE. For personal use only; all rights reserved.

where 0.15m is taken to represent the specification of Korean Gas Corporation. Two
different mesh systems are used in the present model as seen in FIGURE 6. Rectangular
plate meshes are used for the exposed pipe and solid meshes for the concrete protection.
It is assumed that both ends of the pipe and the protection are fixed as below FIGURE 6
(b). The magnitude of the internal pressure of the pipe is 7.85 N / mm 2 . Both drag and
life forces are applied diversely according to the flow velocity distribution. These factors
calculated at the upper and lower sides of the concrete protection are summarized as
below table 2.

(a) Cross section (b) Fixed Boundary Condition of both Ends of the Encased
Pipe
Figure 6.FEA MODEL of exposed river-crossing pipeline enchased in rectangular
concrete

Table 2. Distribution of the velocity , life force and drag


force at the upper and lower sides of the
concrete protection (scour depth = 1m)

x ( m)
v1 (m / sec) v2 ( m / sec) f L ( N / mm 2 ) f D ( N / mm 2 )

0.375 5.207 4.391 1.405 2.397


0.875 5.076 4.281 1.386 2.278
1.375 4.840 4.082 1.352 2.071
1.875 4.500 3.795 1.308 1.790
2.375 4.055 3.420 1.254 1.454
2.875 3.505 2.956 1.195 1.086
3.375 2.850 2.404 1.136 0.718

ICPTT 2009
ICPTT 2009 © 2009 ASCE 110

3.875 2.091 1.764 1.083 0.387


4.375 1.227 1.035 1.042 0.133
4.875 0.259 0.218 1.022 0.006

BEHAVIOR OF EXPOSED PIPELINE IN RECTANGULAR ENCASEMENT


SECTIONAL ANALYSIS CONSIDERING REISSNER EFFECT. In order to
examine characteristics of behavior of pipes encased in rectangular concrete, analytical
Downloaded from ascelibrary.org by Carleton University on 11/01/20. Copyright ASCE. For personal use only; all rights reserved.

and numerical method were concurrently performed. This study has following steps; (a)
FE Analysis, (b) Calculation of Interface Pressure and (c) Local Deformed Diameter
from Hoop Stress. Following figure 7 shows relative local deformed radius for each case
of scour depth 1m and 2m at the center cross section of pipeline in longitudinal direction.
And original radius (381mm, 15inch) is replaced with 1 in figure 7. This model plotted
in figure 7, for pipe in rectangular case, can explain the phenomenon that the maximum
stress occurs at the shoulder parts. A bold line means an original radius and ‘Normal’
means analysis result including self weight and internal pressure. Radial directional
distances from origin indicate the deformed radius of specific point. Because deformed
radius has proportional value with hoop stress, in the following figures, main contributor
to occur stress can be easily found for each location in terms of θ .

LONGITUDINAL ANALYSIS – FLEXURAL BEHAVIOR. To only focus on the


flexural behavior due to current (drag and lifting force), the analysis covering self
weight and internal pressure is conduct in advance and it is plotted in figure 8 as
‘normal’. Based on axial stress from FEA, the Poisson’s effect is eliminated by
differences between ‘normal’ case and each ‘lifting’ and ‘drag’. Figure 8 presents the
contribution of each force component (gravity, lifting and drag force) and degree means
the location on pipe cross section in the clock-wise from top (0°). The attack angle for
drag and lifting force is 90° and 180° (bottom). Due to variation of encasement thickness,
the value for ‘normal’ has maximum stress at shoulder part by Poisson’s effect. The
noteworthy result in figure 8 is maximum tensile stress at 270°, 315° and 360° (0°) and
its stress contributors. At these points, every stress component is under tensile value and
therefore, they have larger level of axial stress. The maximum axial stress develops at
left-upper side shoulder (315°) by both of lifting and drag force.

ICPTT 2009
ICPTT 2009 © 2009 ASCE 111
Downloaded from ascelibrary.org by Carleton University on 11/01/20. Copyright ASCE. For personal use only; all rights reserved.

(a) Scour depth = 1m (b) Scour depth = 2m


Figure 7.Relative local radius according to scour depth at the center section of
crossing

(a) 0° (b) 45°

(c) 90° (d) 135°

ICPTT 2009
ICPTT 2009 © 2009 ASCE 112
Downloaded from ascelibrary.org by Carleton University on 11/01/20. Copyright ASCE. For personal use only; all rights reserved.

(e) 180° (f) 225°

(g) 270° (h) 315°

Figure 8. axial stress according to pipe cross section location

CONCLUSION

The exposed pipeline encased in rectangular concrete on current shows entirely different
behaviors than the plain pipeline. This study defines stress contributors when exposed
pipeline is in a bending. the local deformed diameter is used for cross sectional analysis
and for logitudinal anlaysis, axial stress considering Reisnner and Poisson’s effect is
used. The following conclusions are drawn from this study:
The deformed radius can be an index describes a performance of Pipeline with
encasement. In the view of pipe cross sectional behavior, because its thickness of radial
direction is changeable according to θ , flat spot develops at every shoulder part. The
influence of scour depth is negligibly small and the largest deformed radius occurs at
left-upper side shoulder due to both of lifting and drag force. Main stress contributor at
each point could be easily found in figure 7.
In the longitudinal analysis, lifting and drag force conducts as the only loading condition
by removing stress due to Reisnner and Poisson’s effect. Generally, the maximum
longitudinal (axial) stress occurs at the springing (θ = 90D ,180D ) in exposed pipeline on

ICPTT 2009
ICPTT 2009 © 2009 ASCE 113

currents due to drag force. The noteworthy result is maximum tensile stress occurring at
315°. At that point, every stress component is under tensile value by lifting and drag
force and especially effect of rectangular encasement. This phenomenon of stress
distribution can be explained as a sectional model described in this study.

REFERENCES
Downloaded from ascelibrary.org by Carleton University on 11/01/20. Copyright ASCE. For personal use only; all rights reserved.

Hugh W. O’Donnell (2005). “Investigation of Flood Induced Pipeline Failures on Lower


San Jacinto River.” Pipeline 2005, ASCE, 451-463.
Jong Hwa Won, Moon Kyum Kim, Dong Hyeon Ryu (2007). “Characteristics of an
Encased Gas Transportation Pipeline in Offshore Application. ” ISOPE 2007, in
CD.
K. Abhary, A. Nourbakhsh, and L.L.H.S. Loung (1999). “An exact analytical method
for stress analysis of pipelines.” J. International Journal of Pressure Vessels
and Piping, 76, 561–565
KOGAS R&D Center. (1997). “GSM-2101 Pipeline Design Standard.” , KOGAS.
Reynold King Watkins and Loren Runar Anderson. (2000). “STRUCTURAL
MECHANICS OF BURIED PIPES.” M. CRC Press.
S. P. Timoshenko and J. N. Goodier .(1987). “ Theory of Elastic.” M. McGRAW-Hill.
Ugural. A. C. and Fenster, S. K.(1995). “Advanced Strength and Applied Elasticity.”,
Prentice Hall.
US Army Corps of Engineers. (1999). “Lower Snake River Juvenile Salmon Migration
Feasibility Report / Environmental Impact Statement.”.
Ven Te Chow, “Open-Channel Hydraulics.” McGRAW-Hill, 1959, 9,12.

ICPTT 2009

You might also like