Szilas A.P. Production and transport of oil and gas, Gathering and Transportation
Подождите немного. Документ загружается.
190
7.
PIPELINE TRANSPORTATION
OF
OIL
0
.-
100
200
300
400
I,
km
Fig. 7.4-
5.
Transport system planning in the
case
of
hilly area, after
Young
(
1960)
determined. In order to protect the low sections against overpressure,
in
valley
I
a
thick walled pipe, marked with double line, in valley
2
pressure reducer station,
marked
a,
and for the valley sections
3
and
4
the application
of
thick-walled pipes,
and the building
of
the pressure reducing stations,
b,
and c was planned.
At
pump
station
B,
the pump discharge pressure was reduced, otherwise the required
pressure head
of
station
C
might have been too small. The version is irreal, since the
pressure traverse at terminal Vwould show a “negative head”, even in case
of
the
highest maximum pressure that can
be
realized at pressure regulating station c.
If
the forecasted rate is smaller, the flow gradient is smaller too, and, sometimes only in
principle, the pump stations can transport to a greater distance. The number of
required stations, may be smaller than in case of higher rates. From the Figure
it
can
be seen that, e.g. the transportation
of
a flow rate
q2,
which is less than
4,,
is
realizable and for this task pump station
C
is not needed. In case of an even smaller
flow rate
q3
on
one
hand, pump station
C
is needless too, on the other hand the
version is irreal, since the minimum possible suction pressure before the pump
station
E
is
so
large that the pipeline pressure in the valley section
2,
would exceed
the allowed value.
(d) Optimum diameter and trace
of
the pipeline
Most advantageous is the pipeline, that, between the given end-points, enables
the transportation
of
the prescribed oil flow rate at the minimum transportation
cost. The transportation cost consists of three cost components, differing in
7.4.
ISOTHERMAL
OIL
TRANSPORT
191
character, and these are the depreciation costs
of
the investment? the maintenance
plus operational costs, and the other, additional costs.
In
the next section, only the
first two components
will
be considered, and their sums will be called, as direct costs.
This simplification does not hamper the determination of the relative economic
efficiency of several technical solutions, i.e. the choice
of
the version providing the
minimum transportation cost. The total cost is proportionally greater with the third
component. The consideration of this is justified
if
the costs
of
pipeline transport are
compared to the transport costs of other methods e.g. to the costs
of
transporting by
containers.
In
order to determine the minimum value
of
the direct transportation costs, two
problems must be solved: a) the determination of the optimum pipc diameter, and b)
the designing
of
the optimum trace. Solving these problems, we suppose that the
yearly oil transport is constant during the operational life
of
the pipeline. The real
circumstances are generally more complicated. The pipeline is operating several
decades long.
If
it
transports only the production
of
a relatively known oil
producing area, then the yearly throughput generally changes, but its rate may
approximately forecasted. In such a case the pipe diameter is considered to be the
optimum, at which the aggregate cost
of the transportation is the lowest,
considering the total transportation life. While designing
it
must be also considered,
that from year by year, the
oil
volume will be transported possibly by pump stations
of
various numbers, and capacity. The situation may be further complicated,
if
the
planned annual throughput is modified by unexpected circumstances, e.g. a new oil
field is discovered, and its output should be transported also by the given pipeline.
In
such cases, starting from the existing transport system, we must design and realize a
capacity increase that enables the reconstruction in the most economical way.
(A)
The optimum pipe diameter
If
the throughput to be transported annually is considered to be constant, then
the optimum diameter of the pipeline can be determined by the following
consideration. From
Fig.
7.4-6
it
can be seen that in case of constant yearly
throughput, the amortization cost
A
increases with the increase
in
the pipe diameter,
while the operational cost
B
is reduced. The total cost of transportation,
K,
is
obtained by summing the values of
A
and
B.
The curve
K
=f(di)
has a minimum
value. The pipe diameter
diop,,
corresponding to this minimum point is the most
advantageous, most economical.
diop,
may be determined in one of two ways. a) Estimating costs for various pipe
sizes, a plot similar
to
the one in
Fig.
7.4-6
is prepared and
diop,
is determined
graphically. b) Optimum pipe size is calculated using a suitable mathematical
formula, e.g. the one introduced by Smith-Brady-Donnell.
The total annual cost of transportation is
K=A,+A2+B1+B2.
7.4-
16
I92
-
7.
PIPELINE TRANSPORTATION
OF
OIL
I
I
I
-
Here, depreciation
A,
is defined as
where
a,
is
a cost component independent
of
the pipe size, e.g. ditch cutting, and
back filling within limits, and supervision in Ft/m.
Ft
being the abbreviation
of
the
Hungarian monetary unit (Forint)*;
azdi
is the cost component depending on the
pipe size (pipe price, transportation to the site, welding, painting, insulation,
pressure testing),
a,
is in Ft/mz units;
z,
is the annual depreciation rate
of
the ready-
to-operate pipeline, and
1
is the length
of
the pipeline
in
m.
The annual depreciation
of
a pump station is
7.4
-
17
where pump station output
P
=
~
4Pgh,
W;
b
is unit cost
of
the station,
in
Ft/W;
zz
is the annual depreciation rate
of
the pump station; and
17
is the overall efficiency
of
the installation.
17
Using
Eqs
7.4-2
and
7.4-9
we get
yo.189
1.81
41
h=1.53~
lo-’
d4,81
.
Other
occurrencies,
of
course,
can
be
used in these equations similarly.
7.4.
ISOTHERMAL
OIL
TRANSPORT
I93
Substituting this into Eq. 7.4- 17 we find
7.4-
18
The annual operating cost is
B,
=
Pcef,
where
c
is the operating time over the year in
s;
r
is the specific energy consumption
(J/J for electric motors and kg/J for internal compustion engines);/is the
unit
price
of
power,
or
fuel (expressed
in
Ft/J,
or
Ft/kg).
Using the symbols introduced when deriving Eq. 7.4-
18
we obtain
7.4-
19
The annual cost of maintenance
B2
may be considered constant. Let
us
replace
the terms on the right-hand side of Eq. 7.4-
16
by the above-defined expressions,
and determine the derivative dK/d(dJ. Putting the obtained relation equal to
zero, and solving for the diameter we get, by the rules of calculus the pipe size that
ensures the least cost of transportation
7.4
-
20
In calculating optimal pipe size we have
so
far tacitly assumed the pipeline to
operate without interruption, that is, at
100
percent exploitation.
In
reality,
transportation is most often intermittent to a greater
or
lesser degree. The pauses
in
transportation provide the flexibility needed to handle excess output due to non-
uniform production, breakdown etc., on the input side.
A
lower exploitation
efficiency will, however, augment the specific cost
of
transportation. Transportation
costs may
be
analyzed for various (constant) pipe sizes and variable throughput. Let
us
replace by the above defined expressions the terms of the right-hand side of Eq.
7.4-
16.
Let
us
contract the coefficients except
di
and
4
into constants denoted
Ci;
then we obtain relation
K
=
C,
+
C2di
+
C,di
4'8192'8'
.
Let the length of the line be one km, and let
us
divide both sides ofthe equation by
the annual oil throughput,
C'qp.
Let
K/C'qp=k
denote the unit cost
of
transportation in Ft/(kg km) then
k=C4q-'+C5diq-'+C
6i
d
-4'81
4"81
'
7.4-21
The relationship defines a set of curves,
k
=f(4)di.
The set
in
Fig.
7.4
-
7
has been
plotted, using a similar relationship. The ordinate is calibrated
in
terms of a relative
specific transportation cost
kre,
referred
to
some base value (derived from a diagram
194
7.
PIPELINE TRANSPORTATION
OF-
OIL
d,-508rnm
I
I
I
I
I1
I
,
I
I
I
b’’‘&L
711
2
6801
Ma6
lt2m
658
@
I1
I
I
II
I
762
mm
,I
I
I I
I1
I
I
I
r
I
I
I
I!
17
I
:-p
I
,
I
,
I#
!
by Cabet
1966).
In the Figure, the optimum sizes
in
the individual throughput
ranges are indicated by separate lines. Clearly, the greater the desired throughput,
the larger the optimum pipe size, and the loss the unit transportation cost. This
family ofcurves cannot, ofcourse be used directly in design work, as the constants
in
Eq.
7.4
-
21
will differ from case to case.
d,opI
mm
.a00
~.
700
.600
.
600
.
600
300
Cost components
Fig.
7.4-7.
Optimum pipe size
v.
throughput, after Cabet
(1966)
10”
Francs
%
Table
7.4-1,
after
R.
Cabet
(1966),
lists the main cost components of the Le
Havre-Paris pipeline, built in
1964.
The depreciation costs and other financial
expenses, here, are substantial.
Depreciation is
in
principle the annual decrease in the actual value of an
installation. Book depreciation and actual depreciation often do not agree.
Operating companies prefer book depreciation to be as high as possible. In France,
for
instance, the period for amortization prescribed by law is
30
years for buildings,
Depreciation
Debt servicing
Operating cost
Payroll
Power
Maintenance
Telecommunications
Others
Total
6 534
2 830
I1
931
4 035
2 850
1688
275
3 083
21 295
30.7
13.3
56.0
18.9
13.4
7.9
1.3
14.5
100
7
4
ISOTHLKMAI
011
TKANSI’OKT
I95
20-25 years
for
pipes,
10
years
for
motors, pumps and five years
for
automation,
and control equipment.
In
reality, however, since the introduction of cathodic
protection, pipes have practically unlimited life. Even
so,
although the actual life
of
the equipment and fittings may be longer than the amortization periods mentioned
above, these periods may nevertheless be technologically justified, as the progress of
technology may make obsolete the equipment
in
use by putting on the market
equipment and fittings more advanced both technically and economically. The
relative depreciation (moral wear) of the existing installations
will
thus be
accelerated.
(B)
The
optimum trace
of
piplines
In case of an underground pipeline
of
pre-determined diameter, made of steel
of
a
given grade, the specific investment costs consist
of
two components. The first
of
these is the line cost
for
unit length, while the second component stands
for
the
specificcosts ofdigging the ditch, the laying
of
the pipeline, the backfilling, the clean-
up operations, and the damages to be paid to the owner of the ground, concerned.
Summarizing the different specific investment costs, from section to section
of
different lengths, along the selected trace, we obtain the total investment cost. The
main factors, influencing the optimum trace with the possibly smallest investment
cost, are the following: prohibited areas
for
pipeline laying (e.g. settlements, closed
military areas); the undulation
of
the area; the increase of the pipe-wall thickness at
the section passing through safety zones; the specific construction costs valid for
soils ofdifferent types; the cost
of
the intersection of natural,
or
man-made lakes
or
rivers, marshes; damages; the requirements concerning the branching lines; and
accessibility.
Before the designing
of the trace
of
the pipeline a design map must be prepared
that indicates all the parameters that may influence the construction costs
of
the
pipeline. Previously, the trace on this map was plotted by the designers on personal
considerations. Because
of
the high investment costs, the optimum version should
be determined by applying computers. Such procedures, discussed in the literature
may be classified as the network method, and as the method based on dynamic
programming(Lesic 1973; Babin
et
al.
1972; Shamir 1971). Both from the aspects
of
the data preparation, and the running time on the computer, the procedure based
upon dynamic programming proved to be the preferable one. In the next section the
essence
of
the design method developed at the Department of Petroleum
Engineering
of
the Miskolc Technical University for Heavy Industry, based on
Kaufmann’s procedure (1968) is discussed (Szilas
et
a/.
1978).
The process can be interpreted by the help
of
Fig.
7.4-8.
The head point
of
the
pipeline,
A,
and the tail end-point,
0,
are given. The dashed lines
xo.
.
.
x5
are the
border lines
of
areas
of
different specific construction costs.
For
the sake
of
lucidity,
the boundaries of the prohibited areas and level lines
of
the ground are not
indicated. The former will be referred to whileexplaining the next example, while the
impact
of
the latter is shown by the cost indicating at the connecting lines. On the
196
7
PlPt
1
INL:
?RANSPORTATION
Ot
011
boundary lines points are taken, the number and place
of
which are determined by
the required accuracy. In the Figure these points are marked with symbol
0.
The
first section, plotted between
xo
and
x1
,corresponding
to
the fixed
A
point and the
assumed points
of
B,
C,
and
D
may be a straight line section
of
AB,
AC,
or
AD.
The
construction cost
of
the individual pipe section is obtained that the specific
construction cost
of
the pipe
is
multiplied by the length
of
the pipe section. In the
xo
XI
x2
x3
x4
x5
Fig.
7.4
-
8.
Planning
of
the
optimum
trace
of
the pipeline
Figure these costs are indicated by the figures
S,S,
and
4
written beside the lines. In
general, let
us
mark the construction cost
of
the individual pipe sections by the
formula
u,(x,
~
I
,
x,).
Due
to
similar considerations, the construction costs
of
any
pipeline, crossing the designated points, and connecting the points
A
and
0
may be
calculated from
7.4
-
22
The determination principle
of
the trace
of
minimum cost is that each pipeline
section
must
be optimum, i.e.
of
the minimum construction cost, on the one hand up
to
the point from which the section starts, and on the other hand, from this point
till
the end point
of
the section. Designing is
to
be carried out section by section, by
applying boundary lines.
~~~~o,~I,~2~~3~~4~~s~=~I~~~r~1~+~2~~I~~2~+~~~X.Z,.xj)+
+U,(X,
7
X,)+b(X4,
.xs).
The line section
of
minimum cost between the boundaries
xo
and
xI
is
MIN
uI
(A,
XI)
=
min
(5,3,4)
=
3
where
“MIN”
means the “minimum” restriction referring
to
the mathematical
formula, while “min” indicates that from the numerical values the smallest value is
to
be selected.
XI
designates the point on boundary line
xl.
XI
=
B,
C,
D
,
74
IVTTHFRMAI
011
TRAN\PORT I97
Theconstruction cost of the first, and second line sections between the boundaries
.yo
and
.y2
will be the minimum,
if.
between
A
and
X2
points the traces of minimum
cost
are
determined, and from the values obtained. the smallest is choscn. i.e.
MIN
c~,~(A.
X2),
It
is to be calculated as follows:
MIN
v,,~(A,
E)=MIN
[(r,(A.
X,)+r.z(XI.
E)J=
=min(5+3. 3+4.4+
x)=7
,Y
I
=
H.
c',
n
-L
sign was used where between the two actual points. because
of
prohibitcd
arcas
pipelines may not be layed.
MIN r',,,(A,F)=MIN [(V~(A.X,)+~~(X~.
F)]=
=
min
(5
+
x.
3
+
2.4
+
3)=
5
XI
=
B,
C,
D
MIN
I.~,~(A.
G)=[(~,(A,X,)+Z.~(X,,G)J=
=
min
(5
+
K
.3
+
4.4
+
2)
=
6
.
The minimum values corresponding to the proper points are encircled. The
numerical value
of
MIN
v,,~,~(A,
H);
-(A,
I);
-(A.J)
and
-(A,
K)
may be similarly determined. The minimum is
MIN
C,,~,~(A,
X,)=MIN
L'~,~,~(A.
H)=
I1
.
Considering the first four sections
MIN
111.2.3.4(A.
X4)=MIN r1.2.3.4(A.
4=
14
Finally, for the total line length we can state
Cl.2.3.4.S(A*
O)=MIN
[(1'1.2.3.4(A1
x4)+r5(x4,
O)l=
=min(14+5, 16+4, 18+4)=19
X,=
L,
M,
N
.
The minimum construction cost is determined
by
ACEHLO,
the solid line in the
Figure, the magnitude of which is 19 units.
It can occur with the trace, determined by the above method, that the pipe length
is longer, or/and due to the level differences of the pipelines the demand for specific
pumping power is greater than in case
of
other trace variants ofgreater construction
costs. Such acase may occur first
of
all, ifon the terrain, between the two end points,
the level difference is very significant, or terrain spots
of
very high laying costs are
present. Then, in order to achieve the determination
of
the optimum trace, further
considerations, and calculations are needed.
The design method, discussed above, among the prevailing Hungarian
conditions, decreases the construction cost of the pipeline, by
1.7
-
7.6
percent as
compared
to
the traditional design method.
198
7.
PIPELINE
TRANSPORTATION
OF
OIL
(e) Selection of centrifugal
pumps
For transportation of oil, generally horizontal axis multistage centrifugal pumps
of radial double suction, and radial discharge are applied. The structure, and
operation characteristics of the centrifugal pumps are discussed
in
several manuals.
In this Chapter, only the peculiarities are discussed that are
of
outstanding
importance, regarding the transportation of oil.
The characteristic diagrams of the operation of the centrifugal pumps, given by
the manufacturers, are determined by using water. Curve
H
in
Fig.
7.4
-
Y
is the
so-
t
Fig.
7.4
-9.
Characteristic curves
of
centrifugal pump
called throttling curve representing the change in the (manometric) head,
H
=
Ap/pg
in the function of the transported water flow rate,
q.
q
is the overall efficiency of the
pumps,
q
=
r],qhqu,
where
q,
comes from the friction of the seals, and bearings, and
from the so-called “disc-friction” of the impellers rotating in the space, filled with
liquid;
qh
is the hydraulic efficiency depending on the head losses of the fluid stream
due to friction hitting and the change in the flow direction;
qc.
is the volumetric
efficiency. Curve
P
gives the change in the power requirement of the pump, v.
pumping rate.
The curves may be determined also by in situ tests, measurements. Among the
operation parameters of the pump, the following relations may be determined:
Thus, we may conclude the
-=-
42
=const
JH,JH,
41
7.4-23a, b
7.4
-
24
The efficiency curve has got its maximum. The pump should be selected
so
that its
operating point would coincide with this best efficiency point
(BEP),
correspond-
ing to
%lax.
7.4.
ISOTHERMAL
OIL
TRANSPORT
199
By
the affinity parabolas, expressing
Eq.
7.4-24
in a graphic way,
it
can be
determined on the one hand the congruent, head curve, at
n,
speed, of the pump
where the original head curve was determined for
n,
speed, on the other hand,
it
can
be constructed at what speed the given
q-
H
operating point may be realized. On
Fig.
7.4-
10,
the head curve, represented by solid line is valid for the speed
n,.
Knowing the
BEP
(a) determined by
q,
,
and
H,,
the affinity parabola, intersecting
it
may be determined, and likewise the affinity parabolas, crossing points hand
c
may
I
4,24,1
4
Fig.
7
4
-
10.
be also plotted. Points of the head curve, that are valid at the required
n,
speed may
be obtained
so,
that we read
off
the intersection point coordinates
qal,
Ha,
belonging to
a
affinity parabola and speed
n,
head curve and from the relation,
derived from Eq.
7.4
-
23
we calculate the flow rate
q,,,
whose corresponding affinity parabola point (a')
will
be, one point of the
n,
head curve.
In
the same way, the other points of the new head
curve, with the help of the affinity parabolas, may be plotted.
If, however, the
n,
speed is sought, by which an operating point. defined by the
qa2
and
Ha,
values, may be achieved with the maximum possible efficiency, then
supposed the coordinates of
BEP
are known at
qal
from Eq.
7.4-23
it
can be
calculated that
9.1
4.2
n,=
-n,.
The application ranges of the centrifugal pumps are well shown by the so-called
iso-efficiency diagram
(Fig.
7.4
-
I
I).
Beside the head curves, corresponding to
different speeds (thicker solid line), some curves, indicating iso-eficiency (thinner
solid line) are also shown.