Szilas A.P. Production and transport of oil and gas, Gathering and Transportation
Подождите немного. Документ загружается.
260
7
PIPELINE TRANSPORTATION
OF
OIL
'I
1
2
1
tail ends
4.
From power
unit
5,
alternate current is led into the system. The current
through the cable flows into the heating pipe wall, and the circuit is closed through
the wire
6. The interference of the electromagnetic fields, surrounding the cable and
heater line, results in a so-called skin effect. Due to this, the electric current,
conducted by the heater pipe, propagates through a tight annular cross section.
Inner diameter of
it
is equal to the inner diameter of heater pipe; the outer diameter
is only
0.1
mm larger. Other parts
of
the pipe, such as the external cylindrical
112
3
14
314
314-
I
Table 7.6
-
3.
Compatible sizes
of
oil pipe
and
of
steam pipe used
to
heat
it
(after Pektyemirov 1951)
Oil pipe size
in,
2-3
4-5
6
8
number size, in.
surface, are free of electricity. In the small current conducting cross-sections
considerable heat develops, appearing first ofall
in
the wall ofthe pipe, that, through
the welds. propagates to the external wall of the oil-line, and to its content. In order
to achieve the required heating, the necessary voltage drop is
300-700
V/km, and
the amount of the current is 50-200A/km. The transmitted heat power is
15
-
20 W/m (Ando and Kawahara
1976).
The pipeline is schematically shown
in
Fig.
7.6
-29.
The oil transport line
I
and cable
2,
surrounding heating line
3
are covered
by a common heat insulating polyurethane coating, that, in order to protect against
water and mechanical damages, is coated by polyethylene pipe
5.
The pipeline, due
to the oil flow, and because of the high voltage power transmission line
in
the
vicinity, may de slightly tilled up with some static electricity, that is why the system
is
grounded (see
7
in
Fig.
7.6
-28).
According to Ando and Kawahara, the actually
5
4yj
i+&i7L
1
-
-
-
-
-
-
-
-
----
_-___-_
7
7
2
Fig
7 6
-
28 Scheme
of
the electnc outfits
of
the
SECT system, after Ando and Kawahara (1976)
Fig
7
6
-
29 The structure
of
the SECT
p~pe
line,
after Ando and Kawdhara (1976)
7.7.
METHODS
OF
IMPROVING
FLOW
CHARACTERISTICS
26
1
measured voltage difference between the external pipe wall and the soil is only
100
-200
mV, and through the grounding wire a current of 10-200mA passes. The
mounting and replacement of the cable without damaging the heat insulation, and
later without disturbing the soil, may be realized by applying the "pull-boxes",
6,
as
shown
in
Fig.
7.6-29.
The SECT system proved to be economical. Its economy, as compared to the
steam heating, increases with the pipeline's length, where
it
is to be applied. In
Japan, the heating cost is merely 30 percent of the cost of steam heating for
3700
m
length. System SECT may be also applied for heating oil field flow lines and
transport lines carrying high pour point crude. Myers (1978) furnishes technical and
economic data about field trials, where, beside the SECT, other electric heating
methods were also used. SECT proved to be the more economical
in
each case,
if
the
oil pipe diameter exceeded
3
inches, and at each diameter, where the length of the
pipeline exceeded 1650m, respectively. In Europe, an oil pipeline, heated by the
SECT system, was built first in 1979. This pipeline of 24 inches diameter transports
oil from a port on the Adriatic see to a refinery of4.7 km distance(Ga1ati
et
al.
1979).
Advantages of the SECT system are:
(i)
below 70°C, heating cost is usually
smaller than at any other methods;
(ii)
it
works automatically fulfilling precisely the
temperature prescriptions either while heating up gelled oil
or
at continuous oil
transport;
(iii)
it
requires practically no maintenance.
7.7.
Methods
of
improving
flow
characteristics
7.7.1.
Heat treatment
Flow behaviours of thixotropic, pseudoplastic crudes are discussed in Section
1.3.1. To improve the flowing behaviour of these crudes heat treatment uses the
feature, that the macrostructure of the solidified paraffins can be very different
depending on the rate of cooling.
At
relatively high temperatures, 80- 90 "C, all the
paraffins are in solution.
If
the subsequent cooling is rapid several, independent
paraffin crystals develop, that, at a terminal temperature of about 20 "C, usually do
not unite to form a connected space lattice. The flow properties
of
this dispersed
system may be much more favourable, than that of the oil, in which the paraffin, due
to the slow cooling, forms a more connected space lattice. The macrostructure of the
paraffin is also influenced by the fact, whether the oil, while cooling, is in rest
or
it
is
moving. Laboratory experiments show, that, due to the mixing, during the cool-
down process, paraffin of less connected space lattice, i.e. oil of more favourable flow
behaviour develops.
While cooling the character
of
the developed disperse system is significantly
influenced by the malthene content
of
the oil too. This waxy oil ingredient consists of
heteropolar molecules
of
rather varied composition, which adhering to the
dissolved paraffin crystals, prevent their further growth, and hinder the develop-
ment
of
connected space lattices. Their impact is influenced by their chemical
262
7.
PIPELINE TRANSPORTATION
OF
OIL
7;
N
77
40-
30
composition and amount. Furthermore they exert a very characteristic influence on
the flow properties of the crude during a "reheating and cooling" procedure. This
will be further discussed below.
The
Fig.
7.7-1
shows the results of the experiments, performed at the
Department of Petroleum Engineering
of
the Miskolc Technical University
for
Heavy Industry (Hungary). The crude from Algyo field (Southern Hungary)
of
20
"C
'
PI
I
I
I
20
30 40 50 60
70
80
Initial
temperature
of
heat treatrnenl
T
"C
Fig.
7.7-
1.
Influence
of
starting temperature
of
heat treatment upon apparent yield stress, after Szilas
ef
al.
(
1978)
temperature originally was treated at first in standstill by heating to various
temperatures, then cooling back to
20
"C
at a constant rate, and thereafter the static
shear stress was measured. The Figure shows that the flow behaviour, characterized
by the above mentioned parameter, significantly changes with the end-temperature
of the re-heating. It will significantly worsen before reaching the value of
50
"c;
at
higher temperatures improvement can be seen, but only at a temperature of
70
"C
we attain the flow properties of the original untreated
oil.
In case of re-heating to
80
-85
"C,
the static shear stress was lower, as compared to the original, untreated
value.
Similar result is obtained, if, for the characterizing of the flow properties other
rheological parameters, e.g. the apparent viscosity, corresponding to some given
7.7.
METHODS
OF
IMPROVING
FLOW
CHARACTERISTICS
263
Name
of
crude
Szank
I
Szank
II
Tizlir
Kiskunhalas
I
Kiskunhalas
II
shear rate, are selected. The so-called
critical reheating peak temperature,
offering the
worst flow properties, may change from crude by crude. In the course of our
experiments we found
50
“C
to
be
the most frequent, but occasionally significantly
higher values, such as
70
“C, were also measured. This phenomenon appears only in
the case,
if,
beside the prafin, also resinuous components exist in the crude. The
probable explanation for this may be, that the malthene molecules, as described
above, join the paraffin crystals only in the course
of
a process of cooling down from
a relatively high temperature. This occurs at the first cooling during the production
process too. On reheating, the relatively light homologues of the composite crystals
of the paraffin melt, and the associated malthene molecules liberate again. These,
however, due to the previous binding, have lost their activity, and can regain it only
at a relatively high temperature. The inactive malthene molecules are, in the course
of cooling from relatively small reheating temperature, unable to prevent the
dissolved paraffin molecules from joining the existing space lattice fractures, and
thus a space lattice, more connected than the previous one, and an oil of worse flow
properties develops.
Table
7.7
-
I
presents the results of laboratory experiments,
where the sensitivity of different Hungarian crudes for heat treatment was
examined.
For
characterizing the flow behaviour, the
p,
relative apparent viscosity,
Slow cooling Slow cooling Fast cooling Fast cooling
without mixing with mixing
without mixing with mixing
pa,
P
p,h
Pm
P.h
P.0
Prh
P.“
Prb
1
2
3
4
5
6
7
8
13.8
1.04
0.77 0.79
1.03
1.09
0.70 0.78
0.14
242 0.91
1.86
0.86
2.14 0.80
1.36
31.6
1.42
092
1.10
0.94 0.97 0.52
0.86
0.77
5.31
1.00
3.84
1.24
4.84
1.00
4.21
3.05
1.33
1.00
1.00
0.43
0.94
0.33
1.22
Table
7.7-
I.
Effect
of
heat treatment on several Hungarian crudes
determined at
10
“C, and at
D=
145.5
s-
shear rate, was selected. The adjective
“relative” here means, that the apparent viscosity values are, in each case related in
every line to the
pa
values
of
column
1,
measured after a slow cooling without
mixing. The value for slow cooling was
095
“C/min, while the same value for rapid
cooling
2.20
“C/min.
pro
denotes the “relative” value, achieved by continuous
cooling, while
prb
gives the relative value obtained after a “reheating to
50
“C and
cooling” treatment. The impact of the rapid, and slow cooling, and that of the re-
heating can be seen. Re-heating generally worsens the flow properties. The best flow
properties develop by “rapid cooling and mixing” (column
7),
while the worst ones
by “slow cooling, without mixing” (column
2).
Being aware of the fact that the flow properties
of
the crude, due to the
temperature, and shear history, may significantly change, is
of
great importance.
264
7
PIPELINE TRANSPORTATION
OF
OIL
This knowledge may
be
directly used for industrial improvement of flow properties.
Besides,
it
indicates, that also independently of our intentions, in the course
of
production, transport, and storage, the flow behaviour may change. Rheological
measurements, the results of which may serve as a basis of the calculation of pressure
loss,
may be carried out only with representative samples, where the temperature
and shear history is the same as that of the oil to be transported.
Both from the points of view of the laboratory-scale history simulation, and
the
heat treatment, performed in order to improve the flow properties,
it
is important to
determine the minimum and maximum efficient values for cooling speed. Maximum
efficient cooling speed, expressed
in
"C/min units, is the limit value, above which
the
flow properties cannot be improved any more, while the minimum cooling speed is
the value under which the flow properties already may not be worse. The above
cooling speed values, occurring in practice range from
1
to 3"C/min.
In
the
knowledge of these limits, on the onc hand, the time, required
for
the laboratory
simulation of the flow history may be shortened, and, on the other hand, while
designing the heat treatment process, the cooling speed, that enables us to achieve
the most favourable flow properties by using the minimum cooling energy, may be
determined.
By means of calculation
it
is generally impossible to determine the flow properties
of a mixture of different thixotropic pseudoplastic crudes from the parameters of the
basic oils. Beside the chemical composition
of
the components, the rate of their heat
treatment. the temperatures of the components, and the difference in their
temperatures before mixing play also an important role. The expectable flow
properties may only be determined by measuring samples prepared by laboratory
simulation of real shear and temperature history and mixing circumstances.
In
certain cases simulation may be even more difficult, since the physico-chemical
properties
of
the sample, during stopping
or
storage, may change. This
phenomenon is called "aging", and is assumed to be related to changes
in
the
chemical composition.
The first commercial application
of
heat treatment took place at Nahorkatiya
and Moran, in India
(Oil
and
Gas
Journal
1963).
Even today these are the largest
heater-treater stations, described
in
literature, and they are modernized continuous-
ly
(Chandrasekharan and Sikdar
1970).
The main parameters of the production and
transport of the two oilfields, and the Nahorkatiya plant will
be
discussed as follows.
At
Nahorkatiya, the crude whose pour point ranges from
29
to
34
"C
and whose
paraffin content is
15.4
percent, is transported through a
16"
pipeline of
402
km
length to a refinery where part of
it
is refined; the rest is transported through a
14'
line to another refinery at a distance
of
765
km from the first one. The first
line
segment was designed for a yearly throughput of
2.75
Tg, the second
for
a
throughput of
2.0
Tg year. The lowest soil temperature at pipe depth
1.2
-
1.8
m
is
18
"C. Heat treatment is required between October and April.
It
entails a significant
improvement in flow behaviour. Typically, at
18
"C temperature and a shear rate of
13
s-
I,
the apparent viscosity is
06
Pas in untreated oil, and
0.1
Pas in treated oil.
7.7.
METHODS
OF
IMPROVING
FLOW
<'t1ARA('TI1RISTI~'S
265
2
3
Buffer tonh
Conditioned crude
tonk
4
Exchanger
5
Refrigeration unit
6
cooling
tower
7
Makeup water
constont
8
Selas
heater
9
Steam drum
10
Balance
rank
tCli7k
Mokeup
m
Conditioned crude
to boosters
7'
I
r
Suppltes
to treaters
lntermediote woter
hot
mter
Wokeup-water
Cooling
-
water
..
,I
-
1
3
1
-1
Slops
to tank
Crude
oil
from
storage
tank
___e_
1
I
Crude
oil
w
1"h
Fig.
7.7-
2.
Nahorkatiya crude heat treater station (Chandrasekharan and Sikdar
1970)
266
7.
PIPELINE TRANSPORTATION
OF
011
The Nahorkatiya crude heat treater plant is shown as
Fig.
7.7-2
after
Chandrasekharan and Sikdar (1970). The crude is heated in gas-fired heat-
exchangers to between 90 and
95
“C.
Subsequent cooling to 65
“C
is effected
in
heat
exchangers, using crude to be preheated as the coolant. In the next step, the static
crude is chilled with water to 18
“C.
Each of the 36 treater tanks is an upright
cylindrical tank
6.1
m high, and
of
9.2 m of diameter; each tank contains 127 one-
inch chilling tubes. It is the cooling water that flows
in
the tubes, and the crude that
occupies the space between them. Stationary cooling takes place
in
the following
steps. The empty tank is filled with crude, which is chilled by the circulation of cold
water at a rate ofO.6 “C/min. After cooling to 18
“C,
the crude is pumped to a storage
tank, and the tubes are used to circulate hot water, which again rises the tank
temperature to 65
“C.
This melts the deposited paraffin and prepares the tank for the
reception
of
another batch of crude. These heating and cooling operations are
performed by means of five water circiuts, each separate from the others.
(i)
High-
pressure hot water heated with gas
to
145
OC
and held under sufficient pressure
to
keep
it
below its bubble point is used to heat low-pressure hot water and to
regenerate the concentrated lithium bromide coolant of the absorption cooler.
(ii)
Low-pressure hot water is used
to
heat the treaters
to
65
“C
after the chilling period.
(iii)
“Intermediate” water is used to cool oil in the first phase ofcooling.
It
is kept at
a
temperature of
30
“C
by “cooling” water cooled in a cooling tower.
(iv)
Cold water is
used
to
chill the crude to 18 “C in the treater tanks. This water is cooled to 14
“C
in
an absorption refrigerator.
(v)
Cooling water, circulated
in
an open circuit including
a cooling tower,
is
used to cool the intermediate water. The 324 valves of the treating
plant are controlled by an electric Flex-o-Timer. The water circuits are controlled
by a programme controller which adjusts the required temperatures. The
procedure, because of requiring a great investment, is rather expensive. That is
why
in India now a cheaper solution is being designed that provides continuous heat
treatment during the flow (Mukhopadhyay 1979).
7.7.2.
Solvent addition
The mixing together of two crudes, or products, comparatively low
in
malthenes,
differring from each other in flow behaviour, will result
in
a mixture whose
flow
properties will usually fall between the properties of the two components.
It
is
possible to improve, by this means, the flow behaviour
of
high-paraffin,
high
viscosity gelling crudes, e.g. by the addition of lower viscosity crude, or a liquid
product, such as gazoline, kerosene, or diesel fuel. Attempts have been made at
predicting the behaviour of mixtures starting from those of the components.
No
laws of general scope have been derived as yet, but important research was carried
out
to
determine empirical relationships (Aliev
er
al.
1969).
The effect
of
the solvent upon the continuous transport can be determined by
measuring the flow curves of high viscosity crudes and those of the mixture at
various temperatures, and from the results, in the way, described
in
Section
1.3.5,
the
change
of
the pressure gradient v. throughput at different temperatures is to be
7.7.
METHODS
OF
IMPROVING
FLOW
CHARACTERISTICS
267
0
100
ZM)
300
400
500
D.
s-’
Fig.
7.7-
3.
The elTect
of
condensate on the
flow
curves
of
Algyo-crude, after Szilas and Navratil(1978,
1981)
grad
P
t
0
50
100
150
200
250
q,
m3/h
Fig.
7.7-4.
Isothermal
flow
gradient in a
di=
0.3
pipe line
for
oil
flow
curves characterized by Fig. 7.7
-
3
(Szilas and Navratil 1978,
1981)
268
7.
PIPELINE TRANSPORTATION
OF
OIL
A A
A
q,
=I,O
10~m'/year
,,m3
1
qa=1.3 lo6 m'lyear
,/m3
j
q,
=
1.5
lo6 m'iyear
,/m3
I
0
3
6
91215
0
3
6 9
1215
0
3
6 9
1215
Tt
7
"C
Tt
1
'C
Tt
3
'C
Fig.
7.7-5.
Specific energy demand
of
non
isothermal. solvent using
oil
transport under different
operating conditions. after Szilas and Navratil
(1978.
1981)
calculated.
Figure
7.7-3 shows the steady state flow curves of an Algyo crude,
mixed with natural condensate at temperatures of
0
and
15
"C.
Fig.
7.7-4
shows the flow pressure gradients of the same fluids while passing
through a pipeline of0.30 m
ID.
It
can be seen that the flow curves of the basic crude,
and that of the mixture are of pseudoplastic character. Due to adding the
condensates the friction pressure
loss
is reduced. The rate of this reduction
in
case of
crudes of lower temperature, i.e.
of
worse flow property, is greater.
The mixing of the additives to the basic crude will cause upon the flow pressure
drop two effects of opposite character. On the one hand,
it
increases the flow rate
and, therefore, in case of unchanged other parameters, the flow pressure
loss
increases too; on the other hand,
it
creates better flow properties by which this
loss
is
reduced. Obviously, there is
a
given rate of the solvent at which the decrease
in
the
flow pressure
loss
is the greatest. Values below and above this rate increase the flow
pressure
loss.
The summarized effect of the influencing parameters are well shown
by
Fig.
7.7-5. The specific energy demand, after Szilas and Navratil (1978, 1981)
was calculated for a pipeline of 162 km length and
0.305
m
diameter, from formula
where
Ap
is the flowing pressure
loss;
qm
is the volumetric rate of the mixture, while
qo
is that
of
the original crude; is the efficiency of the centrifugal pump.
It
was
7.7.
METHODS
OF
IMPROVING
FLOW
CHARACTERISTICS
269
assumed that the pipeline was laid horizontally, with an axis
1.2
m below the
surface. The head end temperature of the oil, in each case, was
42
"C,
the thermal
conductivity factor of the soil
2.0
W/(m
K),
and the allowable maximum pressure
loss
in
the pipeline was
58
bars. It was also assumed that the maximum time
exploitation factor was
096.
The six parts of the Figure show the specific energy
demand of the stabilized, non-isothermic flow in the function
of
soil temperature at
different yearly liquid throughputs. In each Figure, the flow energy demand of five
different flow behaviour oils are plotted by using the flow curves of
Fig.
7.7-3.
The
numerals beside the curves show the percentage of the solvent v, the transported
crude. According to
Fig.
7.7-5
(a),
at a relatively small yeary throughput, the
specific energy demand, at any temperature is the smaller, the greater is the rate of
solvent. In case of greater throughputs, however, in Figs
(h)
-
(e),
an inversion
Sovoly
crude
0
5
10
15 20 25
30
35
40
1,
oc
Fig.
7.7-6.
Effect
of
the solvent addition on the Savoly-crude's apparent yield stress