|
|
PHYSICAL
MODEL STUDY FOR THE INTAKE CHANNEL AT THE SUBIYA POWER STATION, KUWAIT
JEN-MEN
LO and A. AL-SALEM
Hydraulics
and Coastal Eng. Dept., Environmental and Earth Sciences Division
Kuwait
Institute for Scientific Research, P.O. Box 24885
13109-Safat-Kuwait
ABSTRACT
To
identify and correct the objectionable features of the entrance shapes
suggested by the conceptual design, three proposed arrangements for the intake
channel of the Subiya Power Station in Kuwait were studied using physical
models. The intake channel was designed
to carry a maximum water supply of 148 m3/s with a maximum velocity
of 0.16 m/s associated with the minimum design water level. This velocity value was calculated so that
no resuspension of sediment already settled in the channel would occur and to
avoid scouring of the channel banks.
The model was designed according to the Froude similarity criterion as
an undistorted model with a linear scale of 1:80. The model was constructed to
simulate an area of 2460 x 1250 m. This
area was found to be sufficient to run the model with a negligible boundary
effect. The model test was carried out
in the Hydraulics and Coastal Engineering Laboratory at the Kuwait Institute
for Scientific Research. The main
objective of the model test was to attain uniform velocity distribution inside
the channel and to minimize separation of flow at the channel entrance. Based on the model study, it was concluded
that the proposed bell-mouth intake channel should be modified by constructing
14 guide vanes at the entrance section to meet the required flow and sediment
conditions.
Keywords: Hydraulic structure,
Intake channel entrance, Sediment deposition, Cooling water, Tidal current.
The
Ministry of Electricity and Water, Government of Kuwait, is planning to
construct a large complex of steam power and desalination plants on the Subiya
Peninsula (Dames and Moore, 1983). Fig.
1 shows the location of the proposed Subiya Power Station. It is intended that
the construction of the Subiya Power Station be executed in two stages. During the first stage, a power capacity of
2400 MW and a capacity for producing drinking water of 12 MIGD will be
installed. The total cooling water
demand for full-load operation during this stage will be 100 m3/s. During the second construction stage, the
drinking water production will be increased to 96 MIGD. Accordingly, the final cooling water demand
will be 148 m3/s. The
cooling supply is from the seawater at Khor Al-Subiya (Fig. 1). The wastewater will be discharged into
Kuwait Bay (Fig. 1).
Fig. 1. Location Map.
In the
conceptual design, three intake arrangements were proposed to convey seawater
from the Khor Al-Subiya to the power plant. Fig. 2 shows the three proposed
intake arrangements. They were
Fig. 2. Three proposed arrangements for the intake channel.
1. A channel of 160-m
bed width and 1:5-side slope guided by two straight dams extending across the
Khor Al-Subiya to a contour of -4.0 m Kuwait land datum (KLD). The bed level of the channel is -6.5 m KLD,
and the top elevation of the guide dams is +6.0 m. This was considered the basic solution.
2. A channel similar
to the basic solution, but with a protruding southern guiding dam to minimize
excessive separation of flow at the channel entrance.
3. A channel with a
bell-mouth entrance that may allow for streamlined flow at the entrance and a
more uniform flow inside the channel.
To
identify and correct the objectionable features of the entrance shapes
suggested in the conceptual design, the three proposed arrangements for the
intake channel of the Subiya Power Station were studied using undistorted
physical models. The intake channel was
planned to carry a maximum discharge of 148 m3/s with a maximum
velocity of 0.16 m/s associated with the minimum design water level. The velocity was calculated so that no
resuspension of sediment already settled in the channel would occur and to
avoid scouring of the channel banks.
Four flow conditions were found to be critical for the hydraulic design
of the channel entrance: the minimum water level of -1.5 m KLD associated with
slack current, the maximum water level of +4.2 m KLD associated with slack
current, and the mean water level of +2.0 m KLD associated with flood and ebb
currents (Abou-Seida and Lo, 1987). The main objectives of the model test were
to reach uniform velocity distribution inside the channel and to minimize
separation of flow at the channel entrance.
The model
was designed according to the Froude similarity criterion as an undistorted
model with a linear scale of 1:80. This
scale was chosen to avoid the effects of surface tension and to ensure that the
flow in the model was turbulent. The
Froude similarity gave the following relationships: model length to prototype
length = 1:80, model velocity to prototype velocity = 1:8.94, and model water
discharge to prototype discharge = 1:57243.34.
The model
was constructed to simulate an area of 2460 x 1250 m. This area was found to be sufficient to run the model with a
negligible boundary effect. The model
was constructed in the Hydraulics and Coastal Engineering Laboratory at the
Kuwait Institute for Scientific Research (KISR) on a steel platform using
lightweight material (styrocrete) covered with a 2-cm cement mortar apron. The supply and drainage systems were
designed to simulate flood and ebb conditions; thus the flow ran from one side
of the Khor to the drain at the other side for flood cases and vice versa for
ebb cases. The water was supplied
through a perforated pipe and flowed over an adjustable weir to the drainage
channel. Plastic perforated boxes were
arranged in front of the pipe to dissipate the excess flow energy and to help
attain the required velocity distribution in the Khor. The intake channel was constructed across
the Khor, and the intake flow was discharged through four pipes by gravity and
adjusted by regulating valves. Dye and
floats were used for velocity measurements.
Fig. 3 shows the general arrangement of the model.
Fig. 3. General arrangement of the model.
A base
test was run for an intake channel 1150 m in length with straight guide dams,
and it was found that the most critical flow conditions were at a water level
of +2.0 m KLD with flood and ebb currents.
Backflow was noticed, and the nonuniformity of the velocity distribution
inside the channel was clear.
The three
proposed entrance conditions shown in Fig. 2a, b, and c were then tested for
the +2.0-m KLD water level for flood and ebb cases. For flood and ebb cases in
the first design (Fig. 2a), the flow was nonuniform, and backflow was observed
along a great portion of the channel length.
For the intake channel shown in Fig. 2b, the flow conditions were still
unacceptable, and the nonuniformity of flow persisted in both the flood and ebb
conditions. When the bell-mouth
entrance (Fig. 2c) was used, a change in the flow behavior was noticed, and the
flow started to be more uniform. But,
separation at the entrance did not disappear, and backflow was noticed during
both the flood and ebb conditions.
Several measures were tried to attain a reasonable uniform distribution
of the velocity as well as keep its value within the acceptable range for no
resuspension of settled sediment. In
addition these measures should have minimized flow separation at the
entrance. The above requirements were
reached by dividing the passage of the flow at the bell-mouth entrance into
fifteen passages separated by fourteen piers each 14.4-m wide and 59-m long
extending up to the +6.0-m KLD level. Fig. 4 shows the final arrangement for
the proposed structures.
The velocity distribution was uniform and minimum
separation was observed. The
arrangement was then tested for a
+4.2-m KLD water level with slack
current and for a -1.5-m KLD water
level with slack current. The
experimental runs proved to be successful for these two cases. To prevent the oil slick movement, a series
of tests for the final arrangement with the installation of skimmer walls
between the guide vanes were also studied.
The lower level of the skimmer walls was 1.5 m below the low water level
(-1.5 m KLD), and the level is above the high water level (+4.2 m KLD). During the test, two different skimmer wall
arrangements were installed. For the
first arrangement, the skimmer walls were installed between the guide vanes,
and for the second arrangement, the skimmer walls were installed between the
guide vanes except at the openings of guide vane nos. 6, 7, 8, and 9. In general, with the installation of the
skimmer walls, the flow pattern inside the intake channel obtained a uniform
flow condition faster than in the case without skimmer walls.
It was
concluded from the model study that the proposed bell-mouth intake channel
should be modified by constructing 14 guide vanes at the entrance oriented at
different angles to the channel is longitudinal axis. The spaces between the guide vanes are not equal, but depend on
guide vanes' location at the entrance section.
Modifications at the tip of the southern guiding dam is required (Fig.
4).
Fig. 4. Final configuration of the intake channel with 14 vanes at the
entrance.
The "Physical Model Study for the Intake Channel at
the Subiya Power Station, Kuwait" was accomplished through the joint efforts of
the Ministry of Electricity and Water, Government of Kuwait, and the Kuwait
Institute for Scientific Research.
Abou-Seida, M. M. and Jen-Men Lo. 1987.
Data for hydraulic studies
(Appendix B).Hydraulics studies and environmental
impact assessment for the Subiya Power Station. Kuwait Institute for Scientific
Research, Report No. KISR2371, Kuwait.
Dames and Moore.
1983. Hydraulic studies,
Vols. II and III. Studies for Sabiya
area, Kuwait Bay, and development of electrical
networks. Ministry of Electricity and
Water, Kuwait, Government of Kuwait, Report No. MEW/CP/PGP/1113 80/81,
Kuwait.