|
|
||
 intermediate level in the pool. The researchers indicated that an optimum ratio of pumped to release
rate was observed for each series of tests.
Holland (1984) developed computational methods to predict the depth of penetration of a
hydraulic jet as well as the entrained flow that crosses the thermocline. In addition, a method for
designing a localized mixing system was developed (Holland 1983). Brown, Mobley, and Nubbe
(1988) formulated a one-dimensional numerical model of a hydraulic jet in a withdrawal zone of the
hydropower project at Douglas Dam. Field measurements of velocity versus depth for a large diameter
plume from the surface water pump installation at Cherokee Dam are reported by Tyson and Mobley
(1996). A two dimensional numerical model using PHOENICS code was developed to optimize the
dissolved oxygen improvement for a given thermal stratification without having the pump plume
disturbing bottom sediments (Hadjerioua 1996).
In 1987, a series of localized mixing tests were conducted at J. Percy Priest Reservoir in
Tennessee (Price 1988, Sneed 1988). These tests were designed to investigate the feasibility of direct-
drive mixers to improve release quality from the hydropower project. Results of these tests indicated
that three pumps generating 45 cfs each improved the release DO by 1 mg/L (release discharge of
4,600 cfs). These tests also demonstrated that the location of the jet relative to the intake structure had
a significant impact on the efficiency of the localized mixing application.
Similar tests were conducted at Douglas Reservoir by the Tennessee Valley Authority (Mobley
and Harshbarger 1987). These tests, using three axial flow pumps, each with a propeller diameter of
4.5 m, indicated that the hydropower release DO could be improved by as much as 2.0 mg/L (Brown,
Mobley, and Nubbe 1988). The results of these tests lead to further testing of commercially available
mixing equipment for this application (Mobley, 1990), and eventually permanent full scale installations at
TVA's Douglas Dam in 1994 (Mobley et al 1995), and Cherokee Dam in 1995 (Tyson and Mobley
1996) . The permanent installations at each dam cost approximately $2M and employ 9 surface water
pumps to improve the release DO of 4 hydroturbine units by 1.5 to 3.0 mg/L depending on reservoir
conditions. The surface water pumps at Douglas Dam have a 15 foot (4.5 m) diameter stainless steel
impeller driven by a 40 HP electric motor. The surface water pumps at Cherokee Dam have
composite impellers that are 16 feet in diameter and are driven by 75 HP electric motors. At each dam
the pumps are designed to allow for up to 65 feet of headwater fluctuation. TVA has obtained good
results from these installations and has found maintenance requirements for the electromechanical
equipment and debris to be significant but reasonable compared with the other aeration alternatives
used at each dam.
4.6.2.5 Summary
Localized mixing is a technique that is being successfully used to improve the release quality
from reservoirs. This technique requires the use of a mechanical pump to jet surface water down into
the hypolimnion with enough energy to penetrate to the center-line elevation of the intake port. A
4.6-10
|
||
|
||
