Investigation of effect of cross-sectional shape of low-level outlet on the flushed sediments rate in the pressure flushing

Sediment related problems are of huge importance in most projects of dam construction and for reservoir management. Worldwide average annual reservoir storage loss due to sedimentation is about 1.0 percent. In order to reduce and remove reservoir sedimentation for sustainable use, different sediment control measures have been developed and used. Some of the more popular methods are watershed management, sluicing, dredging, density current venting, bypassing, and flushing. One of the most effective techniques for removing the deposited sediments from reservoirs is pressure flushing which has less local effects. In flushing methods, the previously deposited sediment would be flushed from the reservoir by opening of the bottom outlets. When the flushing takes place under a sustained water level, only a very limited area of the reservoir is flushed and a scour cone is performed behind the outlet. As the flow around the outlet in the pressure flushing is three dimensional.
The flushing process can be studied in a physical model. This method is, however, relatively costly and time-consuming. Researches for evaluation of geometric characteristics of scouring cone against various shapes of the bottom outlet are necessary, in order to proper design of the bottom outlet. In this study, some experiments performed for investigation effective factors in pressure flushing performance. Experiments were carried out with three outflow discharges (1, 2 and 3 l/s), two water depths above the center of bottom outlet (47 and 64.5 cm), and three bottom outlet shapes (circular, square and rectangular).The experiments were conducted in the hydraulic laboratory of Tarbiat Modares University in Iran. Experiments tests carried out with a flume whose overall dimensions consist of 7.1 m length, 2 m width and 1.5 m height. The front wall of the model will be easy to change to modify different cross sections of reservoir bottom outlets. The sediment height deposits in the main reservoir was 20 cm, with a median diameter of d50=1.15 mm. For the downstream section it was used another stilling basin, which the mixing flow of water and sediment was collected in it. The downstream settling basin was a rectangular flume of 1.4 meter long, 1 meter wide, and 0.8 meter height. At the end of settling reservoir, there was a V-notch weir with an angle of 600 to measure of outflow discharge. For running the experiments, the deposited sediment was flattened and leveled firstly to a specific level above the center of bottom outlet (20 cm), and the model was slowly filled with water until the water surface elevation reached to a desired level. Then, the bottom outlet was manually opened until the outflow discharge becomes equal to the inflow discharge. The equilibrium of scour cone volume and length depends on the reservoir water depth , depth of deposited sediment above the center of outlet , fluid density , sediment density , outlet area , water velocity at bottom outlet , mean deposited sediment diameter , diameter of circle outlet and width of rectangular and square outlets , dam width and gravity acceleration . Therefore, in pressure flushing, scour cone volume may be written as a function of the following variables: .
The pressure flushing method has local effects in sediment disposal and is recommended when local disposal of the sediment deposits is intended. The results of these experiments show that the scouring cone volume depend on outflow discharge, water depth of the reservoir and bottom outlet shape. As for a constant outflow discharge, the decreasing of the water depth leads to increace the scouring cone volume. Also the increasing of the outflow discharge has a positive effect on the flushing cone volume. Scouring cone volume increases 41.3, 41.2 and 19.9 percent with the outflow discharge rise of 150 percent in circular, square and rectangular valves, respectively. Moreover, generally for a constant water depth and outflow discharge, scouring cone volume was greater in square, circular and rectangular outlet shapes, respectively. Finally, based on the statistical analysis on the experimental data, non-dimensional equation is offered for forecasting of the scouring cone dimension. For evaluating the accuracy of the proposed equation, three statistical criteria, including root mean square error (RMSE), mean absolute error (MAE) and R-squared value were used. The comparison between measured and computed scour volume using proposed equations show satisfied agreement. Hopefully, we can share our knowledge to design projects that can sustain the useful lives of existing and future reservoirs.