Enhancement of flushed sediment in pressure flushing using inclined plates upstream of orifice Open Access

The construction of a dam on a river decelerates the water velocity, leading to sedimentation in the reservoir. Moreover, a turbidity current containing large quantities of sedimentary particles accelerates sedimentation. Sedimentation is recognized as the main cause for reducing the useful volume of reservoirs. Reservoir deterioration affects dam performance in flood control, power generation, and water storage for agricultural, industrial, and drinking purposes. As reservoirs are the major sources of water, watershed management methods are performed to prevent sedimentation and sediment transport into reservoirs. Several techniques, including flushing, have been introduced to evacuate sediments from reservoirs. Flushing evacuates sediments through the bottom outlet. Free flushing refers to the evacuation of sediments by discharging the entire stored water. Although it has a high degree of sediment evacuation, but it is not suitable for large dams due to full storage discharge. In pressure flushing, on the other hand, the water level of the reservoir remains unchanged while evacuating sediments through the bottom outlet (Shen 1999). However, pressure flushing has lower sediment evacuation than free flushing and cannot be used to restore a large portion of the storage capacity of a reservoir (Schleiss et al. 2016). The volume of flushed sediment and flow pattern upstream of the orifice in pressure flushing are functions of various parameters, e.g., the water depth over the outlet, outlet size, sediment type and size, and sediment level. There is a large body of research on pressure flushing in the literature (Shammaa et al. 2005; Bryant et al. 2008; Meshkati et al. 2009; Fathi-Moghadam et al. 2010; Emamgholizadeh & Fathi-Moghdam 2014; Powell & Khan 2015).

An increased level of sediments in the vicinity of the dam may occlude the bottom outlet or disturb the functioning of equipment in, for example, the power generation unit inlet. As a result, a solution to increase sediment evacuation and scour cone size near the dam in pressure flushing could be useful. Recently, several methods have been introduced to increase the scour cone dimensions and volume in pressure flushing. Madadi et al. (2016) used cylindrical piles upstream of the outlet and reported 3.5 times larger flushed sediment quantities than the reference test. Madadi et al. (2017) employed a protecting semicircular structure (PSC) upstream of the orifice and raised the flushed sediment quantity by up to 4.5 times compared to the reference test. Haghjouei et al. (2021) utilized a dendritic bottomless extended (DBE) structure upstream of the orifice and reported up to 10-fold sediment removal at a given time compared to the reference test. Beyvazpour et al. (2021) increased the flushed sediment quantity by up to 10 times using a triangular-section pile upstream of the orifice. Shahriari et al. (2022) placed a plate upstream of the orifice and enhanced sediment removal by 11.5 times compared to the reference test.

Sediment movement in open channel flow is due to boundary shear stress (Cooper & Tait 2010) and the transport of sediments depends on velocity distribution (Pu 2021). In pressure flushing, since the flow velocity in the reservoir is almost zero, sediment movement and transport occur in the vicinity of the bottom outlet due to the radial velocity generated due to flowing water through the orifice. Sediment movement in the area upstream of the orifice is mostly due to the generated vortices and radial streamlines. Its mechanism is different from what happens in open channel flow. Sediment movement and removal are performed in two stages in pressure flushing (Powell & Khan 2011). In the first stage, sediments are moved by the shear stress on the bed in the vicinity of the orifice. In a short time, a scour cone forms upstream of the orifice, with vortices appearing. The vortices increase the lift and removal of sediments through the outflow of the orifice. Increased vortex intensity near the orifice leads to further sediment removal through the orifice. The installation of barriers to the flow separates the streamlines and induces downstream turbulence (Melville 2008). Shahriari et al. (2022) employed a plate upstream of the orifice to increase the intensity of vortices and enhance the flushed sediment quantity. Bryant et al. (2008) reported that the flow pattern is radial upstream of the orifice, based on the inflow nature.

A review of the literature showed the importance of the pressure flushing method for saving water during the evacuation of accumulated sediments in comparison with free flow flushing. However, the amount of flushed sediments has always been a problem in this method. Use of plates in front of or around an orifice seems to substantially increase flow turbulence and vortices, enhancing the removal of sediments. So far, how and the effectiveness of this method has not been extensively investigated. Therefore, the present study utilized two inclined plates on the two sides of the outlet to raise turbulence. To evaluate the performance of the inclined plates, the effects of the plate installation distance, size, and inclination on turbulence, sediment removal, and scour cone around the bottom outlet (orifice) were explored.

The sediment level Hs was constant and equal to the lower edge of the orifice at the beginning of each test. In all tests, the water depth from the orifice center (H) was fixed at 49 cm. Sedimentary particles of a medium-size d₅₀ = 0.5 mm, a geometric standard deviation of 1.55, and a specific gravity of 2.65 were used. The tests were carried out at flow rates of Q = 4.17 and 8.34 L/s. The flow rate was measured using an electromagnetic flowmeter with a precision of ±0.5% of the measured discharge. The inclined plates were placed upstream of the orifice, as shown in Figure 1. Three sizes of plate width of 3.5, 4.9, and 6.3 cm (i.e. B/D = 0.5, 0.7, and 0.9) were applied. The radial distance of the inclined plates from the orifice was set to 8.5, 10.5, and 12.5 cm (i.e. R/D = 1.21, 1.50, and 1.79), with α = 30, 45, and 60 degrees. This study carried out a total of 56 tests, including 54 tests with inclined plates and 2 tests without inclined plates (as the reference tests). The maximum values of B and R were adjusted using primary tests such that the water level upstream of the orifice would remain unchanged when the inclined plates were installed with the drain valve fully open. Moreover, the maximum value of R was adjusted such that the inclined plates would induce local scouring, given that the velocity upstream of the orifice declined as the distance from the orifice increased (Bryant et al. 2008). The minimum value of α was set to avoid occlusion upon the placement of the inclined plates in front of the orifice. The upper level of the inclined plates adjusted to the upper level of the orifice. To determine the test time, a set of primary tests were performed by installing the inclined plates upstream of the orifice. It was found that the maximum time required for the predefined flushed volume was 4.5 h. Therefore, all the tests were carried out for 4.5 h. With this time, the dimensions of the flushing cone reached equilibrium conditions and no changes were observed.

In all the tests, the sediments were initially leveled, and water flowed into the flume at a low rate, with the drain valve closed. Once the water depth increased to 49 cm from the orifice center, the flow rate was adjusted using the control valve and flowmeter, opening the drain valve (Figure 2). The water depth was kept unchanged during the test using the drain valve. At the end, the pump was turned off, with the drain valve being closed. Then, the water content of the flume was discharged through the upstream drain and flowed into the main tank through the outlet pipe. Once the flume had been drained, the sediment level upstream of the orifice was measured using a laser meter (Leica, D510) with a precision of ±0.1 mm. The topography of the scour cone was plotted using the free SURFER 16 software, obtaining its volume and dimensions.

As α increased, the free space in front of the orifice rose, leading to a larger quantity of water flowing through the inclined plates at a larger velocity. Flow separation upstream of the orifice increased as the flow collided with the plate edges at a higher velocity, enlarging the low-pressure zone, enhancing the wake vortices, and increasing the flushed sediment quantity. At lower value of α, however, the free space in front of the orifice was smaller; hence, a portion of the flow was deflected to the space behind the inclined plates toward the orifice.

Furthermore, at a given α, a rise in the flow rate from 4.17 to 8.34 L/s increased the flushed sediment quantity. For example, the quantity of the flushed sediments for Q = 8.34 L/s was approximately nine times as high as that at Q = 4.17 L/s for since an increased flow rate would increase the outflow of the orifice. Figure 7(b) plots and Z/D versus α for B = 6.3 cm, R = 8.5 cm, and Q = 8.34 L/s. An increase in α from 30 to 60 degrees led to 22, 17, and 18% rises in Z/D, L/D, and W/D, respectively.

Table 2 represents the sensitivity analysis results. As can be seen, R/D had the greatest effect on the scour cone volume and size; a 1% increase in R/D led to a 3.08% reduction in the scour cone volume.

Table 2

Sensitivity analysis results

.	.	.	.	.
	2.67	0.81	0.90	1.13
	2.20	0.45	0.63	0.83
	−3.08	−1.05	−1.02	−2.00
	1.00	−0.05	0.43	0.54

.	.	.	.	.
	2.67	0.81	0.90	1.13
	2.20	0.45	0.63	0.83
	−3.08	−1.05	−1.02	−2.00
	1.00	−0.05	0.43	0.54

View Large

Furthermore, was found to be the second-most influential parameter. As R/D and influence the velocity at which the flow would collide with the inclined plates, it can be said that the outflow velocity of the orifice is the most important factor affecting the volume and dimensions of the scour cone. The third-most important parameter was found to be B/D, and had the lowest importance; it was observed that had almost no effect on the scour cone length.

All the experiments in this study showed that the main cause for removing the sediments upstream of the orifice is transferring momentum in the form of bed shear stress and vortices. For the open channel flow, only bed shear stress is the main cause of the sediment movements and the natural bedform (water-worked) affect the sediment transport (Cooper & Tait 2010; Pu et al. 2017; Pu 2021). The existence of a plate in the front of the orifice with suitable width and distance from the bottom outlet and as a result flow separation around it, significantly increases the vortices, so it can be concluded that in the pressure flushing process, the vortices are more important than the bed shear stress. However, 3D velocity measurements can prove this hypothesis.

Pressure flushing is not very effective to restore the storage capacity of a reservoir. It is used as a turbidity current discharge method in dam reservoirs during floods (Schleiss et al. 2016). The volume of flushed sediment is low in pressure flushing, and the scour cone occupies only a small area upstream of the orifice. As mentioned, sediments would disturb the functions of equipment in the dam structure. Therefore, it could be useful to increase the quantity of the discharged sediments and increase the scour cone dimensions upstream of the dam bottom outlet. It was found that installing inclined plates led to 19% higher sediment removal and raised the width, depth, and length of the scour cone by 132, 234, and 200%, respectively. They also resulted in an 18-fold increase in the quantity of the flushed sediment at a given time (the initial 20% of the total time), reducing the discharged water during the pressure flushing operation. Inclined plates did not occlude the orifice; therefore, the bottom outlet would not be disturbed during floods and turbidity current. Furthermore, it was found that the velocity at which the outflow collided with the inclined plates and were found to have the highest and lowest effects on the scour cone volume/dimensions, respectively.

Installation of the inclined plates upstream of the dam can be done at the same time as the dam construction. One of the main advantages of this method is that no additional external forces are inserted into the dam body. Due to the separation phenomenon and generation of a low-pressure zone upstream of the orifice, the discharge coefficient increases and this process is very useful, especially for turbidity current venting while flooding.

We are grateful to the Research Council of Shahid Chamran University of Ahvaz for financial support (GN: SCU.WH98.31370). The authors are also grateful to the Center of Excellence for the Improvement and Maintenance of the Irrigation and Drainage Networks for providing part of the data.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.