Attempts to reduce cavitation phenomena at the Ciawi Dam inlet conduit using CFD Modeling

The conduit bottom outlet is prone to cavitation damage on its channel surface, and one preventive measure is the installation of air ventilation. This research aims to grasp flow characteristics comprehensively, predict cavitation likelihood at the intake, and assess the impact of a ventilation pipe’s installation. Utilizing Computational Fluid Dynamics (CFD), the study modeled flow dynamics, focusing on velocity and pressure variables. The Ciawi bottom outlet intake was meticulously modeled at a 1:1 scale under normal water level conditions, and model accuracy was validated through physical testing by BHGK. The research includes two series: Series 0 (intake without ventilation pipe) and Series 1 (intake with ventilation pipe), with gate opening heights varied at 25%, 50%, 75%, and 100% for both series. Simulation results show that in Series 0, potential cavitation damage is indicated at a 75% gate opening in sections 3 and 4 with a cavitation index of 0.11. Series 1, featuring ventilation pipe installation, demonstrated positive outcomes, eliminating potential cavitation damage in the inlet bottom outlet, with all cavitation indices surpassing 0.2. This study provides crucial insights for preventing cavitation damage in conduit bottom outlets through the strategic deployment of ventilation systems.


Introduction
The bottom outlet is one of the dam's outlet structures to release the flow from the reservoir into the original river.The location of the inlet structures is almost the same as the river bed elevation before impounding, as is this outlet.The bottom outlet on the dam is useful, among other things, to reservoir releases in emergencies that require repairs to the dam body or the dam foundation, for river channel replacement that flows the daily discharge as in Ciawi Dam and Sukamahi Dam, sediment flushing facilities such as the Sudirman Dam.Besides that, a function of the bottom outlet is to support the spillway to release excess flood discharge so that overtopping does not occur [1].
The structure of the bottom outlet can be a conduit or tunnel and usually utilizes a former diversion channel that is no longer used.During its lifetime, the bottom outlet structure may be damaged, such as surface concrete lining damage in the area after changing the geometry section or cracking problems in the inlet structure area.Very high-velocity flow will occur along the channel depending on the reservoir water depth and the cross-sectional area of the channel.In order to reduce costs, hydromechanical equipment for discharge regulation at outlets is usually made with economical dimensions.1311 (2024) 012010 IOP Publishing doi:10.1088/1755-1315/1311/1/012010 2 Cavitation appears when a local pressure drop is lower than the vapor pressure, which causes bubbles.When the bubbles carried by the flow burst, the high-pressure impact causes damage to the structure.From various studies that have been carried out, it is found that very high flow velocities will cause cavitation damage phenomenon, that is, structure surface damage [2], [3], [4].High pressure from cavitation repeatedly hitting the channel lining can create holes in the channel surface.In channels that use reinforced concrete material, damage will occur in the concrete cover, starting from small holes and eventually becoming large ones.When the structural reinforcement is exposed, and the bond between the reinforcement and concrete is reduced, the strength of the structure will decrease until the structure collapses.A pressure drop to a negative pressure can form downstream of the gate, which has no air vents, which will cause cavitation and vibration [5].
Prevention of bottom outlet damage is based on difficulty in the repair process and losses caused by the cessation of dam operations during the repair process.In the Ciawi Dam, a dry dam, the bottom outlet functions as a daily outflow channel that operates for its useful life.The risk that arises if the bottom outlet is closed is overtopping during floods.Prevention of cavitation damage has been widely researched, and it is explained that replacing channel structure materials with materials that have high tension, smoothing the surface of the channel, and providing ventilation or aeration to increase local pressure will reduce the possibility of cavitation events [3].In addition, cavitation can be avoided by providing aerators to increase pressure in areas that have cavitation potential [6].This research is to determine the effect of providing air ventilation downstream of the gate slot bottom outlet of Ciawi Dam to prevent cavitation damage.
The cavitation phenomenon is predicted using the cavitation index approach using water vapor pressure variables obtained from surface pressure measurements and flow modeling [7].Modeling using a physical model can provide information on velocity, discharge, and pressure.Additional costs for measuring instruments and model replacement will be required to obtain large amounts of data.Numerical CFD models can illustrate flow characteristics and variables velocity and surface pressure.Another advantage is that numerical modeling is cheaper and faster than physical models.In this research, the numerical model result is validated by the physical model result that the Center for Water Hydraulics and Geotechnics has carried out.Research has been conducted to reduce cavitation in the bottom outlet of Ciawi Dam.This study utilizes a numerical model employing ANSYS-FLUENT to ascertain the flow characteristics of the bottom outlet and evaluate the impact of installing an air ventilation pipe on the potential for cavitation.

Location
The research uses bottom outlet Ciawi Dam data, its geographic location at coordinates 6°39'24.22"S and 106°52'50.20"E. Ciawi Dam is on the Ciliwung River, which is included in the Ciliwung Watershed (Figure 1).
Ciawi Dam is a random fill type dry dam with a 55 m height and crest dam elevation of +551.0 m.Has two spillways, namely the main spillway (side channel spillway type) with a spillway crest at +546.75 m and the secondary spillway (conduit type) referred to as the bottom outlet with an initial elevation of channel base +504.2 m.The bottom outlet is designed to release a 50-year flood discharge of 253.25 m 3 /s, while the side spillway is designed to start operating at a 100-year flood discharge.

Data
The bottom outlet design carried out by Balai Besar Wilayah Sungai Ciliwung Cisadane (BBWS CC) uses a rectangular shape conduit with dimensions 4.2 m x 4.2 m and dimension 7.5 m x 4.2 m for the downstream part, its structure by reinforced concrete, along conduit lined by steel for the inside (Figure 2).The total length of this conduit is 398 m, which consists of an inlet, conduit type 1, conduit type 2, and stilling basin USBR type III for its energy dissipator.The longitudinal section of the conduit is illustrated in Figure 3. From this figure, the hydraulic profile of the channel can be divided into three, i.e., a) upstream channel with channel base slope 0.01 and length 326.16 m; b) downstream channel with channel base slope 0.2857 and length 42.5 m; and c) 42.5 long of energy dissipator with channel base elevation +485.62 m.At the inlet downstream of the gate, there is a 4.2 m long reducer (Figure 4) whose dimension is 4.2 m x 2.8 m, and then the channel dimensions return to 4.2 m x 4.2 m.This reducer for regulating 50-year flood discharge 253,25 m 3 /s.The air vent of intake used three pipes 30 cm in diameter installed downstream of the gate and 3 pipes more downstream of the reducer structure.Water level of reservoir following

Method
In this research, the following steps were carried out: a. Collecting data for modeling, i.e., bottom outlet drawing and design report from BBWS CC and physical modeling results that BHGK has carried out.b.Creating a bottom outlet 3D model according to data from BBWS CC for simulation.Numerical modeling will be carried out using ANSYS-FLUENT in general, and the modeling stages are as follows: 1) Pre-Processor: the process carried out is defining the model geometry for the computational domain, creating a grid, and defining fluid properties.2) Solver is a numerical computing process using one of the numerical methods, namely the approach of known variables to simpler functions, Discretization by approximate substitution into the equations governing the flow, and Solution of algebraic equations.3) Post-Processor: At this stage, the results of numerical computations are visualized and documented for analysis and other purposes.c.Flow simulation bottom outlet by varying the gate opening height (25%, 50%, 75%, and 100%); for 2 model series, intake without air vent and intake with air vent, measuring points for 4 locations around the intake as shown in

Flow Simulation
This research created a three-dimensional numerical model of Computational Fluid Dynamics (CFD) with ANSYS Fluent software to determine the gate's hydraulic characteristics and surface pressure downstream.Flow simulation will be carried out on one conduit hole for two series, namely intake without an air vent and intake with an air vent, with maximum conditions of reservoir water level (Flood Water Level).CFD is a numerical method for simulating flow using the Navier-Stokes and continuity equation, i.e., the continuity equation, momentum equation, and fluid energy equation [9], to obtain an overall flow profile (pressure, velocity, etc).

Cavitation Index
The cavitation index is a number used to predict the level of damage due to cavitation.The equation for the cavitation index [7].Falvey [10] predicts damage levels into three groups, as shown in Table 2.

Table 2. Cavitation damage level
Damage level Cavitation Index Damage potential 1 Revise the design

Boundary conditions and model meshing
The numerical modeling process begins with creating a 3D bottom outlet model from inlet to energy dissipator, as shown in Figure 6.The boundary conditions in this modeling are the inlet at location A and the outlet at locations B, C, and D. The maximum water level at intake upstream has an +550.25 msl elevation.Gravity is determined to be -9.81m/s 2 in the negative y-axis.The simulation method is transient with the multiphase model, namely water and air.The surface tension coefficient is constant at 0.072 N/m.Steady condition is achieved after inlet discharge is equal to outlet discharge.Furthermore, the simulation results presented will focus on the intake section.Based on Table 3, the CFD simulation results are close to the results of the physical model.

Result and discussion
Flow simulations have been carried out on two series bottom outlet models with the following results.
Observations were held on the flow profile, flow characteristics, and pressure distribution.

Series 0 simulation (Intake without air vent) a. Flow Profile
The flow profile investigation was carried out in the direction of the inlet bottom outlet longitudinal section.The experimental results are presented in Figure 10.The water profile downstream of the gate at gate opening 25% to 50% in the channel is approximately at the same level as the gate opening height, whereas after opening the gate 75% to 100%, the water thickness is as high as the channel.The water profile downstream of the reducer (section 3) shows a decrease in the water level for all gate openings due to the presence of a 1.4 m reducer.b.Flow characteristic At 25% to 100% gate opening, the flow downstream of the gate from section 1 to section 4 is supercritical.Table 4 shows that the flow velocity at the bottom channel increases after passing through the reducer, with the highest velocity being 16.9 m/s, namely at 100% gate opening.c.Pressure distribution Figure 12 shows the pressure distribution profile in the channel, and it can be seen that in the directing area, the pressure in the channel is positive, and after passing through the gate, there is a pressure decrease.Pressure measurements at the measuring point at the bottom channel (Table 4) provide values less than atmospheric pressure at the gate opening, ranging from 75% to 100%, with the lowest value being -25 KPa.

Series 1 simulation (Intake with air vent pipe)
The following simulation uses the series 1 model, namely intake added with air vent pipe 30 cm diameter of as many as six pieces in accordance with the site conditions.Series 1 simulation is carried out for 4 gate openings, the same as series 0, with variable inlet gate openings of 25%, 50%, 75%, and 100%.Figure 17 shows the flow profile at the inlet.At 25% to 75% gate opening, the water level downstream of the gate is the same level as the gate opening height, while the water level at 100% height opening reaches the roof.The decrease in water level occurs after passing through the reducer for all gate height openings, and the flow is free surface flow.The flow profile is shown in Figure 14.It can be seen that the flow velocity passing through the inlet from upstream to downstream has increased, especially after passing through the sluice gate and reducer.Table 5 shows that the highest velocity at the bottom channel of 32.8 m/s occurs when the height opening is 100%.After passing through the gate, the flow is supercritical, starting from 25% to 100% gate opening.c.Pressure distribution Figure 16 depicts the pressure contour at the inlet.Observably, the pressure in the approach channel is positive, decreasing post-passage through the gate and reducer.Table 5 displays the pressure at each measurement point.In the lower channel, all pressures are positive, while at the middle of the channel height, the minimum negative pressure registers at -13 KPa.The recorded discharges corresponding to each intake gate opening are as follows: 90 m 3 /s for a 25% gate opening, 170 m 3 /s for a 50% gate opening, 292.57m 3 /s for a 75% gate opening, and 302 m 3 /s for a 100% gate opening.These values represent the volumetric flow rates achieved at different levels of gate opening, providing a comprehensive overview of the system's hydraulic performance under varying conditions.The data obtained serves as a crucial foundation for understanding the hydraulic behavior of the inlet and is instrumental in assessing the system's effectiveness at different operational configurations.This information aids in optimizing flow control strategies and contributes to the overall enhancement of the dam's operational efficiency.

Discussion
The flow simulations conducted at the intake bottom outlet for Series 0 and Series 1 reveal noteworthy alterations in flow velocity and pressure variables.The analyses of these series provide detailed insights into the dynamic behavior of the flow, showcasing how the implementation of different configurations, namely the absence and presence of a ventilation pipe, influences the flow characteristics.The comparison between Series 0, representing the intake without a ventilation pipe, and Series 1, representing the intake with a ventilation pipe, highlights substantial flow velocity and pressure variations.This comprehensive examination allows for a more nuanced understanding of the hydraulic dynamics within the bottom outlet, emphasizing the significance of the ventilation system in influencing the flow parameters.These findings contribute to the broader understanding of the system's performance and serve as a foundation for informed decision-making regarding optimizing and enhancing the bottom outlet's operational efficiency.In sections 1 and 2, located upstream of the reducer, the application of the air vent pipe increased the pressure, although negative pressure still occurs.From Figure 18, it can be seen that a fairly high increase in pressure occurs when the gate is opened at 75% and 100%.After adding the air vent pipe, the lowest pressure is -4 kPa, which occurs on the surface of section 1. Table 4 and Table 5 showed that velocity at the bottom increased, and a fairly large increase occurred in section 1 for all gate openings, with average velocity from 11 m/s increase to 26 m/s.From Figure 19, it can be seen in sections 3 and 4 that the application of ventilation pipes increased the pressure at all measuring points.Negative pressure occurs on the surface but is not as big as before the pipe was given.The lowest pressure is -19.5 kPa on the surface of section 4 when the gate opening is 100%, where the initial pressure is -52 kPa.
We can compare changes in velocity from Table 3 and Table 4. Velocity at the bottom has increased.The highest velocity is 32.8 m/s from the previous 16.9 m/s.
The calculation of the cavitation index is shown in Table 6.In sections 3 and 4, the intake without an air vent pipe when the gate opened 75% is predicted structural damage will occur due to cavitation, which is discovered by a cavitation index value < 0.2, whereas in sections 1 and 2, there is no potential for cavitation damage.

Figure 4 .
The variables observed were flow velocity and surface pressure.d.CFD modeling results were validated by comparing the pressure variable of CFD simulation at 100% gate opening to the physical model carried out by BHGK to obtain a similar phenomenon.e. Analyze the simulation results by comparing the variable values in series 0 and series 1, and then make conclusions.

Figure 4 .
Figure 4. Intake profile points A -L are points of modeling measurement flow variables.

Figure 5 .
Figure 5. Pressure measurement locations of physical model carried out by BHGK Figure 4 shows the location of flow variable measurements carried out by CFD results, while Figure 5 shows the location of pressure measurements carried out by the physical model.Measurement point G in the CFD results has the same position as measurements in the physical model at point 12. So, these measurement points will be used to validate the CFD results.

Figure 6 . 7 . 6 Figure 8
Figure 6.3D model of bottom outlet Ciawi DamThe research will use two models: intake without an air vent and intake with an air vent, as shown in Figure7.

Figure 8 .Figure 9 .
Figure 8. Model meshing 2.7.Validation CFD Result The CFD results were validated with a physical model test carried out by BHGK.The variable used for comparison is static pressure at the same location between the simulation CFD and the physical model.A physical model was conducted for a 100% inlet gate opening without an air vent.Data obtained from physical model testing is shown in Figure 9.

( a )Figure 13 .
Figure 13.Water volume fraction profile in intake without air vent

Figure 14 .
Figure 14.Velocity contour profile in intake with air vent pipe a. Flow ProfileFigure17shows the flow profile at the inlet.At 25% to 75% gate opening, the water level downstream of the gate is the same level as the gate opening height, while the water level at 100% height opening reaches the roof.The decrease in water level occurs after passing through the reducer for all gate height openings, and the flow is free surface flow.

Table 3 .
CFD Result Validation at 100% Inlet Gate Opening

Table 5 .
Velocity and pressure in intake, series 1 (with air vent)