Summary

Funagira Dam has nine gates. The closing of the gates during the falling limb of the flood hydrograph has two objectives: first to maintain the flushing level and second to raise the reservoir water level up to the desired level. This translates to closure in two stages, each stage following a different closure approach to meet the two objectives. In this assignment, both closing approaches are investigated. The model shows that the first stage of the current practice of gate closing, which has been designed to meet the flushing level objective, is fulfilling its objective best. Nevertheless, the second stage of the current practice of gate closing has some room for improvement to reduce the formation of eddies at the downstream side. A new gate closing approach for the second stage is proposed and compared with the current 2^nd gate closing manner.

The stilling basin has recently been protected to reduce the risk of bed erosion. The new bed elevation of the stilling basin slightly increases the velocity magnitude within the first 200 m downstream of the dam. However, the model results show that the protection does not have much impact on the water level downstream the dam. The model results show that lowering of the bed in front of gate 1 is effective. However, further investigation would be needed to ensure that for the whole stilling basin.

Furthermore, a new 2D hydrodynamic model has been built for Funagira Reservoir using the new Delft3D flexible-mesh suite. The model has been validated against the Delft3D4 software and the data. The new model will be upgraded and used for further studies of the dam and its reservoir. The flexible-mesh software has many new functions and added value to improve the modelling of Funagira Dam and its reservoir.

   Figure a Funagira Dam

The objectives

 The aim of the project is to investigate the gate closure procedures and the recently implemented riverbed protection at the downstream of the dam. In addition to that, a new flexible mesh model will be built for Funagira Reservoir.

 During the project progress meeting (dated February 2019), J-Power requested additional analysis to provide insights about using free flow operation during the flood season for discharge higher than 2,000 m³/s. This topic will be briefly addressed in this study in preparation for further investigations in the future.

This project objectives may be listed as the following components:

1. Investigate the closing of gates during the falling limb of the flood peaks of various discharges
2. Investigate the free flow operation during the flood season.
3. Investigate the flow behaviour after the new protection of the stilling basin. 
4. Investigate the energy dissipation of the stilling basin in front of gate one.

Findings

Part of the research project findigns are:

To improve the operation of Funagira Reservoir during flood peaks, new gates operation is introduced and tested in a model. Moreover, the new bed protection is evaluated in order to ensure proper functioning of the stilling basin. In this chapter, the findings are summarized and recommendations concerning reservoir operation and the efficiency of bed protection are provided. 

This assignment has been divided into five components; Based on the analysis the following might be concluded: 

1. Component A: Operation of gates during peak flows.
   For that, the closing of gates has been divided into two phases: 

• • • Closing the gates to maintain the desired flushing water level during the flood season. The model shows that the current approach is good (giving the current constraints), and no further optimisation is needed.
    • Closing the gates to fill the reservoir to level 54.8 masl have some room of improvement. The model shows that the closing of Gate 4,5 and 6 as last gates creates a jet towards the left bank (see the upper plot of Figure ‎3.13). This might be contributing to the erosion due to the movement of soil in between the tetra-blocks used to protect the left bank. Accordingly, we proposed a new operation protocol that leads to closing gates 7, 8 and 9 as the last
    • The new approach (see Figure ‎3.6) seems to be useful to mitigate the erosion of the left bank. However, before giving a firm recommendation to change the operation protocol, we would like to verify the findings of the model based on dedicated field measurements; including a test case for proposed protocol.

 2. Component B: Operation during flood season (free flow stage)

Based on the model results we recommend starting the free flow stage once the discharge is > 2,000 m³/s. This is expected to be more effective in term of sediment flushing and operating efforts. Nevertheless, decision and examination must be made for selecting the flushing water level if the discharge is less than 2,000 m³/s. The two levels proposed are 50.6 masl and 47 masl. This issue needs to be investigated further. The results show that the new operation rule during flood will: 

• • Lower the water level in the reservoir by around 4 m compared to the current operation when the discharge is 2,000 m³/s. This 4 m declines when discharge increases. When the discharge reaches 8,000 m³/s the old and new operation rules have similar water levels.
  • Increase the velocity magnitude within the reservoir. Within the first 200 m upstream the dam, the velocity is significantly increased. The velocity magnitude is not changing much as the discharge increases from 2,000 m³/s to 8,000 m³/s.
  • Not have much effect on the water level downstream of the dam. Its effect on the velocity downstream is minor. 

3. Component C: new bed protection for the stilling basin

Three simulations have been conducted to investigate the effect of the new bed protection. Based on the topography of November 2017, the bed level is raised. This does not have much impact on the downstream water level; but it may increase the velocity magnitude. The increase in velocity ranges from 25% to 10% as the discharge increase from 1,800 m³/s to 11,200 m³/s. The model results also show that the impact is mainly within the first 200 m downstream of the dam. The reflection of this increase on the velocity after the first 200 m is minor. Furthermore, it seems that the increase of the bed roughness does not have much impact on the hydrodynamic condition downstream the dam.

 4. Component D: Energy dissipation downstream of gate 1

The bed level downstream of gate 1 is relatively high. This leads to a shallow water depth there. Consequently, this has an influence on the efficiency of the downstream area to dissipate the energy. Therefore, it might be beneficial to lower the bed downstream of gates 1 and then protect it. This may also have an effect on damping the eddies moving towards the left bank. The model results indicate that deepening downstream gate 1 reduces the velocities locally. Still, it is required to study the effects in detail to ensure that the lowering is effective within the whole stilling basin.