Numerical Simulation of Long Tailrace Tunnels in Hydropower Pumped Storage Plants

This thesis aims to improve the current situation in the Duge pumped-storage hydropower plant in Norway. In the Duge power plant, which has an installed capacity of 200 MW, the rotor of one of the two Francis reversible pump-turbines (Unit 1) experiences excessive lifting with high loads. During high discharge, the resulting force acting on the turbine is directed upwards, threatening the turbine to lift and go out of its bearings. This problem only arose a few years ago when the turbines were refurbished in 2017 and since then, the maximal load has to be restricted to 80%. One of the challenges in Duge is its long tailrace tunnel. The resulting friction losses highly affect the pressure in the tunnel system. Even though Norconsult, a Norwegian consulting company, already investigated a mechanical engineering solution, this work aims to find a civil engineering solution for this problem. In order to achieve this, a 1D numerical model of the power plant is made to simulate its behavior. The numerical model is then used to test and propose different measures to improve the existing situation. The numerical tool used in this work is LVTrans, a freeware using LabVIEW, which allows to calculate hydraulic transients in pipe systems. The different steps of the work are carried out as follows: first, the model is calibrated and validated with field measurements; further, the worst-case scenario for the lifting problem is found from a dynamic and steady point of view. Finally, modifications of the power plant are suggested to solve the lifting problem. Three solutions for the improvement of the existing situation are selected: 1. Implementation of a downstream air cushion surge tank and an additional parallel tunnel in the tailrace. 2. Increase of the upper chamber volume in the main surge tank and an additional parallel tunnel in the tailrace. 3. Only an additional parallel tunnel in the tailrace. The first solution aims to avoid any problems with the worst load case and to be able to run the power plant at 200 MW, but is an expensive measure. The second solution is lighter from a technical and economic point of view and aims to avoid any problems in a normal operation case, with the maximal load also being increased to 200 MW. Finally, the third solution is the most economical and aims to improve the current situation by increasing the maximal load to 185 MW. Having in mind that Duge will receive new runners in the long term, it seems that the best option is the third solution, which is lighter on a technical and economic point of view.