Numerical and experimental study of a very low head (VLH) turbine in an open channel at no-load conditions

Very low head (VLH) turbines are axial turbines with the capability of extracting power at a high efficiency of over 80% from very low heads (less than 4.5 m) and high flow rates of up to 30 m3/s per unit, with a significant reduction in construction costs, compared to conventional hydropower turbines. At no-load conditions, the turbine reaches its runaway speed, which is its theoretical maximum speed, causing damage to the equipment and the turbine structure, thereby imposing a considerable financial burden. As a result, understanding the performance of the turbine at the runaway speed is of utmost importance. The present paper centers its experimental and simulation efforts on predicting the runaway speed of the turbine rotor. An experimental study is conducted to determine the relationship between the runaway speed of the turbine and the flow rate, as well as compare its value with the rotational speed at the design point. Through the conducted experiments, it has been observed that at the mass flow rate corresponding to the design point, the runaway speed is 2.12 times higher than the rated value. Transient simulations using Computational Fluid Dynamics (CFD) are employed to investigate the performance under no-load conditions by estimating the generated torque as a key parameter. Simulations are carried out in an open channel configuration using the Ansys CFX software, and the results are compared to the experiments. In this study, a portion of the work focuses on simulating the free surface flow over the turbine, which allows for more realistic simulations of the turbine's operational behavior. The results of these simulations indicate that the homogeneous model is able to correctly predict free surface flow downstream and upstream. The numerical approach can also predict the runaway state of the VLH turbine with high accuracy, exceeding 95%; however, this variation is due to mechanical and hydraulic losses that have not been factored into the simulation. In this CFD modeling, setting up the boundary conditions accurately is vital, especially for the water levels at the inlet and outlet.