Cost-effective CFD simulation to predict the performance of a hydrokinetic turbine farm

After the performance assessment of a new hydrokinetic turbine prototype in the tailrace channel of a run-of-river power plant (Lavey, CH), the possibility to install a farm of such hydrokinetic turbines on this pilot site is investigated. In order to identify the optimal number and size of the machines as well as their location on this site, numerical simulations are carried out. Such a process requires to simulate a set of several free surface flow computations including different configurations of the farm in order to maximize the energy tapped by the machines. However, free surface flow computations are time and resources consuming. To allow such flow investigations, a different approach has been selected. In the present study, the turbines were replaced by a model mimicking the resistance effect of the machines. This model consists in a pressure drop applied at the turbine location to emulate the disturbance effect on the flow, especially the wake effect that will modify the available kinetic energy for the neighboring machines. Finally, it results that such a low CPU approach allows us to estimate the best location of the hydrokinetic turbines maximizing the energy harvested