Kinetic turbines harnessing tidal and ocean currents make use, in some designs, of nozzles and/or diffusers. Nozzles come at a cost, but they can help from the structural, hydrodynamic or positioning points of view. In those cases, they might make sense as long as they drive the LCoE (Levelized Cost of Energy) down, which is the ultimate objective of energy-harnessing devices. The design must then optimize the combined performance of both blades and nozzle. However, the interaction between turbine blades and nozzle is not always fully clear, and even less its optimization. A relevant amount of efficiency can be lost if the design spiral is not appropriate.
The authors have suggested in  an approach for the optimization of turbines within nozzles. This approach was followed in  and validated with model tests. In the approach, the turbine is initially substituted by an actuator disc that applies a radially constant pressure drop. But in these references, the optimum pressure drop in the actuator disc was the same as if there was no nozzle at all, i.e., 4/9ρv2. This is equivalent to considering the nozzle coefficient does not depend on the pressure drop, and thus, on the induced velocity field. Hence it is a somewhat arbitrary assumption.
This paper describes, using actuator disc theory, how nozzles affect the disc optimum pressure drop in uniform flow conditions. The effect of a hub is also analyzed. Then, using a viscous FVM CFD code, the variation of the pressure drop is quantified for two different acceleration nozzles, one suffering flow separation and the other one not. As the pressure drop increases, so does the flow expansion downstream. This rises the average radial component of velocity at the nozzle, increasing the thrust and nozzle coefficient. Therefore the optimum pressure drop goes up compared to that without nozzle. The increment in efficiency that can be obtained with this approach is quantified for the studied nozzles. Finally, the integration of this effect into the blade design is discussed.