On the potential of the ideal diffuser augmented wind turbine: an investigation by means of a momentum theory approach and of a free-wake ring-vortex actuator disk model

Abstract

Diffuser-augmented wind-turbines are drawing increasing attention since they can beat the Betz-limit referred to the rotor-area. However, their diffusion is still prevented by some issues including: 1) the attainable power has not yet been shown to be larger than that of an open-turbine with the same frontal-area, 2) the classical analysis methods rely on the one-dimensional-flow and no-tip-gap assumptions whose impact has never been quantified. The paper addresses these two items investigating the potential of ideal diffuser-augmented wind-turbines using a newly-developed Axial-Momentum-Theory approach, and an extended version of a free-wake ring-vortex actuator-disk model. In comparison with similar methods, the novelty of the first approach is that it accounts for the two-dimensional effects and the tip-gap presence. Since this approach cannot evaluate the performance of a turbine for a given duct-geometry, a ring-vortex method is also developed. This is the first low-computational-cost method relying on the exact solution of the inviscid-flow through a uniformly-loaded ducted-turbine with a finite-size tip-gap. It strongly couples the flow induced by the duct and the wake which are modelled as the superposition of ring-vortices. The combined use of axial-momentum and ring-vortex methods leads to the following results. Firstly, it is clearly shown that an ideal diffuser-augmented turbine can extract more power than a Betz disk with the same frontal-area. To strengthen this statement, a new duct geometry with a remarkable value of the exit-area power-coefficient equal to 0.6098 is presented. This value is significantly higher than that of a base-line NACA5415 duct profile, i.e. 0.4800. Secondly, the impact of the one-dimensional-flow and no-tip-gap assumptions is evaluated. It is also shown that the tip-gap has negligible effects. Moreover, the one-dimensional-flow hypothesis has a low impact for high values of the rotor load, while the errors grow up decreasing the rotor thrust.