Proposta de uma metodologia para o projeto aerodinâmico de turbinas eólicas e hidrocinéticas

Abstract

The main objective of the present study was the development of a pragmatic methodology that allows to evaluate, with some precision, the performance of wind and hydrokinetic turbines. For this evaluation, the QBlade software was applied, which is based on the Blade Element Momentum (BEM) methodology and requires the lift and drag coefficient curves as a function of the angle of attack for the turbine blade aerodynamic profile. Such curves can be obtained in the literature or by computational simulation. In both cases, the Reynolds number considered is the one that occurs at a position equal to 70% of the blade span. For the computational simulation, two-dimensional meshes were generated, whose dimensions are based on the profile chord length. The calculation of such coefficients is performed only for an angle of attack range, and for the others an extrapolation is made within the QBlade software applying the Viterna method. The solution of the Navier Stokes equations was obtained by applying the Reynolds Averaged Navier-Stokes methodology. Before being inserted into the QBlade software, the coefficient curves must be treated by one of the two models: Aerodas or Stall Delay, so that behaviors not captured in the two-dimensional simulation are considered. For sections with a circular profile, the drag coefficient is graphically obtained. All considerations presented were applied in the study of two turbines that have experimental results available in the literature. The hydrokinetic turbine has two blades composed of the aerodynamic profile NACA 63-618, rotor radius equal to 0.4 m and Reynolds number equal to 4×10^5. Similarly, the wind turbine has two blades with S809 airfoil, rotor radius equal to 5.029 m and Reynolds number equal to 1×10^6. The Langtry-Menter Shear Stress Transport transitional turbulence model presented the best results for the turbines’ aerodynamic profiles. The power curves obtained with the coefficients treated with the Aerodas model, for both turbines, showed a good approximation to the experimental values obtained in the literature. In addition, the results did not show significant differences when applying the curves of lift and drag coefficients complete or limited to stall angle. On the other hand, the estimated power curves with the coefficients treated with the Stall Delay were even more accurate than those obtained with the coefficients treated with the Aerodas. However, for the wind turbine it was necessary to estimate lift and drag coefficients for angles of attack greater than the stall angle. Finally, a flowchart for the energy evaluation of wind and hydrokinetic turbines was proposed.