Proposta de uma metodologia para avaliação de desempenho de fazendas de turbinas hidrocinéticas por métodos de ordem reduzida

Abstract

The growing demand for alternative renewable energy sources has increased interest in hydrokinetic turbines as a means of harnessing the high energy density available in riverine and marine currents. To improve energy extraction efficiency and power generation capacity, these devices are commonly deployed in arrays, also known as farms. In this context, the main objective of this work was to develop a methodology for evaluating the performance of hydrokinetic turbine arrays, with an emphasis on the application of low-order models. To this end, the Blade Element Momentum (BEM) method was assessed for the prediction of turbine power and thrust coefficients. This methodology requires lift and drag coefficient curves as functions of the angle of attack for the airfoil composing the turbine blades. Initially, the procedures used to obtain these aerodynamic coefficients were investigated, focusing on their predictive capability. The conventional BEM methodology was validated, and an improvement was proposed. The proposed enhancement consisted of employing the local Reynolds number of each blade section to calculate the lift and drag coefficients, rather than using a single representative Reynolds number evaluated at 70 % of the blade span for the entire blade. This modification was implemented in Fortran, validated, and registered as a new software named HydroBEM. Experimental validation of HydroBEM was performed using two different hydrokinetic turbines with experimental data available in the literature. The first one was a three-bladed turbine of 1.6 m diameter and blades based on a thickened NACA 63-415 airfoil, which was evaluated under two different turbulence intensity conditions. The second turbine had a diameter of 0.444 m and two blades employing the NACA 4415 airfoil. The study of turbine arrays requires accurate knowledge of wake velocity profiles, as these constitute the inflow conditions for downstream turbines. For this purpose, the conventional Jensen wake model and its cosine-modulated formulation were evaluated. Although widely adopted in the literature, these models present important limitations. To overcome these shortcomings, a new hybrid Jensen–Gaussian wake model was proposed. The model was experimentally validated and demonstrated superior performance compared with the standard Jensen formulation. Validation was conducted using literature data for a 0.7 m diameter turbine with three blades based on the NACA 63418 airfoil, as well as experimental data from the 0.444 m diameter turbine with two NACA 4415 blades previously employed in the HydroBEM validation. Finally, the latter turbine was also used to validate the coupled application of HydroBEM and the proposed wake model. The combined methodology provided satisfactory predictions of the performance of coaxial hydrokinetic turbines while maintaining a low computational cost.