Chaired by: Joshua A. Paquette | SANDIA
Topic: DSE. Design, System Engineering and New Concepts
Form of presentation: Oral
Duration: 90 minutes
An airfoil database was developed in the present work. The database includes a multivariate, unstructured grid, Radial Basis Function (RBF), optimization-compatible, interpolator for airfoil aerodynamic coefficients. Its use in a simplified blade optimization is demonstrated in the present abstract. The database was designed to aid complex blade design optimizations carried out in a novel optimization framework, HAWT Opt 2, being currently developed at DTU Wind Energy. In the database, the lift, drag and moment coefficients are stored as functions of airfoil family, thickness, Reynolds number and angle of attack. Aerodynamic add-ons and airfoil modifications are included in the database as separate airfoil families. Additionally, the database includes two different 3D correction methods and two different 360 degree extrapolation methods.
Marco Caboni, Edmondo Minisci, Annalisa Riccardi
This paper presents different approaches to design wind turbine airfoils accounting for the uncertainty affecting the input parameters of the numerical models used to assess the airfoil performance. The approaches are specifically applied to the aerodynamic design optimization of a wind turbine airfoil considering the uncertainty associated to the XFOIL’s parameter NCRIT. Subject to a set of aerodynamic and structural constraints, the uncertain response of the airfoil is optimised by means of both probabilistic and imprecise probability based approaches. The aim of the work is to show how different techniques to treat uncertainties can be used for the design of airfoils, depending on the available information, and the associated computational costs. The ultimate goal is to demonstrate that the airfoils designed by uncertainty based approaches are actually more robust and reliable than the ones designed deterministically.
Jared J Thomas, Andrew Ning
This paper presents a method for reducing multi-modality in the wind farm layout optimization problem. The reduction was achieved by increasing and then slowly decreasing the wake spreading angle while maintaining the wake velocity decay rate. The results distribution using the new method had a mean annual energy production (AEP) improvement of 11%. The typical multi-start method had a mean AEP improvement of only 7%.
Frederik Zahle, Niels N Sørensen, Michael K McWilliam, Athanasios Barlas
This article presents a design study into the redesign of a wind turbine blade tip seeking to increase the energy production subject to the loads constraints of the existing blade. The blade shape is parameterized to allow for planform changes with respect to chord and twist, as well as blade length extension, and additionally three parameters that allow to explore winglet-like shapes. The design strategy uses 3D CFD computations to create a surrogate model, after which numerical optimization is used to optimize the tip shape subject to a number of geometric and loads-based constraints.