Supervisor: Sergio Campobasso, University of Glasgow
Sponsor Company: Gaia-Wind Ltd.
PhD Student: Marco Caboni
This student has graduated and the project is complete.
This project is developing a variable-fidelity system for the analysis and design of wind turbine rotors under uncertainty due to manufacturing and assembly errors, combining the low- and high-fidelity approaches. More precisely, a Blade Element Momentum (BEM) theory-like model coupled with a novel highly accurate and efficient 2D Navier-Stokes (NS) solver and incorporating statistic analysis tools is being developed. This system will provide a probabilistic prediction of the turbine power production. This stochastic aerodynamic performance tool is ideal for daily industrial use due to its very low execution times, and can also be used for robust design optimisation, namely to define a rotor design with minimal sensitivity to manufacturing and assembly errors.
The main benefit of low-fidelity methods is their high computational speed. Their drawback is the poor accuracy (and thus reliability) when dealing with complex aerodynamic phenomena (e.g. stall) and/or when analysing blade geometries differing from those on which the low-fidelity models were calibrated. An insufficiently accurate performance of BEM code predictions may also occur when the actual rotor geometry differs from the nominal design due to geometric errors. The system under development removes this modelling weakness by incorporating a novel efficient and accurate 2D NS solver and probability theory tools in a novel BEM-like code.