Accuracy and consistency of CFD and engineering models for simulating vertical axis wind turbine loads. (1st September 2020)

• Record Type: Journal Article
• Authors: Bangga, Galih; Dessoky, Amgad; Wu, Zhenlong; Rogowski, Krzysztof; Hansen, Martin O.L.
• Is Part Of: Energy. Volume 206(2020)
• Publication Date: 2020-09-01
• Subjects: Aerodynamics -- CFD -- Energy -- Momentum -- VAWT; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2020.118087
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: The present work is intended to assess the ability of state-of-the-art approaches with various fidelity levels for accurate load predictions on vertical axis wind turbines (VAWT). The assessments are conducted by employing the Double-Multiple-Streamtube (DMS), Improved-DMS (IDMS), Unsteady Blade Element Momentum (UBEM), Vortex Model and fully resolved computational fluid dynamics (CFD) approaches. For the later case, three different codes are employed, namely FLOWer, TAU and Ansys Fluent. Three different turbines from low up to high rotor solidity (0.23, 0.53 and 1.325) are selected as the case studies. The prediction results are compared with experimental data at various operating ranges in terms of integral and azimuthal loads. The studies demonstrate that there is consistent agreement between engineering models at lightly loaded cases for the power curve prediction. The discrepancy at high tip speed ratio (λ) is caused by wake expansion, unsteady and decambering effects. In contrast, CFD predictions hardly show consistent power prediction but deliver accurate thrust values. Highlights: Three VAWTs at various solidity are simulated using state-of-the-art prediction approaches. The predictions make use of the DMS, IDMS, UBEM, Vortex model and CFD approaches. The consistency and accuracy of each model are presented and discussed. Low order models are consistent and accurate for lightly loaded cases. Wake expansion term is important for low order models at high TSR.