High-fidelity CFD Simulations of Two Wind Turbines in Arrays using Nonlinear Frequency Domain Solution Method

Shine Win Naung, Mahdi Erfanian Nakhchi Toosi, Mohammad Rahmati*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Aerodynamics of a wind turbine within windfarms is strongly influenced by the wake of neighbouring turbines. In particular, the performance of a wind turbine can be dramatically reduced depending on its location in the wake region of an upstream turbine. A detailed investigation of the effect of the upstream turbine on the downstream turbine with respect to their distances is essential for the design and optimisation of wind farm layouts. Conventional time domain solution methods, such as Unsteady Reynolds Averaged Navier Stokes (URANS) based Computational Fluid Dynamics (CFD) model of wind turbines in arrays, can provide a detailed analysis of this interaction effect. These methods are, however, impractical due to the high computational cost required for modelling turbines in array configurations. In this paper, a novel modelling and computational method is proposed to simulate two wind turbines in arrays by considering them as a multi-stage turbine. A nonlinear frequency domain solution method is then employed to model flow nonlinearities due to their interactions. The distances between the turbines are varied, and the effects of the upstream wind turbine on the downstream one are thoroughly investigated. Extensive validations of the nonlinear frequency domain solution method against the conventional time domain solution method reveal that the proposed frequency domain solution method provides accurate results while reducing the computational cost by one to two orders of magnitude.
Original languageEnglish
JournalRenewable Energy
Early online date22 Apr 2021
DOIs
Publication statusE-pub ahead of print - 22 Apr 2021

Fingerprint Dive into the research topics of 'High-fidelity CFD Simulations of Two Wind Turbines in Arrays using Nonlinear Frequency Domain Solution Method'. Together they form a unique fingerprint.

Cite this