RANS simulations of neutral atmospheric boundary layer flow over complex terrain with comparisons to field measurements

Abstract

Micro-scale Reynolds-averaged Navier-Stokes (RANS) simulations of the neutral atmospheric boundary layer (ABL) over complex terrain and a comparison of the results with conditionally averaged met-tower data are presented. A robust conditional sampling procedure for the meteorological tower (met-tower) data to identify near-neutral conditions based on a criterion for the turbulence intensity is developed.The conditionally averaged wind data on 14 met-towers are used for the model validation. The ABL flow simulations are conducted over complex terrain which includes a prominent hill using the OpenFOAM-based simulator for on/offshore wind farm applications (SOWFA) with the k−ϵ and the SST k−ω turbulence models. The discretization of the production term in the transport equation for the turbulent kinetic energy (TKE) is modified to greatly reduce the commonly observed nonphysical near surface TKE peak. The driving inflow is generated through an iterative approach using a precursor method to reproduce the measured wind statistics at the reference tower. Both of the RANS models are able to capture the flow behavior windward of the hill. The SST k−ω model predicts more intense flow separation than the k−ϵ model downstream of the steepest sections of the hill. The wind statistics predicted at the location of the met-towers by both of the RANS models are fairly consistent. Overall, the comparisons of the direction, mean, and standard deviation of the wind between the simulations and the tower data show reasonable agreement except for the differences of the mean wind speeds at four met-towers located closer to the main ridge of the hill in a region of strong terrain variations. KEYWORDSatmospheric boundary layer, complex terrain, field measurements, k−ϵ model, neutral stability, Reynolds-averaged Navier-Stokes (RANS) simulations, SST k−ω model terrain with hills, ridges, and even mountain slopes. The interaction of these complicated landforms with the atmospheric boundary layer (ABL) turbulent flow may greatly affect both the character of the large scale flow over wind farms and the local flow features such as flow acceleration, separation, and recirculation. Therefore, a detailed wind resource analysis for a prospective wind site is important for the wind engineers to accurately predict the power output from the entire wind plants.The most direct way to obtain wind resource information for a prospective wind farm is to conduct field measurements with meteorological towers (met-towers). The monitoring duration for the met-tower observation typically lasts for at least 1 year to obtain a sufficient number of samples so that reliable statistical analysis of diurnal and seasonal variability of the wind can be determined. 2 Due to cost limitations, the mettowers are usually sparsely distributed across the wind farm site, and thus, they only provide a highly localized assessment of the wind conditions. Extrapolation of the wind conditions from the tower locations to other regions of the wind farm is difficult in complex terrain...

Analytical three‐dimensional wind flow model for real‐time wind farm simulation

“…Wind flow modelling and energy estimates : To analyse and evaluate potential wind sites, LES‐based assessments enable a thorough analysis of wind flow even in complex terrain with exceptionally high detail and have been compared against meteorological data as presented in [30]. Thus, numerical simulation techniques are applicable for comprehensive flow modelling if computation time is not an issue.…”

“…The National Renewable Energy Laboratory's (NREL) high fidelity Simulator fOr WF Applications (SOWFA) [22], which is based on the OpenFOAM toolbox [23], is widely used to analyse aerodynamic interrelationships under different conditions or as comparison model reference [24][25][26][27][28][29][30][31][32][33][34]. In addition, the Fatigue, Aerodynamics, Structures, and Turbulence (FAST) tool for non-linear aero-hydro-servo-elastic simulation of WT rotors, structural loads and drive train components can be coupled to SOWFA to cover all relevant processes and energy conversions with high detail [35].…”