Validation of Wind Power Plant Modeling Approaches in Complex Terrain

The effect of terrain on wind-power-plant aerodynamics is often neglected due to the complexity of terrain-atmosphere-turbine interactions and the computational effort required to adequately resolve the relevant physics. In this work, the effects of topography on plant performance are investigated using a comprehensive set of measurements from the Wind Forecast Improvement Project 2 and three modeling approaches operating at different levels of fidelity. A high-fidelity approach consists of large-eddy simulations on a terrain-resolving mesh to produce unsteady, heterogeneous flow that is sensitive to the local topography, with turbines modeled as actuator lines. In comparison, an engineering approach makes use of a steady-state wake model that has been modified to account for terrain effects. An intermediate level of fidelity is achieved through the use of a multiphysics model that couples wake dynamics to the aeroelastic response of each wind turbine and uses inflow from a large-eddy simulation. This paper details a case study of nine turbines in a near-neutral atmospheric boundary layer after a wind-speed ramp. In situ and remote-sensing measurements upstream of the wind plant are used to prescribe initial and boundary conditions to the models. Wind turbine power measurements are used to assess the ability of the different models to capture the effects of terrain on turbine wake dynamics and therefore power production in an array. The study demonstrates that accounting for unsteady turbulent inflow and averaging over time periods longer than 10 minutes are necessary for realistic predictions of mean power output, due to unsteady, local variations in wind speed and direction. Engineering models that account for local uphill terrain steepness may not be sufficient in all cases as favorable pressure-gradient effects may be overestimated and downwind terrain-induced vertical wake displacement is neglected.