Power Prediction of a Fermeuse Newfoundland Wind Farm

Abstract

Wind power forecasting within time intervals ranging from few minutes to hours is essential for utilities. Such information is required for planning operation and shut down of thermal units in the system. This research involves finding a simulation model and developing software that could be used for power prediction of Fermeuse wind farm. The model includes a simple individual wind turbine dynamic model, variation in wind speed with height, variation in wind speed over the site area and the wake interaction between wind turbines. Detailed Computational Fluid Dynamics (CFD) based flow models of a wind farm including dynamic model of a wind turbine are too complex and are not suitable for real time power prediction and need a supercomputer for calculation. Wind power physical model takes into account topographical map, a wind farm layout and the long term site wind and atmospheric data and uses this information to calculate wind speed at all wind turbines. Using the wind turbine physical model the output power of each individual turbine in a wind farm is established and the wake effects are calculated. The developed power model is applied to Fermeuse, Newfoundland 27MW wind farm and the results are presented in the paper.