Forecasting wind power and turbine energy efficiency
Loading...
Date
2024-10-10
Authors
Μίχος, Δημήτριος
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This dissertation presents a comprehensive methodology for modeling, simulating and forecasting the behaviour of turbulent airflow over complex terrain in the vicinity of wind farms, with a particular focus on addressing challenges associated with limited wind data availability. The goal of this approach is to enable accurate short-term energy production forecasting for wind turbines operating in such challenging environments. A case study of the Lavrio wind farm in Greece is utilized to illustrate and validate the methodology of this PhD research. The approach leverages vertical wind LIDAR measurements, coupled with a novel spatial extrapolation technique based on a CFD model (Wi.Sp.Ex), to reconstruct the airflow above the complex terrain of the wind farm. The k-ε turbulence model is employed to efficiently capture the near-wall flow behavior. Through a series of steady-state simulations with varying inlet conditions and initial values, the temporal evolution of the turbulent flow field is investigated, providing valuable insights into the complex interactions between wind and terrain. The Wi.Sp.Ex. model serves as the core physics-based component of
the Wind Energy Extraction Latency (W.E.E.L.) model, which forecasts the 15-minute power production of a wind turbine by identifying its wind energy extraction latency and utilizing the Wi.Sp.Ex simulation results as wind speed forecasts. This integrated approach enables a more accurate and reliable prediction of short-term energy production, facilitating improved wind farm management and grid integration strategies. This methodology offers a promising solution for enhancing the accuracy and efficiency of wind resource assessment and wind farm design, particularly in complex terrain settings where traditional measurement and modeling approaches might fall short. By addressing the challenges associated with limited wind data availability and incorporating the temporal dynamics of the wind field, this approach contributes to the advancement of wind energy technology and its integration into the broader energy landscape.
Description
Keywords
Wind energy yield forecasting, Short term forecasting, Computational fluid dynamics, Turbulent flow, Wind energy, Wind turbines, Wind parks