Predictive engineering in wind energy: A data-mining approach

The large-scale wind energy industry is relatively new and is rapidly expanding. The ability of a wind turbine to extract power from the wind is expressed with the power curve. The key parameter determining wind turbine performance is wind speed and it is normally measured by an anemometer placed at the nacelle of a turbine.;The dynamic nature of wind is a barrier that calls for applying predictive engineering. Traditional approaches based on physics and mathematical modeling are not fully handle the variable nature of the wind.;Data mining is a promising approach for modeling in wind energy, including power prediction and optimization, wind speed forecasting, power curve monitoring, and fault diagnosis. It involves a number of steps including data pre-processing, data sampling, feature selection, and dimensionality reduction. This Thesis focuses on applying data-mining to predictive engineering in wind industry. Models for prediction of wind speed and wind farm power, turbine, and fault diagnosis are built. However, the approach and methods discussed in this research are also applicable to other industrial processes.;Chapter 2 introduces a methodology for short-term wind speed prediction based on wind farm data. Chapter 3 and Chapter 4 present prediction models for wind turbine parameters. Chapter 5 proposes strategies for dynamic control of wind turbines. Chapter 6 explores the fault diagnosis and prediction using SCADA data.