I finally convinced myself that part of my recent works, ideas and experience with wind power forecasting should be compiled into a book. The writing of this book is still in progress – it will come along with datasets and R functions, for educational purposes.
Waiting for the book to be finalized and published, here is its synopsis:
Wind energy is increasingly recognized as a necessary feature of modern and future energy portfolios. While this form of renewable energy already has a significant share in the electricity production of countries like Denmark, Spain or Ireland, it is also expected to contribute to ambitious objectives of CO2 emissions reductions for a large number of other countries worldwide. But even though wind energy may be seen as an ideal candidate in the search for sustainable and clean energies, its actual integration in the electricity networks and markets comes as a complex challenge. This is the case for nearly all actors of the energy sector such as power producers, utilities, network operators, energy traders, or even policy-makers. A main reason for that is the completely different nature of wind energy compared with more conventional types of energies like gas, coal or hydropower. Wind power generation is a direct function of the meteorological conditions, which we humans have no control of, and is hence highly fluctuating. An essential contribution to an optimal integration of this renewable energy into power systems and markets is then the forecasting of its power output at lead times ranging from a few minutes up to 10 days ahead. The book introduces the general mathematical and statistical framework for the problem of wind power forecasting. It also considers a range of practical aspects ranging from the evaluation of forecasts to the communication of forecast uncertainty and it use in decision-making. The authors start by covering the essential aspects of the data used as input (of both electrical and meteorological nature), then laying down the general models and methods of wind power forecasting with emphasis on statistical methods. The complementary approaches of point forecasting, probabilistic forecasting and forecast scenarios are dealt with, while insisting on the importance of forecast uncertainty, its communication, as well as its use and value in decision-making. This book addresses scientists and engineers working in wind energy related R&D and industry, as well as graduate students and non-specialist researchers with interest in the fields of forecasting, statistical science, meteorology and renewable energy.