Wind Forecasting in Railway Engineering

Strong wind represents one of the most significant risks to railway safety. If winds can be forecast, early-warning can be given to running trains. Failure to forecast dangerous winds, on the other hand, can lead to sudden cross-wind incidents. Consequently, accurate wind forecasting is vital. However, wind signals are difficult to track with statistical or physical methods. With new hybrid intelligence systems, nonlinear wind signals can now be predicted, using intelligent models. Bringing together the latest developments in railway engineering, wind engineering, and wind forecasting theory and technique, is critically important to this aspect of safe railways.

Wind Forecasting in Railway Engineering presents core and leading-edge technologies in wind forecasting for railway engineering. The title is the first book to bring together wind speed forecasting and railway wind engineering, offering solutions from both fields. Key technologies are presented, and theories, modelling steps, and comparative analyses of forecasting technologies for railway wind engineering are given. Each chapter presents case studies and applications. The book consists in nine chapters, covering an introduction to typical applications and key issues; analysis of wind field characteristics; optimization methods for the placement of a wind anemometer; single-point time series along railways; deep learning algorithms on single-point wind forecasting; reinforcement learning algorithms; ensemble single-point wind forecasting methods; spatial wind; and data-driven spatial-temporal wind forecasting algorithms. This important book offers practical solutions for railway safety, by bringing together the latest technologies in wind speed forecasting and railway wind engineering into a single volume.