Mining frequent spatio-temporal patterns in wind speed and direction

Abstract

Wind is a dynamic geographic phenomenon that is often characterized by its speed and by the direction from which it blows. The cycle’s effect of heating and cooling on the Earth’s surface causes the wind speed and direction to change throughout the day. Understanding the changeability of wind speed and direction simultaneously in long term time series of wind measurements is a challenging task. Discovering such pattern highlights the recurring of speed together with direction that can be extracted in specific chronological order of time. In this chapter, we present a novel way to explore wind speed and direction simultaneously using sequential pattern mining approach for detecting frequent patterns in spatio-temporal wind datasets. The Linear time Closed pattern Miner sequence (LCMseq) algorithm is constructed to search for significant sequential patterns of wind speed and direction simultaneously. Then, the extracted patterns were explored using visual representation called TileVis and 3D wind rose in order to reveal any valuable trends in the occurrences patterns. The applied methods demonstrated an improvement way of understanding of temporal characteristics of wind resources.