Analyzing and forecasting lightning flashes and the related wind gusts at a wind energy power plant in a hilly region of western Greece

Wind power plants are vulnerable to abrupt weather changes caused by thunderstorms associated with light-ning activity and accompanying severe wind gusts and rapid wind direction changes. Due to the damages that such phenomena may cause, the knowledge of the relationship between storm systems and the produced wind field is essential during the construction and operation phase of a plant. In the first part of this study, the relationship between severe wind gusts and lightning activity in a power plant in Greece is investigated. Wind data are measured at the wind turbines for a 3-year period (2012-2014); the corresponding lightning data come from the ZEUS lighting detection network. Wind gusts are well correlated to lightning strikes. This correlation is maximized during winter when well organized weather systems affect the area and minimized in summer as a result of local storms due to thermal instability. The second part of the study focuses on the development of an artificial neural network (ANN) model in order to forecast these two parameters in a 1-h ahead horizon based on wind speed, wind direction, and maximum observed wind gust measured at the nacelle of a wind turbine and four other variables, namely CAPE, TTI, wind speed at the 500 hPa isobaric level, and the 0-6 km vertical wind shear. The proposed model could be considered as a promising tool in simulating the occurrence both of wind gusts and lightning flashes, providing a relatively good evidence of the possibility of occurrence of such events.