Intelligent Lightning Hazard Warning System for a Wind Farm

Abstract

Wind energy is one of the most important forms of renewable energy, and with the progress in this field, as the production capacity of wind turbines has increased, their height has also increased significantly. The height of wind turbines, number of them in a wind farm, and their specific location have increased the probability of lightning strikes and made them one of the most important hazards for wind turbines. Given the importance of maintenance and inspection for wind farms, creating a system for detecting safe time for these operations with low lightning probability is very useful. In this regard, by analyzing local meteorological data (pressure, temperature, wind speed, wind direction and humidity) and the lightning location system data an intelligent warning system for lightning hazard in a wind farm is developed based on machine learning methods. It is applied and tested on a case study of a wind farm in Croatia. The results show the success of this lightning hazard warning system in predicting the safe times with low lightning probability for the wind farm.