Multi-Source Meteorological Data Integration for Lightning Risk Assessment in Wind Power Systems: A Systematic Review

Abstract

The increasing deployment of wind power plants has heightened their exposure to lightning-related hazards, which pose significant risks to turbine integrity, operational continuity, and economic performance. Traditional lightning protection systems are often limited by reactive mechanisms and reliance on single-source data, resulting in insufficient detection of complex lightning phenomena such as upward and intra-cloud discharges. This study aims to develop and evaluate a real-time multi-source meteorological data integration framework for advanced lightning risk detection and protection in wind power systems. The proposed approach integrates satellite observations, ground-based Internet of Things sensors, radar and light detection and ranging technologies, lightning detection systems, and historical datasets within a unified architecture. Machine learning models, including logistic regression and Extreme Gradient Boosting (XGBoost), are applied to classify lightning types and compute a composite risk index for decision-making. The results demonstrate that the integrated framework significantly improves prediction accuracy and enables timely identification of lightning precursors, achieving a response latency of less than three seconds. The system supports automated protective actions, such as turbine shutdown and repositioning, thereby reducing potential damage and downtime. Furthermore, the modular and cost-effective design allows scalable deployment across different operational environments. In conclusion, the integration of multi-source meteorological data with artificial intelligence and real-time monitoring technologies provides a robust and adaptive solution for enhancing lightning protection and resilience in wind energy infrastructure under increasingly variable climatic conditions.