Enhanced structural health monitoring for large-scale offshore photovoltaic systems based on response pattern recognition

Offshore floating photovoltaic systems are becoming increasingly prominent due to their energy efficiency and operational reliability. Ensuring their structural integrity, particularly the connections between floating modules, is essential to prevent instability and failures. Traditional inspection methods are impractical for large-scale installations with multiple connections and numerous connecting structures. To address this challenge, we propose a data-driven structural health monitoring approach using artificial neural networks (ANN). In this paper, we introduce an anomaly identification technique employing ANN trained on datasets comprising the motion of floating bodies. Hydrodynamics-based simulations validate the effectiveness of this method, demonstrating its potential to enhance structural health monitoring in offshore photovoltaic installations.