Fault Diagnosis Method for Photovoltaic-Storage Power Stations Based on Multimodal Data Fusion

With the rapid development of photovoltaic (PV) power generation, PV energy storage power stations are facing the challenge of frequent electrical faults. Traditional fault diagnosis methods suffer from slow response and insufficient accuracy. This paper proposes a fault diagnosis method for PV energy storage power stations based on multimodal data fusion. By deploying NFC sensors to collect real-time data such as current, voltage, and temperature, and combining it with the Transformer model for fault diagnosis and localization, this method addresses these challenges. Experimental results demonstrate that the system can effectively identify and locate faults such as inverter failures, grounding faults, cable breaks, and PV module faults. It significantly improves the speed and efficiency of fault response, providing reliable technical support for the intelligent operation and maintenance of PV energy storage power stations.