REAL-TIME FAULT DETECTION AND DIAGNOSIS OF SOLAR PHOTOVOLTAIC SYSTEMS FOR RAILWAY MICROGRID USING IOT AND MACHINE LEARNING

Abstract

 The rapid adoption of solar photovoltaic (PV) systems as a sustainable energy source has made effective monitoring and maintenance crucial for ensuring reliability and performance. Traditional PV fault detection methods often rely on manual inspection or offline analysis, leading to delayed diagnosis, energy losses, and reduced system lifespan. This research presents an Internet of Things (IoT)-based framework for real-time fault detection and diagnosis (FDD) in solar PV systems. The proposed system continuously monitors key parameters such as voltage, current, temperature, and irradiance using low-cost sensors, microcontrollers, and cloud-based platforms. By applying advanced machine learning algorithms and anomaly detection techniques, it effectively identifies faults such as partial shading, open circuits, line-to-line short circuits, and PV module degradation. Furthermore, the IoT-enabled design supports remote accessibility, data visualization, and predictive analytics for proactive maintenance. The proposed FDD framework is also extended to railway microgrid applications, where integrated PV and energy storage systems power both traction and auxiliary loads. This integration enhances energy efficiency, system reliability, and fault resilience, contributing to the development of smart and sustainable railway infrastructure.