Hybrid simulation and experimental framework for real-time fault detection in PV boost converters using fuzzy logic and LoRa connectivity

The reliable operation of photovoltaic (PV) energy systems depends heavily on the performance of DC–DC boost converters, which are susceptible to faults in their semiconductor switches and passive components. This paper introduces a hybrid simulation and experimental framework for real-time fault diagnosis in photovoltaic boost converters, integrating fuzzy logic-based diagnostic intelligence with LoRa-enabled remote monitoring. A Python-based time-domain simulation was developed to model converter dynamics under normal and faulty conditions, encompassing scenarios of diode, inductor, and MOSFET degradation. Unlike existing approaches that rely mainly on simulation or limited laboratory validation, the proposed framework combines real-time hardware implementation with IoT-based monitoring, enabling both accurate fault classification and scalable remote supervision. Simulated voltage–current signatures were validated through a hardware prototype based on an Arduino Mega controller, voltage and current sensors, and a LoRa32 module for long-range wireless transmission to a Firebase cloud platform. The fuzzy inference system, implemented in both simulation and hardware environments, successfully classified component-level faults with sub-second response time. The results indicate reliable detection performance, reaching fault diagnosis within 1 s and cloud-based reporting in approximately 1 s, while exhibiting robustness under diverse irradiance and load conditions. The proposed hybrid framework improves diagnostic reliability, facilitates scalable IoT-based monitoring, and creates a replicable basis for the intelligent monitoring of renewable energy converters.