Enhancing fault detection in bifacial photovoltaic systems: a two-stage CNN-RF approach with I-V curve analysis

Rising global energy demand, driven by population growth and industrialization, threatens environmental sustainability, necessitating renewable solutions to combat climate change and resource depletion. Bifacial photovoltaic (PV) modules, capturing sunlight on both sides, yield 20–30% higher energy in high-albedo environments than monofacial modules but suffer 10–40% efficiency losses from faults like shading, dust, aging, degradation, and cracks, with dual-sided designs complicating detection. This study pioneers a multi-fault classification framework for bifacial PV systems, integrating mathematical modeling of dual-sided fault impacts on I-V curves and a 180-day synthetic dataset with randomized severities. Combining I-V curve and maximum power point-based power profile analyses, it demonstrates bifacial resilience, achieving up to 100% power advantage under severe shading and 11.7–30% superior outputs across faults. The proposed two-stage CNN-RF model delivers 100% accuracy in Stage 1, classifying baseline, degradation, and obstruction, and 97.6% accuracy in Stage 2, identifying specific faults such as dust, shading, aging, and cracks, with an area under the curve (AUC) of 0.999 and false positive rate (FPR) of 0.006. It surpasses standalone CNN at 89.7% accuracy in Stage 2 with an 8.8% accuracy gain, AUC of 0.994, and FPR of 0.026, as well as RBF-SVM at 96.1% and GBRT at 91.8%, due to its synergistic feature extraction and robust ensemble classification. The model's low computational footprint enables real-time scalability for large PV farms through edge computing or cloud integration, enhancing reliability and maintenance strategies for bifacial systems.