Solar energy generation is often hindered by various surface conditions on solar panels, such as dust, snow, bird droppings, and physical or electrical damage, which reduce efficiency. The growing reliance on renewable energy sources emphasizes the need for effective maintenance solutions to sustain optimal performance. This research aims to address the gap in automated fault detection for solar panels by developing an IoT-driven deep learning model to accurately identify and classify these faults. Utilizing MobileNetV3 architecture, combined with data augmentation techniques, the model is trained to detect six fault types, demonstrating robust accuracy in distinguishing between conditions. The innovative use of data augmentation, transfer learning, and MobileNetV3 contributes to the model's enhanced performance and precision, providing significant advancements over traditional manual inspections. Experimental results reveal a classification accuracy of 94%, confirming the model's effectiveness in real-time fault detection. These findings underscore the model’s contribution to optimizing solar energy efficiency and maintenance cost reduction, marking a meaningful impact on renewable energy management.