In a refereed scientific journal within the publishing house (Springer) with an impact factor of 1.2 and Sitescore 3.4.
The study aimed to enhance the thermal efficiency and predictive accuracy of photovoltaic thermal (PVT) systems by utilizing advanced artificial intelligence algorithms—namely Artificial Neural Networks (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS), and Relevance Vector Machine (RVM). Experimental data were collected from a PVT system using key variables such as solar radiation, inlet temperature, wind speed, and ambient temperature.
The comparative analysis showed that the ANN model outperformed the other algorithms, achieving the highest predictive accuracy. The ANFIS and RVM models also demonstrated moderate predictive capabilities. The results highlight the significant potential of AI-driven models—particularly ANN—in improving the performance of photovoltaic thermal systems, thereby contributing to global renewable energy objectives by enhancing system adaptability and energy output.
This research contributes to the advancement of renewable energy technologies, demonstrating that AI algorithms can effectively manage complex and dynamic interactions within PVT systems.
