Experimental and AI-based prediction of a solar air heater with novel recycled interlocking channel fins

Engineering notes

Engineering notes will be added by the Power for Solar editorial team.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为光伏效率、钙钛矿太阳能电池、储能技术、太阳能热利用、BIPV、并网技术等高价值论文补充中文说明。

Original abstract

Solar air heaters are a promising method used for drying and heating applications owing to their low operating costs and simple design. However, these solar air heaters exhibit limited thermal performance due to the low heat transfer coefficient between the absorber and the air passing through the duct, resulting in a decrease in thermal and exergy efficiency. The present work aims to address this issue by conducting a single-pass solar air heater with novel recycled aluminum interlocking channel fins, along with using artificial intelligence approaches to predict thermal and exergy efficiencies. The novelty of the present study is represented by integrating a comprehensive experimental assessment using the 4E analysis (Energy, Exergy, Economic, and Environmental analysis) with the development and comparison of artificial neural network (ANN) and deep neural network (DNN) models to predict efficiency under different operating conditions. Two solar air heaters were tested: a conventional (C-SAH) and a modified finned (M-SAH) heater, under natural and forced convection conditions at mass flow rates of 0.0046, 0.008, and 0.012 kg/s. The study findings reveal a significant enhancement in the modified heater. Under natural convection conditions, the outlet air temperature increased from 78 °C to 84 °C for the modified heater. Regarding thermal efficiency, the modified heater exhibited the highest efficiency of 48.6% at a flow rate of 0.012 kg/s. Daily thermal efficiency also increased from 32.91% to 44.36% at the same flow rate. The exergy efficiency reached a maximum of 2.80% for the modified heater. The AI models achieved high predictive performance; the DNN model achieved an R² of 0.924 for thermal efficiency, while the ANN model performed best in predicting exergy efficiency with an R² of 0.971. These results demonstrate the reliability of the proposed AI models in predicting the performance of SAHs. Additionally, the use of recycled aluminum fins significantly enhances the performance of SAHs, offering a low-cost and sustainable tool for solar air heating applications. The integration of artificial intelligence methods enhances the design and operation of highly efficient solar air heater systems.

Links and sources

• PDF from original source

Comments