A solar hybrid photovoltaic thermal (PVT) system is a set of combined solar collectors that include a photovoltaic module (PV) and a solar panel in the same frame. When the absorption of solar radiation on the PV cell's surface is low, the system's efficiency decreases. Hence a novel Selective Response Surface Methodology for Hybrid Anti Reflective Coated Nanofilm Filter Thickness Optimization is proposed for increasing the electrical and thermal efficiency of a Hybrid PVT System. Several nano-film filters are used on the surface of the existing hybrid PVT system to capture solar energy, but they degrade over time, and temperature changes reduce the spectrum radiation absorption capabilities and path length of light rays inside the filter. To overcome these issues a novel Hybrid Anti Reflective Coated Nanofilm Filter is proposed in which a Polycarbonate nano-film filter is utilized since its refractive index is unaffected by temperature changes. This nano-film filter is coated with Zirconium oxide (ZrO2), which improves the path length of light rays inside the nano-film filter, and a Cerium Oxide (CeO2) nano-coating is utilized over this coating to lower the quantity of light reflected off the surface of the solar panel. Furthermore, existing approaches for calculating nano-film filter thickness do not consider multi-collinearity , resulting in inaccurate forecasts and disappointing outcomes. To resolve these concerns, the Selective Neuro RSM-Taguchi Optimization method is developed, which employs an MFFNN (Multilayer Feedforward Neural Network) to anticipate the Electrical and Thermal Efficiency of a Hybrid PVT system based on input parameters. Then the Response Surface model is used to choose the parameters for optimization, and Taguchi Optimization is used to selectively identify the Nano-film filter thickness while taking multi-collinearity into account. The ideal thickness obtained by these optimization methods enhances the efficiency and performance of the PVT system.
Read full abstract