Development of a Predictive Model for Evaluating the Performance of Low-Concentrator Photovoltaic Thermal Module Using Design of Experiments

The performance of photovoltaic modules operating under low-concentrated illumination is influenced by several factors including wind speed <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(U_{wind})$</tex>, solar radiation <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(G)$</tex>, inlet water velocity <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(V_{in})$</tex> and temperature <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(T_{in})$</tex>, adhesive thermal conductivity <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(K_{adh})$</tex>, solar concentration ratio <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(CR)$</tex>, and module reference efficiency <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\eta_{ref})$</tex>. This study proposes a prediction model for evaluating three responses including the average photovoltaic module temperature, electrical power generation, and gained thermal power of a Low Concentrator Photovoltaic Thermal (LCPV/ T) module in terms these influencing factors. The prediction models are obtained using the response surface methodology (RSM). The RSM tool created 152 combinations of these designs and operating conditions. The responses at these 152 runs were evaluated using a validated computational model. The RSM tool also evaluates the most significant factors influencing these responses. Finally, various single and multi-objective optimization scenarios are proposed to optimize the <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\text{LCPV}/\mathrm{T}$</tex> system performance. The results showed that at the worst operating condition of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$G$</tex> of 1000 <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{W}/\mathrm{m}^{2}, U_{wind}$</tex> of 0 m/s, <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$CR=3, T_{in} =20^{\circ} \mathrm{C}, V_{i n}=0.01 \mathrm{m} / \mathrm{s}, K_{adh}=3.7 \mathrm{W} / \mathrm{m}. \mathrm{K}$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\eta_{ref}=22\%$</tex>, the maximum generated electrical and thermal power are around 439.051 W/m<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> and 2018.1 W/m<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>, respectively.