Spectral-absorptance-based temperature model and its differentiated performance in commercial crystalline silicon PV modules

This study proposes a spectral-corrected PV module temperature model that explicitly accounts for wavelength-dependent absorption over 280–2500 nm, including sub-bandgap absorption as a direct thermal source. The model is parameterized using measured absorptance spectra of bifacial PERC, HJT, and TOPCon modules and validated with outdoor measurements from six commercial modules. The results demonstrate improved predictive performance, with root-mean-square errors reduced by up to 0.11 for PERC modules and maximum temperature deviations limited to 1.64 °C. Incorporating spectral absorption consistently enhances temperature prediction accuracy and mitigates systematic bias across a wide range of irradiance levels, air masses, and ground reflectance conditions. Neglecting spectral mismatch leads to temperature overestimation of up to 0.3 °C at large air masses (AM 10), while installation-related effects further amplify deviations to 0.94 °C over high-albedo surfaces such as snow. Beyond accuracy gains, the model reveals distinct technology-dependent thermal behavior, with HJT modules operating at slightly higher temperatures than PERC and TOPCon, and these differences becoming more pronounced under high irradiance and high-albedo environments.Overall, the proposed approach provides a physically consistent framework for PV thermal modeling, enabling more reliable temperature inputs for energy yield simulations and improved assessment of temperature-driven degradation risks in modern crystalline-silicon PV systems.