Recent conversion efficiency breakthroughs in double-junction (tandem) perovskite/crystalline silicon solar cells demand advanced opto-thermo-electrical simulations, that are critical for translating laboratory results into realistic photovoltaic module and system performance. A holistic framework is here developed and presented, combining cell-level simulations, spectral analysis, PV module and PV system modelling. After validating the deployed physics models against measured cells and modules, hourly spectral irradiances for Delft, the Netherlands, and Catania, Italy, are generated and clustered into representative “blue-rich” and “red-rich” spectra. The effects of spectral variations on the current-matching and energy yield of tandem modules are quantified. Realistic module architectures are simulated, integrating dynamic temperature and spectrum data. Temperature coefficients are derived as a function of both irradiance and module temperature, significantly improving upon traditional indoor-derived values. Results show that standard indoor-derived coefficients under-/overestimate values in realistic conditions, highlighting the ultimate need for location-specific power matrixes. This study offers a robust pathway to predict tandem module energy yields across seasons and climates, supporting optimized design choices for industrial production and future PV installations.

Photovoltaic (PV) technology is raising attention as a low-cost green energy source. It mainly finds applications in solar fields, on building facades and on rooftop. One of the main issues that can occur is the shading of solar cells inside the photovoltaic module which could affect the maximum power output of the PV panel and the lifetime of the cell itself. In order to predict the behaviour of PV panels in partial shading conditions, simulations and then measurements on two different photovoltaic modules have been carried out and compared. Data have shown that the maximum power output of the panels under 1sun illumination can be predicted by simulation with a 3% discrepancy from measured values, independently from the type of technology and interconnections of the PV module.