Comprehensive VIPV energy yield and MPPT evaluation under realistic dynamic shading in urban environments

This study evaluates the performance of a vehicle-integrated photovoltaic (VIPV) module under realistic dynamic shading along urban routes. A multistage framework combines image-based shadow extraction, irradiance and thermal modelling, and electrical simulation of two common interconnection schemes (series and total-cross-tied, TCT). Five routes covering different seasons and times of day were analysed, and four zone types—dense trees, scattered trees, open low-rise, and open midrise—were identified to contextualize shading behaviour. Results show that shading factor is the primary driver of performance losses, with winter routes exhibiting lower yield due to longer shadows despite cooler temperatures. TCT consistently outperforms series, particularly under highly non-uniform, dendritic winter shadows. Power spectral density analysis reveals that most power fluctuations occur below 24 Hz, enabling effective tracking by well-tuned P&O algorithms. Fixed-voltage control provides a strong baseline, while an artificial neural network evaluated on highly-dynamic sections offers modest improvements only in the TCT configuration.