A Computational Comparison and Validation Between Ray Tracing Techniques Under Special Light-Sharing Trade off Scenarios in Photovoltaics

Emerging production models like agrivoltaics offer a promising path for expanding renewable energy and achieving climate-neutral goals. However, these innovative systems introduce an unexplored light-sharing tradeoff, creating uncertainties in both energy and crop production. Profitability hinges on both outputs, requiring accurate modeling of ground-level irradiance to predict future crop yields. Thus, the economic assessment of agrivoltaic systems depends heavily on the precision of irradiance models for further analysis. Ray tracing techniques offer an efficient solution for simulating light distribution in complex scenarios, using parallelized ray trajectories to estimate irradiance beneath solar modules and across diverse environments. However, their high computational demand presents a challenge to widespread adoption. Balancing accuracy with computational efficiency is crucial for advancing research in this area. Validation experiments can help optimize algorithms, making these tools more accessible and cost-effective. This study presents a computational analysis of two ray tracing tools: the Python Agrivoltaic Simulation Environment (PASE) and NREL's Bifacial Radiance. We assess their computational resource requirements, estimate effort, and compare results with field data from Werbomont, Belgium, under different PV array locations. This analysis provides insights into light distribution under the modules, with implications for crop production and biodiversity cultivation.