AGRO – Agrivoltaics Layout GeneRation with Optimization Framework

The widespread adoption of ground-mounted photovoltaic systems is facing increasing opposition due to conflicts over land use with agriculture. Agrivoltaics (AV) offers a synergistic solution; however, current design methodologies often treat agricultural constraints and panel layouts as static boundaries, resulting in suboptimal trade-offs between PV-energy conversion and the irradiance available to crops. This study addresses this issue by proposing AGRO: a comprehensive optimization framework that combines the Non-dominated Sorting Genetic Algorithm (NSGA-II), the EnergyPlus dynamic simulation engine, and Python algorithms, treating both agricultural and PV layouts as active decision variables rather than fixed boundary conditions. Applied to a representative Mediterranean case study constrained by Italy’s regulatory framework, the co-optimized layouts improve the trade-off between PV energy generation density and cumulative solar irradiance on cultivated soil, with the system adapting its morphological configuration to protect crop irradiance even during critical seasonal periods for PV-energy conversion. The variable-layout strategy proves effective not only in raising ground-level irradiance but, more distinctly, in making it more uniform across the cultivated cells. This work shows that dynamic spatial allocation enables PV systems to fully comply with Italian legislative constraints while enhancing the solar irradiance available to crops.