A digital-twin and machine-learning framework for the design of multiobjective agrophotovoltaic solar farms

This work develops a computational Digital-Twin framework to track and optimize the flow of solar power through complex, multipurpose, solar farm facilities, such as Agrophotovoltaic (APV) systems. APV systems symbiotically cohabitate power-generation facilities and agricultural production systems. In this work, solar power flow is rapidly computed with a reduced order model of Maxwell’s equations, based on a high-frequency decomposition of the irradiance into multiple rays, which are propagated forward in time to ascertain multiple reflections and absorption for various source-system configurations, varying multi-panel inclination, panel refractive indices, sizes, shapes, heights, ground refractive properties, etc. The method allows for a solar installation to be tested from multiple source directions quickly and uses a genomic-based Machine-Learning Algorithm to optimize the system. This is particularly useful for planning of complex next-generation solar farm systems involving bifacial (double-sided) panelling, which are capable of capturing ground albedo reflection, exemplified by APV systems. Numerical examples are provided to illustrate the results, with the overall goal being to provide a computational framework to rapidly design and deploy complex APV systems.