Physics-guided machine learning predicts the planet-scale performance of solar farms with sparse, heterogeneous, public data

The photovoltaics (PV) technology landscape is evolving rapidly. To predict the potential and scalability of emerging PV technologies, a global understanding of these systems’ performance is essential. Traditionally, experimental and computational studies at large national research facilities have focused on PV performance in specific regional climates. However, synthesizing these regional studies to understand the worldwide performance potential has proven difficult. Given the expense of obtaining experimental data, the challenge of coordinating experiments at national labs across a politically divided world, and the data privacy concerns of large commercial operators, a fundamentally different, data-efficient approach is desired. Here, we introduce a physics-guided machine learning (PGML) approach for PV to demonstrate that: (a) the world can be divided into a few PV-specific climate zones, called PVZones, illustrating that the relevant meteorological conditions are shared across continents; (b) by exploiting the climatic similarities, high-quality monthly energy yield data from as few as five locations can accurately predict (with a root mean square error of less than 8 kWh m−2) global yearly energy yield potential at high spatial resolution. Moreover, by homogenizing noisy, heterogeneous public PV performance data, the global energy yield can be predicted with less than 6 % relative error compared to physics-based simulations, provided that the dataset is representative. This novel data-efficient PGML scheme for PV is independent of both PV technology and farm topology, allowing it to adapt seamlessly to emerging PV technologies and farm configurations. The results pave the way for physics-guided, data-driven collaboration between national policymakers and research organizations in developing efficient decision support systems to accelerate PV deployment worldwide.