Evaluating uncertainty of shared energy in solar energy communities using a stochastic simulation framework

Properly sizing renewable energy sources is crucial for ensuring their techno-economic viability, especially under policies promoting solar power through photovoltaics (PV). The inherent variability of PV production requires assessing energy yield and self-consumption at different time scales, along with their associated uncertainties, to evaluate technical performance and financial risks. This challenge is critical for solar renewable energy communities (RECs), where energy sharing determines performance quality. This work introduces a framework that quantifies the impact of PV's stochastic nature on energy sharing uncertainty in RECs. Tested across seven locations in Italy leveraging PVGIS data, the framework integrates pathintegral and Fokker-Planck formalisms with a Monte Carlo approach, and is demonstrated to effectively capture production variability and energy yields. For each location, 10,000 synthetic profiles were generated for a 50 kW peak power plant connected to the grid, serving 100 residential consumers with a typical consumption profile representative of the area. The relative uncertainty in yearly shared energy proved to range from 2 % to 3 %. Comparisons with benchmark methods, like averaged hourly production (AHP) and typical meteorological year (TMY) profiles, revealed an systematic overestimation of shared energy during months of production surplus, underscoring the need of accounting for stochasticity in energy modeling.