Energy Simulation Models of a Photovoltaic-powered Energy Community

A Renewable Energy Community (REC) is a group of users, consumers, and prosumers who come together to produce and consume renewable energy in order to reduce the costs and consumption of non-renewable energy. Photovoltaic production, net of CO2 emissions related to the construction of the plant, is a clean energy source, ideal for REC, to produce energy to be shared among members, however, the variability of solar radiation still represents a major challenge in managing photovoltaic energy production. The aspect of energy production forecasting is increasingly a crucial aspect to limit imbalances in the electricity grid, optimize the operation of generation, load and storage resources. The approach we propose is effectively adapted to solar plants such as small-scale ones installed in condominiums. For the latter, it is possible to develop an initial forecasting model based on the technical characteristics of the plant (such as nominal power and panel orientation), which can then be refined using machine learning techniques and the use of historical generation data. Once trained, the best-performing models require relatively limited computational resources to generate accurate forecasts. A photovoltaic forecasting model for RECs must be accurate to allow optimal management of energy production and consumption within the community. These models predict photovoltaic daily output to ensure that the energy produced meets the needs of the community, minimizing losses and increasing efficiency. The study aims to develop integrated energy systems, optimizing the interaction between different energy sources (renewable and conventional) and energy vectors (electric and thermal) to maximize overall efficiency and improve the operational management of energy microgrids, thus accelerating the transition to a sustainable energy future.