An Agent Based Simulation and Optimization Model for Minimizing Costs of Renewable Energy Communities

Abstract

This paper presents a novel approach to modelling Renewable Energy Communities (REC) that integrates agent-based modelling, discrete-event simulation, and system dynamics. The model captures individual Points of Delivery (PoD) in a REC and represents aggregated REC dynamics over time, including energy flows, feedback loops, and long-term trends such as the balance between energy generation and consumption. Additionally, an optimisation model has been formulated to maximise REC performance. The model defines an objective function in order to maximise the Net Present Value (NPV) of investments while considering inflation rates and projected cash flows. As part of this work, a simulation tool has been developed that is both adaptable and scalable. Its adaptability enables parameter adjustments such as geographical location, installed photovoltaic capacity, PoD consumption, and energy prices, while its scalability supports a wide range of REC sizes, from small-scale projects to large, complex systems. The tool also integrates regulatory incentives, such as subsidies or penalties, to ensure realistic and context-sensitive outcomes across diverse regulatory frameworks. This tool has been validated through a real-world case study in Val d’Aosta, Italy. Scenario optimisation under varying energy price conditions demonstrated the tool’s effectiveness achieving substantial economic benefits. Results reveal that higher energy prices drive increased installed power, reduced external energy purchases, and higher self-consumption.