Assessing the potential of forecast-based optimization in renewable energy communities with flexible electricity, heat and mobility resources

Renewable Energy Communities (RECs) allow members to share self-produced renewable electricity with other consumers via the public grid. In this way, RECs can effectuate a more efficient local electricity use and reduce the load on the higher-level grid. However, to unlock this potential, intelligent control systems must be established to coordinate operational decisions within communities with foresight. The central inputs for such systems are forecasts, which inevitably contain errors. Thus, it is crucial to gain a better understanding of their effects on forecast-based optimization. We developed a comprehensive framework combining an optimization model with a simulation environment that permits a realistic representation of forecasting errors. The optimization model is a Mixed-Integer Linear Program that covers all relevant energy sectors for consumers, namely flexible appliances, stationary batteries, electric vehicles with bidirectional charging, and electric heating systems with thermal energy storage. The optimization model uses short-term forecasts to determine optimal control actions for all elements of the REC. These are fed into the simulation, which determines the actual outcomes based on updated realization values of the environment, possibly deviating from original forecasts. As a comparative benchmark we also implemented a static rule-based algorithm as default control option in the simulation model. A scenario analysis showed that forecast-based optimization leads to considerable performance improvements, especially in RECs with a high degree of flexibility. In this case, the community-wide self-sufficiency and self-consumption could be increased by approx. 10 and 20 %-points, respectively. As a positive side effect, the maximum grid feed-in is reduced by 14 %. • Smart coordination of time-flexible resources for all members of an energy community. • Optimization of shiftable loads, batteries, electric vehicles and heating systems. • Open-source simulation framework with realistic representation of forecast errors. • Forecast-based optimization increases self-sufficiency and reduces network load.