A three-stage stochastic optimization model for the design of smart energy districts under uncertainty

We propose a three-stage stochastic integer programming model to tackle the design of smart energy districts, including electricity and heat storage, conversion and distribution systems, under uncertainty. The model allows to account for the uncertainty in the short-term forecasts, the day-ahead electricity bidding, the day ahead scheduling of large power plants and the possibility of real-time scheduling adjustments of flexible energy systems (integer recourse). The application to a case study shows the complexity of the associated mixed integer linear program (MILP) and the need for ad hoc decomposition techniques.