A compositional framework for energy management of a smart grid: A scalable stochastic hybrid model for cooling of a district network

The goal of this paper is to introduce a compositional modeling framework for the energy management of a smart grid that operates connected to the main grid. The focus is on the cooling of a district network composed of multiple buildings that possibly share resources such as storages, chillers, combined heat and power units, and renewable power generators. We adopt a modular perspective where components are described in terms of energy fluxes and interact by exchanging energy. Model dimension and complexity depend on the number and type of components that are present in the specific configuration. Energy management problems like the minimization of the electrical energy cost or the tracking of some electrical energy profile can be addressed in the proposed framework via different control strategies and architectures.