Energy management in a multi-building set-up via distributed stochastic optimization
We address a cooling energy management problem in a multi-building setting where buildings need to maintain comfort conditions for the occupants by keeping their zones temperature within a certain range. To this purpose, each one of them has its own chiller and is connected to a shared cooling network. The goal is to minimize the overall district electricity cost over some finite time horizon by optimally setting the temperature set-points in the buildings and the energy exchange with the cooling network, compatibly with comfort and actuation constraints, while accounting for uncertainty, mainly due to outside temperature, people occupancy, and solar radiation. To this purpose, a distributed version of the scenario approach to stochastic constrained optimization is adopted, which allows to guarantee by design a predefined robustness level of the obtained solution against uncertainty.