Optimization Under Uncertainty in Building Energy Management

The introduction of decentralized energy resources as well as energy storage systems to the energy system calls for new control and coordination mechanisms and systems. This is also true for buildings. An optimized operation of buildings comprising decentralized generation and energy storage systems can be achieved by a building energy management system. It controls and coordinates the operation of individual devices in a building's energy system to achieve given goals, such as the increase of energy efficiency, the decrease of carbon emissions, the minimization of operating costs or the provision of demand response measures.

This thesis picks up on this idea and extends the ongoing research by presenting an approach to the optimized operation of building energy systems that includes the uncertainties in the predictions of the future energy generation and consumption into the control scheme of a building energy management system. To do so, this thesis identified the use of a scenario-based consideration of the uncertainties to be best suited. The presented approach uses a rolling horizon optimization approach with a stochastic two-stage optimization problem, which considers several forecast scenarios in the optimization.