Distributed Optimization Methods for Coupled Energy Systems

Efficient control of energy systems is an important factor in achieving the
CO2-emission goals. District heating (DH) networks are an especially relevant
example of such energy systems. State-of-the-art control of small
and medium-sized DH networks, however, still mainly rely on simple
rule-based control concepts. Handling future challenges such as varying
prices and intermittent renewable production is difficult to achieve with
such control concepts. Optimization-based energy management systems (EMS)
are a promising high-level control approach for the efficient operation of DH
networks and complex energy systems in general. An especially interesting
challenge arises when DH networks grow, as often the opportunity arises to
interconnect them. However, if they operated by different owners, the control
task becomes challenging, especially for optimization-based EMS. This is
because, in the overall objective function, the cost and revenue for any
exchange of energy would cancel out. This thesis presents a solution to this
challenge. The main focus of this thesis is on the application of distributed
optimization methods for EMS in the context of coupled energy systems,
operated by multiple owners, especially interconnected DH networks. The
presented methods and ideas are evaluated on a practical application of
three DH networks in Austria.