AbstractsPhysics

The energy transition, from a centralized to a decentralized and sustainable power supply using small scale power plants, presents new challenges to the distribution grid provider who is responsible for maintaining the stability of the electricity network. Furthermore, the rapid uptake of power generation from residential photovoltaic panels and wind turbines, together with decreasing prices for residential storage devices, is likely to lead to a reorganization of the energy market. Thus, new procedures to ensure the overall network stability need to be developed, which are flexible with respect to the underlying network and scalable, to be able to handle the amount of data of a fast growing network of renewable energy producers. To this end, we consider (distributed) model predictive control (MPC) and hierarchical distributed optimization algorithms. We examine a network of residential energy systems (RESs) where every resident is equipped with solar photovoltaic panels and local storage devices, i.e., each RES is consuming, generating, and storing power. The RESs are connected through a grid provider responsible for the stability of the overall network. We propose three different hierarchical distributed optimization algorithms. The flexibility of the algorithms allows for a plug and play manner of implementation. Scalability is obtained by solving the optimization problems on the level of the RESs and not on the level of the grid provider. Furthermore, with respect to a specific centralized optimization problem, convergence of the distributed optimization algorithms to the central optimum can be proven. In addition, we show how distributed optimization can be used to obtain a real-time pricing scheme depending on the power supply and the power demand, in contrast to the static pricing schemes in current widespread use. It is verified numerically that the properties of the open-loop solutions carry over to the closed-loop by embedding the distributed optimization algorithms in receding horizon schemes. The results are illustrated using a dataset on power generation and power consumption of residential customers of the company Ausgrid. Deutsche Fassung: Die Energiewende, die von der Energieerzeugung aus zentralen Großkraftwerken zur Energiegewinnung aus dezentralen Kleinkraftwerken führt, stellt die Netzbetreiber vor neue Aufgaben bei der Sicherung der Netzstabilität. Der Anstieg der Stromerzeugung aus privaten Solarzellen und Windkraftanlagen zusammen mit dem immer größer werdenden Angebot an privaten Energiespeichern führt zu einer Umstrukturierung des Energiemarktes. Dies macht neue Methoden bei der Gewährleistung einer stabilen Stromversorgung erforderlich, die flexibel bezüglich des Netzwerkes einsetzbar sind und gleichzeitig gut skalierbar sein müssen, um die Datenmenge eines schnell wachsenden Netzwerkes aus erneuerbaren Energieerzeugern handhaben zu können. Diese Arbeit setzt sich in diesem Kontext mit der modellprädiktiven Regelung (MPC), beziehungsweise mit der verteilten modellprädiktiven… Advisors/Committee Members: Grüne, Lars (advisor).