Residential city districts as flexibility resource : analysis, simulation, and decentralized coordination algorithms

  • Wohnquartiere als Flexibilitätsressource : Analyse, Simulation und dezentrale Koordinationsalgorithmen

Molitor, Christoph; Monti, Antonello (Thesis advisor); Moser, Albert (Thesis advisor)

1. Aufl.. - Aachen : E.ON Energy Research Center, RWTH Aachen University (2015, 2016)
Book, Dissertation / PhD Thesis

In: E.On Energy Research Center : ACS, automation of complex power systems 32

Zugl.: Aachen, Techn. Hochsch., Diss., 2015


This dissertation centers on residential city districts as source of flexible electrical energy demand and generation. More specifically, the focus is on the exploitation of operational flexibility provided by so-called electro-thermal heating systems like heat pumps (HPs) and combined heat and power (CHP) systems in combination with inexpensive thermal storage. Due to the high share of primary energy related to space heating in residential buildings, the electrification of the heating system is considered as a significant source of flexibility. First, this dissertation investigates the load shaping capability of clusters of Building Energy Systems (BESs) with different shares of electro-thermal HSs under a coordinated operation. The presented results show that heterogeneous clusters, comprising equally BESs with CHP and HP systems, provide the most load-shaping capability. Furthermore, the analyses revealed that clusters comprising less than 50 BESs are already sufficient in order to significantly provide load shaping. Thus, the clusters of BESs are capable to address local grid conditions like imbalances, related to the associated city district.Second, this thesis proposes a decentralized coordination approach, which has several advantages over existing approaches regarding data privacy, plug and play functionality and reduced demand for central computing units. In this approach, each BES, represented by a software agent, solves in a first step a local optimization problem, yielding a set of optimal or near-optimal operating schedules. In a second step, each agent determines by message exchange with other agents the actual schedule supporting the system objective the most. The simulation-based analysis shows that the algorithms exhibit a fast convergence and a good scalability. Third, a co-simulation platform is presented enabling the analysis of holistic simulations of city district energy systems. The simulation platform, referred to as MESCOS, allows simulating city district energy systems, comprising dynamic buildings models, energy supply infrastructure and control and energy management algorithms. In order to enable the simulation of large city district energy systems comprising hundreds of BESs, the platform exploits parallel computing features of modern simulation servers while at the same time reusing existing modeling and simulation tools. Finally, the simulation of a sample city district demonstrates the insights which can be gained by means of the introduced simulation platform.