Colloquium lecture: 13 December 2022, Matthias Vietz

Symbolic picture for the article. The link opens the image in a large view.
Bild der Präsentationsfläche

QoS Analysis of Smart Distribution Grids with Network Calculus


Distribution grids are changing. Due to the expansion of renewable energies and the shift of system responsibility to the operators of these systems, several so-called “Smart Grid Services” (SGSs) need to be implemented in the distribution system itself, in contrast to the previous centralized implementation. Those functions have Quality of Service (QoS) requirements which vary in strictness and tightness of
the guaranteed bandwidth and latency. To ensure a reliable operation of the smart distribution grid, the QoS requirements must be analytically proven. This work bases its approach of analytically verifying the QoS requirements of a network on two steps, with the first step being outside of the scope of this work. In the first step, a constraint satisfaction problem (CSP) containing the network and
computing resource requirements is created and solved. In the second step, the solution of the CSP is used and transferred to a network calculus formulation. In this step, the links are represented by service curves in the network calculus and the different SGS are modeled by arrival curves. The delay bounds are calculated for each SGS and compared to its requirement. If one of the requirements is not met,
the result is returned to the CSP solver to generate a new feasible solution. The result of my work is a flexible Network Calculus framework that can be used to analyze arbitrary communication networks. It offers the possibility to modify its parameters in order to influence the model. To account for different service priorities, scheduling policies such as strict priority, weighted fair queuing, weighted round
robin, and deficit round robin are implemented. The best-performing parameters are derived for a sample network.


Time: 10:15 am

Place: room 04.137, Martensstr. 3, Erlangen



Meeting-ID: 631 7461 3617
Kenncode: 044678