Software-Defined Networking for Real-Time Capable, Reliable and Cost-Efficient Smart Grid Communication Infrastructures

Changeover to green energy generation requires precise monitoring and control of power systems at all times, relying on appropriate Information and Communication Technology (ICT) infrastructures. Such communication networks need to fulfill diverse requirements, ranging from ultra-reliable, real-time data transfers on the transmission power grid level, to cost efficiently connecting tens of thousands of intelligent distribution grid devices per square kilometer. This dissertation aims at providing a comprehensive view on power system communications, applying different methods and techniques for evaluating and enhancing its overall performance. This involves analytical and simulative approaches as well as empirical measurements. Network Calculus (NC) is utilized to obtain analytical worst-case bounds on the delay of time-critical intra- and inter-substation communications in transmission power grids. Reliability and real-time capability of power system communications are advanced, applying the concepts of Software-Defined Networking (SDN) for dynamic network control. Finally, a techno-economic modeling framework is created to assess the cost-efficiency of different communication solutions under consideration of technical constraints. In particular, the economic benefits of utilizing shared, SDN-enabled communication infrastructures are analyzed.