Co-simulation based performance evaluation of ICT infrastructures for smart grids

The traditional electrical transmission system is currently fundamentally changing towards a more intelligent network, the so called smart grid. This change is caused by both, an increasing number of decentralized, renewable energy (including photovoltaic, wind energy, etc.) and controllable loads with high demands (including electric vehicles, combined heat and power generation, etc.). For the power system this change is accompanied by the need for novel mechanisms, enabling both, real-time analysis of the system's state and intelligent adjustments of the power flow in order to guarantee power system stability at all times. For realizing such systems, an efficient and high-performance communication network is necessary, which can deliver the arising information quickly and reliable, even in case of emergencies. For evaluating new protection and control systems, simulations are a common way to model the behavior of those systems, but, however, are only available for individual parts of the system. Especially with regard to smart grids, a combined evaluation becomes increasingly important to evaluate overlapping respectively mutual effects and to include the performance of the communication network, which strongly influences the efficiency of protection and control systems. The goal of this thesis is the design and evaluation of a combined modeling approach taking into account both power system and communication network simulators. An essential base for this has been presented by introducing the so called Hybrid Simulator Architecture (HSA), which provides generic functionalities and has been used for implementing the Integrated Co-Simulation of Power and ICT Systems for Real-Time Evaluation (INSPIRE) co-simulator. Furthermore, in the context of this thesis, a generic approach for enabling inter-simulator communication has been proposed and illustrated, relying on communication protocols and data models common for substation automation systems to enable an evaluation, which is extensible for future enhancements. As time management has been identified as a critical requirement for co-simulators, this thesis especially pays attention to this and presents a detailed evaluation for the applied heterogeneous simulators. Here, results have shown, that the limited access regarding the continuous-time model of the power systems simulator can lead to the introduction of synchronization errors. These errors have been evaluated within this thesis and corresponding approaches for their minimization have been proposed. The capabilities of a hybrid simulation environment, combining power system and communication network simulators, are highlighted by various case studies modeling different failure scenarios, illustrating the need for a joint reflection. Only this combined evaluation will enable the appropriate dimensioning of an efficient and reliable communication network infrastructure in future networks. Results and approaches, developed in the course of this thesis, have been documented within more than 10 publications, which have been presented at international conferences as well as published as part of national and international journals. Finally, together with other associates of research unit 1511, the resulting simulator has been applied for patent and - due to its general approach - will be adopted for future research projects within the context of smart grids.