ResiNet

Projektseite auf Deutsch

Project description

The collective nonlinear dynamics of power grids and their nonlinear response characteristics to external disturbances underlie the design and operation of our electrical power supply. Due to an increasing share of renewable energy sources and the related expansion of power grids, the latter are exposed to external dynamic disturbances as well as internal parameter changes across multiple time scales. Computer simulations of specific example systems in concrete states are currently a frequently used method for investigating power grid dynamics and their resilience to disturbances. However, the more meshed the grids become and the more varying parameters and system states they have, the less exhaustively can computer simulations cover all conceivable combinations of scenarios. In addition, a fundamental mechanistic understanding is not possible purely on the basis of simulations.

In this short project, we are developing approaches to counteract this limitation. We combine methods from nonlinear dynamics, statistical physics and network dynamics to simultaneously analyze entire sets of scenarios in the model. On the one hand, we design an approach that facilitates the identification of those nodes in the network that causally influence the loss of phase-locked states, i.e., normal operating states, the most. On the other hand, we are developing a concept, which mathematically predicts how power grids react to external fluctuations on average for all system parameters simultaneously. Such methods can help to predict the resilience of power grid dynamics against both parameter (i.e. component or operational state) changes and external fluctuations, thereby contributing to resilient design and operation.