Efficient and Certified Model Reduction for Parametrized Dynamical Systems

In the current presentation, we show that some characteristic components of Reduced Basis Methods can be transfered to model reduction of parametrized linear dynamical systems. The key for efficiency of the reduced scheme is an offline/online decomposition, which is obtained in case of affine parameter-dependence of the system components. Additionally, certification of the reduced system solution is possible by guaranteed, rapidly computable and tight a-posteriori error bounds. Experiments demonstrate the applicability of the reduction scheme, the reliability of the error estimators and the runtime gain.