Critical infrastructure can be defined as organizations and facilities that are of vital importance to the state community and whose failure or impairment would lead to permanent supply bottlenecks, significant disruptions to public safety or other dramatic consequences. Just think of transportation (bridges, train stations, etc.), electricity and water grids or - in the collective consciousness since the start of the war in Ukraine in March 2022 - the gas supply. In short, critical infrastructures are the indispensable lifelines of modern societies and their protection is of the utmost importance for both our prosperity and security. In this talk, the central research themes and application fields alongside some first real-life results of the research consortium RISK.twin established at the University of the Bundeswehr Munich in 2021 will be presented. The RISK.twin project brings together scientists from the engineering sciences, computer science and mathematics, as well as from the social and political sciences. The focus of the talk will be on two application fields, namely digital twins for health monitoring of bridges and for control and incident handling in wastewater treatment facilities, thus representing two central sectors of critical infrastructure. Among the highlighted developments, we will primarily emphasize the underlying enabling technologies such as physics-based, data-driven and hybrid modeling, model order reduction and surrogate modeling as well as physics-informed machine learning (ML). The aim of the presentation is to span an arc from basic research to concrete application and the benefits that can be achieved through simulation methods and ML in critical infrastructure protection.