Data-Driven Reachability Analysis of Digital Twin FMI Models.

Digital Twins are an emerging technology which makes it possible to couple cyber-physical assets with their virtual representation in real-time. The technology is applicable to a variety of domains and facilitates a more intelligent and dependable system design and operation. In this paper, we address the challenge of analysing Digital Twins by proposing a simulation-based reachability analysis of models based on the Functional Mock-Up Interface standard. The analysis approach uses simulations to obtain the Lipschitz constant of the model which is then used to compute reachable states of the system. The approach also provides probabilistic guarantees on the accuracy of the computed reachable sets that are based on simulations of the system from random initial states.