Real-Time Capable Model Updating of Digital Twins: A Case Study of a Cantilever Beam

Digital twins are virtual representations of physical systems and are becoming increasingly important in structural health monitoring of engineering structures. However, the applicability of a digital twin highly depends on its ability to stream and process real-time measurement data from the physical system to the digital representation. A data streaming strategy based on Apache Kafka for a real-time capable implementation of digital twins is presented, focusing on model updating using a cantilever beam as a case study. In this setup, measurement data are transferred to a server every second, and modal parameters are identified every 10 s. These identification results are fed into the digital twin of the system, which is continuously updated to accurately reflect the physical behaviour of the cantilever beam. This contribution demonstrates that the condition of the structure can be accurately identified using a digital twin, allowing for real-time decision making.