Condition Monitoring, Prognostics & Digital Twins

Beyond detecting that a fault exists, ISACLAB researches how to predict when a failure will occur and how to track asset health continuously. We build physics-based digital twins for supervisory monitoring of turboshaft engines, nonlinear-observer and Bayesian prognostic algorithms for compressor failure prediction in heavy-duty gas turbines, and multisensor security-management frameworks for steam turbines. This extends into hardware: we design and fabricate our own sensing instrumentation, such as low-frequency inductive velocity transducers for large-scale steam turbine monitoring, closing the loop between physical measurement and predictive modeling. Work in this area has appeared in Digital Twin and the Journal of Theoretical and Applied Vibration and Acoustics.