Reusing Model Validation Methods for the Continuous Validation of Digital Twins of Cyber-Physical Systems

TL;DR

The paper tackles the challenge of maintaining digital twin validity for evolving cyber-physical systems by reusing traditional model validation techniques in a runtime, continuous-validation setting. It presents a generic architecture and workflow that compare real-world operation with simulated data, using metrics such as RMSE, normalized Euclidean distance, and relative error to detect deviations and drive twin updates. Through a case study on a lab-scale gantry crane, it demonstrates that several validation metrics can reveal divergence and can be paired with parameter estimation to re-align the twin, though metric effectiveness varies by scenario and initialization. The work highlights practical considerations, such as threshold design and data limitations, and argues that continuous validation can enable the twin to evolve in tandem with its physical counterpart, offering a concrete path toward continual digital twin credibility in CPS contexts.

Abstract

One of the challenges in twinned systems is ensuring the digital twin remains a valid representation of the system it twins. Depending on the type of twinning occurring, it is either trivial, such as in dashboarding/visualizations that mirror the system with real-time data, or challenging, in case the digital twin is a simulation model that reflects the behavior of a physical twinned system. The challenge in this latter case comes from the fact that in contrast to software systems, physical systems are not immutable once deployed, but instead they evolve through processes like maintenance, wear and tear or user error. It is therefore important to detect when changes occur in the physical system to evolve the twin alongside it. We employ and reuse validation techniques from model-based design for this goal. Model validation is one of the steps used to gain trust in the representativeness of a simulation model. In this work, we provide two contributions: (i) we provide a generic approach that, through the use of validation metrics, is able to detect anomalies in twinned systems, and (ii) we demonstrate these techniques with the help of an academic yet industrially relevant case study of a gantry crane such as found in ports. Treating anomalies also means correcting the error in the digital twin, which we do with a parameter estimation based on the historical data.

Figures (18)

• Figure 1: Visual representation of the difference between a (base) computerized model, and a minimal digital twin. The twin is updated throughout its lifecycle to accurately represent the real-world system, this is achieved through continuous data collection. In contrast, the base model does not evolve.
• Figure 2: General architecture needed to enact the workflow.
• Figure 3: Visualization of the workflow. The continuous validation workflow does not include the alternative model search step, as indicated by the dotted line.
• Figure 4: System state of an intermittent and continuous process over time.
• Figure 5: Abstract illustration of the routine of a gantry crane picking up and dropping off a container.