Deep-sea Argo floats are essential tools for humanity's exploration, development, and utilization of deep-ocean resources, and they form a key component of the global Argo program. However, the large-scale deployment of these floats faces major challenges in reliable anomaly detection due to the interdisciplinary nature, the complexity of the marine environment, and numerous uncertainties. With the growing convergence of information technology and ocean applications, digital twin–driven approaches have shown great potential in addressing these challenges. This paper proposes a data-and model-driven digital twin framework for deep-sea Argo floats, integrating adaptive model updating with anomaly detection. Using China's independently developed Xuanwu deep-sea Argo float, certified by the international Argo program, as a case study, we demonstrate the framework's typical applications in model adaptation and anomaly detection. Sea trials data validate the feasibility and effectiveness of the proposed system.
Gao et al. (Wed,) studied this question.