Generalized graph foundation models as versatile data-driven digital twins for complex technological systems

作者：Benjamin G. Pierce, Hein Htet Aung, Thomas G. Ciardi, Kristen J. Hernandez, Raymond Wieser, Weiqi Yue, Yangxin Fan, Alexander C. Harding Bradley, Balashanmuga Priyan Rajamohan, Erika I. Barcelos, Jayvic C. Jimenez, B. K. Spears, Brian Giera, Robert X. Gao, Mengjie Li, Kristopher O. Davis, Laura S. Bruckman, Yinghui Wu, Pawan K. Tripathi, Roger H. French · 发表于：Scientific Reports · 年份：2026 · DOI：10.1038/s41598-026-61113-5 · 研究领域：Digital Transformation in Industry、Model Reduction and Neural Networks、Graph Theory and Algorithms

Digital twins are comprised of computational models that mimic the 'as built' characteristics of devices, systems, and networks of systems whose performance in the real world warrants quantitative and critical assessment.The literature on constructing digital twins is historically focused around task-specific, physics-based models that seek to understand the device from first principles, thereby constructing an idealized digital twin, based on the known physics, of the system.However, these "physics-based digital twins" (pbDT) can be quite difficult to construct, as the level of detail required to accurately model the complex physics of many devices is often missing or expensive to obtain, especially for systems with widespread deployment.Additionally, pbDTs generally assume the device is working as intended; in practice, many systems experience some form of performance degradation, or derating, that causes them to operate off-specification, in manners such that the basic physics is undetermined.As it is often the goal of a digital twin model to quantify these departures from the idealized system, it is quite difficult to separate assumptions from the expected model output.In contrast, data-driven digital twins (ddDT) seek to model the system as it actually is based on real observations and datastreams arising from the device in question.ddDTs enable agility in responses because they learn system dynamics directly from data at operational timescales. As a result, complex phys...