Learning stable reduced-order models for hybrid twins

To view the full version of this page (or to download files), please log in or create a new account

2021年9月15日 System Modeling

Abstract. The concept of “Hybrid Twin” (HT) has recently received a growing interest thanks to the availability of powerful machine learning techniques. This twin concept combines physics-based models within a model-order reduction framework—to obtain real-time feedback rates—and data science. Thus, the main idea of the HT is to develop on-the-fly data-driven models to correct possible deviations between measurements and physics-based model predictions. This paper is focused on the computation of stable, fast and accurate corrections in the Hybrid Twin framework. Furthermore, regarding the delicate and important problem of stability, a new approach is proposed, introducing several sub-variants and guaranteeing a low computational cost as well as the achievement of a stable time-integration.


Abel Sancarlos, Morgan Cameron, Jean-Marc Le Peuvedic, Juliette Groulier, Jean-Louis Duval, Elias Cueto and Francisco Chinesta


Hybrid Twin, machine learning


ダウンロードsancarlosetalupdated.pdf 1.53 MB