Bayesian inverse problems with conditional Sinkhorn generative adversarial networks in least volume latent spaces
Year: 2025
Type: article
Source: Neural Networks
Authors: Qiuyi Chen, Panagiotis Tsilifis, Mark Fuge
Language: English
Subfield: Statistical and Nonlinear Physics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: bronze
Awards: G2442722421, CAREER: Learning Design Representations: The Effect of Differential Geometric Manifolds on the Inference of Design Structure, MASSACHUSETTS INSTITUTE OF TECHNOLOGY BASE AWARD
PROJECT TITLE:
''HIGH FIDELITY DIGITAL TWINS FOR BWRX-300 CRITICAL''
CONTROL NO: 21741542
THE PROPOSED WORK WILL SUPPORT THE ADVANCEMENT AND DEMONSTRATION OF NEW PREDICTIVE MAINTENANCE APPROACHES AND MODEL BASED FAULT SYSTEM DETECTION TECHNIQUES BY GEH AND GER. THE PROPOSED DIGITAL TWINS FUNDAMENTALLY ADDRESS MECHANICAL AND THERMAL FATIGUE FAILURE MODES WHICH DRIVE O&M ACTIVITIES WELL BEYOND THE SELECTED COMPONENTS IN BWRX-300, AND ARE EXTENDABLE TO ALL ADVANCED REACTORS (ARS) WHERE A FLOWING FLUID IS PRESENT.
OpenAlex ID: W4411589436