Yantao Luo
PhD thesis title: Development of Data-Driven Model Order Reduction Techniques for Multi-Physics Problems in Nuclear Systems
Academic Tutor: Francesca Celsa Giacobbo
Academic Supervisor: Antonio Cammi
PhD cycle: 40° (see all student profiles of the same cycle > LINK)
BSc: Nuclear Engineering and Nuclear Technology, Southwest University of Science and Technology
MSc: College of Nuclear Science and Technology, Harbin Engineering University
Thesis abstract
My research focuses on developing Model Order Reduction (MOR) techniques for multi-physics (MP) problems in nuclear systems, specifically targeting the coupled neutronics-thermal-hydraulics framework. Due to the inherently high computational cost of such simulations, this project proposes MOR integrated with machine learning to significantly enhance computational efficiency. This advancement aims to facilitate the development of digital twin nuclear reactors, enabling fast-response predictions while maintaining observable accuracy.
Personal interest in my research theme
I am deeply fascinated by multi-physics problems because of their complex and intriguing mutual feedback mechanisms, which merit thorough investigation across various reactor types. While my master’s thesis concentrated on explicit two-code coupling, the one-code coupling scheme greatly appeals to me, as it eliminates the need for data mapping by sharing a unified mesh for both physical fields. Given the time-consuming nature of these simulations, I am highly motivated to employ ROM and machine learning approaches to reduce computational burdens and accelerate innovation.
