GEOMETRIC AND AI APPROACHES TO MODELLING THE MULTI-SCALE JOVIAN MAGNETOSPHERE
Seminars
Semester 2
The Jovian magnetosphere, governed by multi-scale coupling, nonlinear interaction mechanisms, and intricate geometric configurations, stands as one of the most complex planetary plasma systems in the Solar System. Traditional modeling frameworks face challenges in balancing the system's inherent complexity with long-term simulation stability and computational feasibility. This talk presents an integrated approach combining structure-preserving geometric algorithms to resolve cumulative errors in long-term simulations with high-performance computing to realize multi-scale magnetohydrodynamic and kinetic hybrid simulations. Additionally, we explore the potential of artificial intelligence (AI) to unify heterogeneous datasets—including in situ measurements, remote observations, laboratory experiments, and existing simulation outputs—into a cohesive knowledge framework. This methodology establishes a dynamic, data-adaptive research paradigm for the Jovian magnetosphere, providing new insights into the cross-scale coupling mechanisms that govern its evolution.
Biography Prof. Jian LIU is a Taishan Scholar Distinguished Expert, CAS Outstanding Young Scientist, and Shandong University Distinguished Young Scholar, as well as a Zhai Guanglong Scholar at USTC. He serves as a Standing Committee Member of the Simulation Algorithms Committee at the Chinese Association for System Simulation, Director of the Advanced Algorithms Joint Laboratory at the National Supercomputing Center in Jinan, and member of the Coordinating Group for Magnetic Confinement Fusion Integrated Simulation Platform. He earned his B.S. in Physics (inaugural Yuanpei Honors Program) from Peking University in 2005 and a Ph.D. in Plasma Physics (2012) from Peking University, with a joint doctoral fellowship at Princeton University (2010–2012). His research focuses on plasma artificial intelligence, large-scale numerical simulations, high-performance computing, and structure-preserving algorithms. He has pioneered work in AI for Science, simulation software development, and multiscale physics of complex systems. Prof. LIU has led 15 national research projects, including the National Key R&D Program, NSFC grants, and CAS Frontier Science Initiatives, while contributing to multiple major scientific engineering projects.
Additional information: Prof. Binzheng ZHANG, binzh@hku.hk