Shunyuan Mao

University of Victoria

Shunyuan Mao is an astronomy Ph.D. student at the University of Victoria working with Dr. Ruobing Dong. He used machine learning to solve partial differential equations describing the evolution of protoplanetary disks in his master’s study at the University of Victoria. He earned his bachelor’s degree in astronomy from the University of Science and Technology of China, where he became fascinated by the universe and artificial intelligence.

Shunyuan Mao developed a novel tool with state-of-the-art machine learning techniques that extraordinarily speed up the simulations of protoplanetary disks, where planets like Earth are forming. With this tool, astronomers can predict the state of protoplanetary disks under various physical conditions on their laptops, one second for each system. Before this tool came out, simulating one system took hundreds of hours on professional computers.

Planets form in protoplanetary disks, a gaseous and dusty environment rotating around young stars. Studying the planets in extrasolar systems helps astronomers to infer the history of the solar system and understand the formation of our homeland, Earth. Although large telescopes can directly observe formed planets around old stars, it is still challenging to image forming planets near young stars. Alternatively, astronomers infer the planet and disk conditions by observing the planets’ perturbations on disks. One crucial step of this indirect approach is to build and evolve disk-planet models accurately so that the models faithfully describe the stories in the observed systems. The challenge of traditional methods for modeling is that they are usually computationally expensive — millions of CPU hours are needed for each system. Shunyuan’s work is an initial exploration of how to reduce the cost. The work was published in the Astrophysical Journal Letters in June 2023.