Scientists are exploring renewable energy (RE), but energy storage (ES) solutions have become increasingly untenable. There has been some ES research on advanced functional materials, focusing on nanoscale material design, but traditional scientific experimental methods take a lot of time and money to develop these new materials.
Therefore, we designed a deep-learning system to identify the most promising materials in the multiple disciplines. On one hand, our research-based system has demonstrated remarkable advantages in automated scientific discovery, including about a 90% reduction in time required for literature reviews and a 70% rank correlation with experimental ranking, the latter of which is more than twice as effective as traditional methods. On the other hand, our experiment-based system powers multi-dimensional parameter optimisation in nanomaterial fabrication in 75% less time.
Through this project, we plan to develop our AI solutions to a higher level for practical applications. We will develop a platform on which users can customise AI solutions for material design or use our developed system for insights into material design and experiment parameters.
Team members
Mr Xie Tong (University of New South Wales)
Mr Wang Shaozhou (University of New South Wales)
Mr Linghu Qingyuan (University of New South Wales)
Miss Peng Jiejun (Johns Hopkins University)
* Person-in-charge
(Info based on the team's application form)
- CityU HK Tech 300 Seed Fund (2022)
- Innovation and Entrepreneurship Growth Award, The “Chunhui Cup” Oversea Students Innovation and Entrepreneurship Competition (2022)
- Top 12 project, Innovation Australia 2021 (2021)