Using artificial intelligence algorithms, quantum chemical theory and molecular dynamics, Zhu Tong, an associate researcher with the ECNU School of Chemical and Molecular Engineering, has completed high-precision computer simulations of fuel combustion with his students. Their research, “Complex reaction processes in combustion unraveled by neural network-based molecular dynamics simulation, was published in Nature Communications.
ECNU is the first author affiliation for the paper. Zeng Jinzhe, an ECNU undergraduate who graduated in 2019, is the first author. ZhuTong and Prof. Zhang Zenghui are the corresponding authors.
The corresponding authors, ZhuTong and Prof. Zhang Zenghui
Combustion occurs at extreme physical conditions, with high pressures and high temperatures up to several thousand degrees. Many of the elementary reactions during the process of combustion typically occur on sub-picosecond time scale. These extreme physical conditions make it very difficult, if not impossible, to carry out real-time experimental studies of combustion.
For instance, traditional computational simulations cannot accurately and efficiently make quantum chemical calculation of a large sample of reaction paths as a result of violent chemical reactions in the combustion process.
More researchers have decided to seek the help of machine-learning (ML) methods, especially artificial neural networks, thereby providing the possibility to construct simulated algorithms with quantum chemical precision and high-level efficiency. This research provides a database construction scheme designed specifically for the combustion reaction, adopting an artificial neural network model to simulate the molecular dynamics (MD) of methane combustion lasting up to 1 nanosecond with 0.1 sub picosecond resolution.
How the artificial neural network model and the major substances change in methane combustion reaction
The first author, Zeng Jinzhe
This research was supported by the Basic Research on Turbulent Combustion for Engine under the National Natural Science Foundation of China, Key R&D Program of the Ministry of Science & Technology, National Innovation and Entrepreneurship Training Program for Undergraduates, and ECNU Supercomputer Center.
Source: School of Chemistry and Molecular Engineering
Photos by Li Zhendong
Copy editor: Joshua Mayfield
Editor: Yuan Yiwei
