A5103072011- Combustion and flame dynamics
- Advanced Combustion Engine Technologies
- Wind and Air Flow Studies
- Fire dynamics and safety research
- Model Reduction and Neural Networks
- Nuclear reactor physics and engineering
- Nuclear Engineering Thermal-Hydraulics
Central South University
2025
Peking University
2024
Within the scope of reacting flow simulations, real-time direct integration (DI) stiff ordinary differential equations for computation chemical kinetics stands as primary demand on computational resources. Meanwhile, number transport that need to be solved increases, cost grows more substantially, particularly those combustion models involving coupling chemistry and such transported probability density function model. In current study, an integrated graphics processing unit-artificial neural...
The application of deep neural networks (DNNs) holds considerable promise as a substitute for the direct integration combustion chemistry in reacting flow simulations. However, challenges persist ensuring high precision and generalization across various fuels conditions, particularly posteriori time-evolving flame This study performs comprehensive learning with multi-fuel computational fluid dynamics (CFD) validations. process begins generating thermochemical base states from low-dimensional...