Physics-constrained machine learning for thermal turbulence modelling at low Prandtl numbers
International Journal of Heat and Mass Transfer(2022)
摘要
•A data-driven algebraic model for the turbulent heat flux is developed.•The ANN is trained with DNS data of near unity and low Prandtl number flows.•Invariance and realizability properties are embedded into the model structure.•The model adapts satisfactorily to different Prandtl numbers and flow conditions.•The model is highly sensitive to the accuracy of the momentum modelling.
更多查看译文
关键词
Low-Prandtl,Turbulence,Heat flux,Artificial neural networks
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要