期刊文献+

大温差气固相互作用模型的分子动力学研究

Numerical study of gas-solid interactions with large temperature differences based on molecular dynamics
原文传递
导出
摘要 气固相互作用是许多物理问题的基础,描述气体分子反射速度分布的散射核模型在宏、微观流动研究中都有重要的应用.经典散射核模型大都依赖于待定的调节参数,且仅适用于气固温度相同或相近的情形.本文针对不同温度气体分子在固定温度壁面上的相互作用问题,利用分子动力学(MD)方法直接模拟气体分子的散射过程,获得气体分子反射速度与入射状态的对应关系,进而应用机器学习方法,以Cercignani-Lampis-Lord模型为基础,从直接模拟数据中重建出更精确的、考虑气固温度差影响的修正散射核模型,分析气体分子散射特性及模型参数随气固温度比的变化规律.结果表明,在所讨论温度范围内,与MD模拟结果相比较,修正散射核模型能准确重现反射速度与同方向入射速度的概率密度分布;气固温度差异对散射核的影响体现在特征温度调整为气体温度、调节系数随入射速度及气固温度比变化等两个方面.随着气固温度比变大,气体分子入射速度对散射特征的影响增强,当温度比超过2.5时切向反射速度分布不再受温度影响,而法向速度持续缓慢变化. The scattering of gas at different temperatures on a solid surface with a specified temperature was directly simulated using the molecular dynamics(MD) method. Then, based on the Cercignani-Lampis-Lord(CLL) model, a more accurate modified scattering kernel considering the effect of the gas-solid temperature difference was reconstructed from the MD simulation data using the machine-learning method. The scattering characteristics of gas molecules and the dependency of the parameters in the scattering kernel on the gas-solid temperature ratio were analyzed. The results showed that the probability density of the reflection velocity with the incidence velocity in the same direction predicted by the modified scattering kernel agreed very well with the value counted from MD simulation data. Unlike the traditional kernels in which wall temperature is considered, the characteristic temperature in the modified kernel is the gas temperature. The adjustment coefficients(ACs) depend on both the incident velocity and gas-solid temperature ratio. The analytic expressions for ACs are obtained. The larger the gas-solid temperature ratio, the stronger is the influence of the molecular incidence velocity on the scattering characteristic. When the temperature ratio is greater than 2.5, the probability density of reflection velocity in the tangential direction remains almost constant, but that in the normal direction changes slowly with temperature.
作者 吴晓妍 王子敬 秦丰华 罗喜胜 WU XiaoYan;WANG ZiJing;QIN FengHua;LUO XiSheng(Advanced Propulsion Laboratory,Department of Modern Mechanics,University of Science and Technology of China,Hefei 230026,China)
出处 《中国科学:物理学、力学、天文学》 CSCD 北大核心 2022年第2期76-86,共11页 Scientia Sinica Physica,Mechanica & Astronomica
基金 国家自然科学基金(编号:11972338,11625211,U1730124,11621202)资助项目。
关键词 气固相互作用 散射核 分子动力学 机器学习 gas-solid interaction scattering kernel molecular dynamics machine learning
  • 相关文献

参考文献3

二级参考文献66

共引文献63

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部