摘要
用统一二阶矩模型（ＵＳＭ）模拟了旋流数为０４７和１５的气粒两相流动，并和实验结果以及ｋεｋｐ模型的模拟结果进行了对比．研究结果表明，提高旋流数减小了轴向速度反流区，增大了切向速度似固核区．ＵＳＭ和ｋεｋｐ模型预报旋流数为０４７时的两相速度场差别不大，并都和实验结果接近，但前者预报的旋流数为１５的两相速度场比后者有改进，在两种情况下，前者都能揭示出后者无法预报的两相湍流各向异性规律．
Numerical simulation of swirling gas particle flows are carried out for the development of innovative cyclone separators, hydrocyclones, cyclone furnaces and swirl burners in our laboratory, using two fluid models. The key problem in two fluid models is the two phase turbulence model. The two phase turbulence in swirling flows is obviously anisotropic. The isotropic k ε k p model proposed by us can give reasonable two phase velocity fields in case of swirl number s =0 47, but cannot predict the anisotropy of two phase turbulence. Therefore, for strongly swirling gas particle flows anisotopic two phase turbulence models should be used. One of such models is the “unified second order moment (USM) model” i. e. the two phase Reynolds stress transport equation model, proposed by the present authors several years ago. In order to clarify the effect of swirl number on the two phase flow behavior, in this paper the USM model is used to predict both weakly swirling gas particle flows with s =0 47 and strongly swirling gas particle flows with s =1 5. Basically, the USM model consists of gas phase and particle phase Reynolds stress equations together with transport equations of other relevant correlations, such as the two phase velocity fluctuation correlation, using the idea of second order moment closure adopted in single phase flows. The numerical procedure is an extension of SIMPLEC algorithm Eulerian Eulerian predictions for two phase flows. The boundary conditions for the gas phase are the same as in usual simulations, and the boundary conditions for the particle phase are: given inlet conditions by experiments; zero normal velocity, zero gradients of axial and tangential velocities, zero particle mass fluxes, zero gradients of Reynolds stresses at the wall. The numerical results are compared with those using the k ε k p model and PDPA measurements. They indicate that increasing swirl number leads to reduction of the central reverse flow zone and increase of the size of solid body rotation zone. The comparison shows that both USM and k ε k p models can predict well the two phase velocities of weakly swirling flows with s =0 47, but the USM model can predict better the strongly swirling flows with s =1 5, and the USM model can properly reveal the features of anisotropic two phase turbulence for both two cases, while the k ε k p model fails to do so.
出处
《力学学报》
EI
CSCD
北大核心
1998年第4期385-390,共6页
Chinese Journal of Theoretical and Applied Mechanics
基金
国家自然科学基金
高校博士点基金
清华大学煤燃烧国家重点实验室开放课题
关键词
旋流
两相流动
湍流模型
统一二阶矩模型
swirling flows, gas particle flows, turbulence model[ST〗[WT〗[HJ〗[LM〗
