基于变施密特数的障碍环境气体扩散数值仿真  

Numerical simulation of gas dispersion in obstacle environments based on variableschmidt number

作　　者：孟庆浩 杨子祯 侯惠让 井涛 MENG Qing-hao;YANG Zi-zhen;JING Tao(School of Electrical and Information Engineering,Tianjin University,Tianjin 300072,China)

机构地区：[1]天津大学电气自动化与信息工程学院,天津300072

出　　处：《安全与环境学报》2022年第2期919-925,共7页Journal of Safety and Environment

基　　金：天津市自然科学基金项目(20JCZDJC00150,20JCYBJC00320)。

摘　　要：通过气体扩散仿真获得气体分布是应对城市环境毒害气体泄漏和扩散问题的重要手段。城市多障碍环境下,风场呈现湍流特点,对气体扩散分布有明显影响。湍流施密特数(Sc_(t))描述了风场湍流特性与气体扩散分布的关系,其取值对湍流环境气体扩散仿真的准确性十分重要。针对多障碍物环境下气体扩散数值仿真问题,提出了一种基于变Sc_(t)的计算流体力学仿真方法:通过调用仿真风场数据自适应计算Sc_(t),进而利用变Sc_(t)进行气体扩散仿真。通过Ansys Fluent计算仿真结果,与开源障碍物阵列风洞试验数据集进行对比,验证了使用本方法获得的仿真气体质量浓度分布与风洞试验数据的一致性,且仿真准确性优于基于常量Sc_(t)等仿真方法,可以应用于复杂障碍物环境下的气体扩散仿真。In an urban environment with multiple obstacles,the wind field exhibits turbulent characteristics,which has a significant impact on gas diffusion and distribution.The turbulent Schmidt number,denoted by Sc_(t),describes the relationship between the turbulence characteristics of the wind field and the gas diffusion distribution,and its value is very important for the accuracy of simulation of gas diffusion in a turbulent environment.To solve the problem of gas diffusion simulation in multi-obstacle environments,a computational fluid dynamics(CFD)simulation method based on the variable Sc_(t)is proposed.By calling the computational results of simulated wind field with the realizableκ-εmodel,including turbulent momentum diffusion coefficient v;,turbulent kinetic energyκ,and turbulence dissipation rateε,the variable Sc_(t)is calculated adaptively by user-defined function(UDF),a C language interface of CFD software Fluent,with which the users can make custom calculation formula.The Sc_(t)varies among different grids and different iterations.Accordingly,the gas diffusion simulation is carried out using the calculated variable Sc_(t).All the computational processes are carried on by Fluent without any external intervention.By comparing with the open-source obstacle-array wind tunnel experimental data set using the statistic index mean squared error(MSE),it is verified that the simulation concentration distribution obtained by the method proposed in this paper is of more consistency with the experimental distribution,and the accuracy of the concentration simulation is better than the simulation method based on the constant Sc_(t).Therefore,the proposed method could be applied to the simulation of gas dispersion in complex obstacle environments.Besides,the distribution of Sc_(t)is plotted by user-defined memory(UDM),which indicates the relationship between obstacles arrangement and Sc_(t)values qualitatively.Obtaining the quantitative relationship will be the future work,combined with machine learning methods.

关 键 词：公共安全 气体扩散分布 障碍环境 湍流施密特数 计算流体力学仿真(CFD) 

分 类 号：X51[环境科学与工程—环境工程]