基于车身绕流的低雷诺数湍流模型改进研究
RESEARCH ON IMPROVEMENTS OF LRN TURBULENCE MODEL BASED ON FLOW AROUND AUTOMOBILE BODY
-
摘要: 本文将汽车绕流模块化为各典型局部流动,通过常用湍流模型对各典型局部流动进行数值模拟,结果验证了湍流模型对转捩的捕捉能力是准确模拟汽车绕流的关键. 在分析汽车绕流分离及转捩机理的基础上,优化了稳态和瞬态求解方法,改进了湍流模型对转捩的预测能力,进而提高了湍流模型在汽车流场模拟上的精度. 针对汽车绕流的稳态问题,将流线曲率因子及 响应阈值引入 LRN k-\varepsilon (low Reynolds number k-\varepsilon ) 模型,获得了一种能够更准确预 测转捩的改进低雷诺数湍流模型 (modified LRN k-\varepsilon ),改善了原模型对湍流耗散率的过强依赖性及全应力发展预测不足等问题;针对汽车绕流瞬态求解,通过分析 RANS/LES 混合湍流模型的构造思想及特点,引入约束大涡模拟方法,结合本文提出的改进的 LRN k-\varepsilon 湍流模型,提出了一种能准确捕捉转捩现象 的转捩 LRN CLES 模型. 分别将改进的模型用于某实车外流场和风振噪声仿真中,通过 Ansys Fluent 求解器计算,并将计算结果与常用湍流模型的仿真结果、HD-2 风洞试验结果和实车道路实验结果进行对比,表明改进后的湍流模型能够更准确模拟复杂实车的稳态和瞬态特性,为汽车气动特性的研究提供了可靠理论依据及有效数值解决方法.Abstract: The flow around automobiles was modularized into typical local flows in this paper. Through analyzing the characteristics of typical local flows, it is verified that the capture ability of turbulence model to transition is the key to accurately simulate the flow around automobiles. The paper optimized the solutions of steady-state and transient-state problems by analyzing the separation and transition mechanism of the flow, promoted the prediction ability of turbulence model for transition and improved the accuracy of turbulence model for automobile flow field simulation. For the steady-state solution of the flow around automobiles, by introducing the streamline curvature factor and the response threshold into the low Reynolds number (LRN) k-\varepsilon model proposed by Jones and Lauder, a modified low Reynolds number turbulence model (modified LRN k-\varepsilon ) which can predict transition more accurately was obtained. This model alleviated the problems of the original model's over-dependence on the turbulent dissipation rate and the insufficient prediction of the total stress development. For the transient-state solution, by analyzing the characteristics of the RNAS(Reynolds-averaged Navier-Stokes equations)/LES(large eddy simulation) mixed turbulence model, introducing the constrained large eddy simulation (CLES) method and the modified LRN k-\varepsilon turbulence model proposed in this paper, a transition LRN CLES model that can accurately predict the transition was proposed. These improved models were applied to the simulation of the external flow field and buffeting noise of a real automobile model respectively. Computations were carried out using the ANSYS Fluent solver. The calculation results were compared with the simulation results of the commonly turbulence models, HD-2 wind tunnel test results and real vehicle road test results, it show that the improved turbulence models can more accurately simulate the steady-state and transient-state characteristics of the complex real automobiles, which provides a reliable theoretical basis and effective numerical solution method for the study of automotive aerodynamic.