一种改进的均匀各向同性湍流初始化方法
发布时间:2018-05-31 01:33
本文选题:均匀各向同性湍流 + Rogallo初始化方法 ; 参考:《力学学报》2016年06期
【摘要】:均匀各向同性湍流是一种最简单的湍流理想状态,也是湍流基础理论研究的最重要对象之一.为了用数值方法产生均匀各向同性湍流场,一般采用Rogallo提出的方法在谱空间生成初始场,然后再转换到物理空间.研究表明,由该方法生成的初始湍流场在3个棱向上呈各向异性,在结构函数和速度概率密度分布上均有体现.尽管在初始场样本很多时,这种各向异性可以在平均意义上消除,但作为数值模拟采用的单个流场则波动较大,不利于在实际计算中作为单个初始场生成各向同性湍流.在此基础上提出一种改进的Rogallo初始化方法,称为模量平均法,将Rogallo方法在3个轴向分别进行,并进行模量平均,最后采用能谱进行模量控制.这种方法可以一方面保持初始场能谱,另一方面减小单个流场的各向异性波动,以产生各向同性程度更佳的单个初始场.在统计意义上,新方法可以分别将结构函数和速度概率密度的相对标准差减小约10%.
[Abstract]:Homogeneous isotropic turbulence is one of the simplest ideal turbulent states and one of the most important objects in the basic theory of turbulence. In order to generate homogeneous isotropic turbulent field by numerical method, the method proposed by Rogallo is generally used to generate the initial field in spectral space and then to be converted to physical space. The results show that the initial turbulence field generated by this method is anisotropic up to three edges and can be found in both the structure function and the velocity probability density distribution. Although this anisotropy can be eliminated in the average sense when there are a large number of initial field samples, the single flow field, which is used for numerical simulation, fluctuates greatly, which is not conducive to the generation of isotropic turbulence as a single initial field in the actual calculation. On this basis, an improved Rogallo initialization method, called modulus averaging method, is proposed. The Rogallo method is carried out separately in three axes, and the modulus is averaged. Finally, the modulus is controlled by energy spectrum. This method can preserve the energy spectrum of the initial field on the one hand and reduce the anisotropic fluctuation of the single flow field on the other to produce a single initial field with better isotropy. In the statistical sense, the new method can reduce the relative standard deviation of the structure function and the velocity probability density by about 10 parts respectively.
【作者单位】: 北京航空航天大学中法工程师学院;法国里昂中央理工大学流体与声学实验室;
【基金】:国家自然科学基金资助项目(11202013,11302012,51420105008)
【分类号】:O357.5
【相似文献】
相关期刊论文 前2条
1 刘鲭洁;陈桂明;刘小方;占君;;BP神经网络权重和阈值初始化方法研究[J];西南师范大学学报(自然科学版);2010年06期
2 ;[J];;年期
相关会议论文 前1条
1 朱明华;刘元安;吴惠兰;;一种用于CMA阵列的初始化方法[A];开创新世纪的通信技术——第七届全国青年通信学术会议论文集[C];2001年
,本文编号:1957754
本文链接:https://www.wllwen.com/kejilunwen/lxlw/1957754.html