大涡模拟中大气边界层紊流风场的生成方法研究

发布时间：2018-12-12 16:14

【摘要】：大涡模拟是计算风工程中的重要研究方法和手段，能够比较准确的模拟脉动风。在大涡模拟中脉动入口边界条件尤为重要，生成脉动风入口的方法有很多，其中随机序列合成法应用广泛。然而随机序列合成法中的DSRFG方法由于脉动成分的加入导致平均风剖面变化，模拟的风场达不到预想效果。大涡模拟中网格对计算域中风场存在过滤，在入口处生成的脉动速度存在的最高频率应该与计算域中过滤的截断频率相统一，减少计算域中不可分辨频率对计算的影响。随机序列合成法中的DSRFG方法还存在较为明显的压力梯度问题。本文有以下主要研究内容：基于von-Karman谱推导出大涡模拟中网格尺寸与截断频率之间的关系，并用具体公式描述其关系，保证程序中脉动成分的时间过滤和计算域中的空间过滤相统一，减少高频成分对计算结果的影响。计算多个FLUENT模型，，分析每个计算模型870个检测点处的脉动风速。使用曲线拟合和曲线拟合优度检验理论找出最能够代表计算域内平均风速的风剖面，分析DSRFG方法程序中给定平均风剖面与计算域中平均风剖面的差异，通过修改原有DSRFG程序使计算域中平均风剖面满足规范要求。在MATLAB中分析每一个CASE870个检测点处的脉动风速，对比RFG方法和DSRFG方法脉动风功率在计算域中的变化，以及湍流强度在计算域中的变化情况，并分析过滤高频成分对DSRFG方法脉动特性的影响。将DSRFG方法的原有程序改编成MATLAB可识别的语言，并通过MATLAB生成脉动速度，通过MATLAB过滤掉脉动速度的高频成分，将处理后的速度重新加入FLUENT中，用以减弱DSRFG方法存在的严重压力梯度问题。
[Abstract]:Large eddy simulation is an important research method in wind engineering, and it can simulate fluctuating wind accurately. The boundary condition of pulsating inlet is very important in large eddy simulation. There are many methods to generate pulsating wind inlet, among which the random sequence synthesis method is widely used. However, the DSRFG method in the stochastic sequence synthesis method can not achieve the desired results because of the variation of the mean wind profile caused by the addition of pulsating components. In large eddy simulation, the wind field is filtered by grid, and the maximum frequency of pulsating velocity generated at the entrance should be consistent with the truncation frequency of filtering in the computational domain, so as to reduce the influence of the indiscernible frequency in the computational domain on the calculation. The DSRFG method in the random sequence synthesis method also has obvious pressure gradient problem. The main contents of this paper are as follows: based on the von-Karman spectrum, the relationship between mesh size and truncation frequency in large eddy simulation is derived, and the relation is described with specific formulas. In order to reduce the influence of high frequency components on the calculation results, the temporal filtering of pulsating components and the spatial filtering in the computational domain are unified in the program. Several FLUENT models are calculated, and the pulsating wind speed at 870 detection points of each model is analyzed. The theory of curve fitting and curve fitting goodness test is used to find out the wind profile which is the most representative of the average wind speed in the computational domain. The difference between the given average wind profile and the average wind profile in the computational domain is analyzed in the DSRFG program. By modifying the original DSRFG program, the average wind profile in the computational domain can meet the requirements of the specification. The fluctuating wind speed at each CASE870 detection point is analyzed in MATLAB. The variation of fluctuating wind power in the computational domain and the turbulence intensity in the computational domain are compared between the RFG method and the DSRFG method. The influence of filter high frequency components on the pulsation characteristics of DSRFG method is analyzed. The original program of the DSRFG method is adapted into a MATLAB recognizable language, and the pulsation velocity is generated by MATLAB, and the high frequency component of the pulsation velocity is filtered out by MATLAB, and the processed velocity is added to FLUENT again. It is used to reduce the serious pressure gradient problem in DSRFG method.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TU312.1

【相似文献】