摘要
为了更准确地分析轴流泵的压力脉动特性,基于结构化网格,在考虑叶顶间隙的情况下,以时均N-S方程为基本控制方程,采用标准k-ε双方程湍流模型和SIMPLEC算法,对轴流泵模型进行了不同的采样频率和采样时间下的全流场非定常数值计算,并分析其对计算结果的影响.结果表明:在进行压力脉动分析时,若取静压作为傅里叶变换的纵坐标,频域图中频率为0时的物理意义即该点平均静压值;采样频率和采样时间对轴流泵压力脉动时域特性和频域特性具有重要影响,指出了因采样频率过小而导致的混叠现象在时域和频域中的表现;并推导出当保证采样时间与叶轮周期和叶片数乘积的比值达到整数倍关系时,基频处不会发生频谱泄漏,并建立了普遍适用轴流泵压力脉动数值计算采样时间的选取公式,进而从公式得出针对不同转速的轴流泵,应选择不同的采样时间.结果为进一步分析轴流泵压力脉动提供了一定参考.
In order to precisely analyze the pressure fluctuation in axial-flow pumps, the unsteady flow in such a pump was calculated based on the time-averaged N -S equations, the standard k -ε turbu- lence model and SIMPLEC algorithm with a structured mesh under a specific tip gap, then the pressure signals monitored were sampled with different sampling frequencies and times to show their effects on the pressure fluctuation characteristic. The results show that the point at f = 0 stands for the average static pressure physically if the pressure was used as the Y-axis of FFT in a frequent domain diagram. Besides, the sampling frequency and time exhibit a significant effect on the pressure fluctuation feature and a too small sampling frequency is responsible for the aliasing phenomenon in time- and frequency-domain. It was derived that a spectral leakage at the basic frequency doesn＇t occur when the sampling frequency is an integer multiple of the product of the time for impeller rotating one revolution and the number of blades. Furthermore, a general equation for detemlining sampling time was established for the pressure fluctuation signals obtained in an axial-flow pump by numerical computation. According to that equation, sampling time is dependent on pump rotational speed. Obviously, such a finding forms the base for pro- cessing of pressure fluctuation signal obtained in CFD simulations in an axial-flow pump.
出处
《排灌机械工程学报》
EI
北大核心
2013年第3期190-194,共5页
Journal of Drainage and Irrigation Machinery Engineering
基金
国家自然科学基金资助项目(51079063
51109093)
江苏高校优势学科建设工程项目
关键词
轴流泵
压力脉动
采样频率
采样时间
时域特性
频域特性
axial-flow pump
pressure fluctuation
sampling frequency
sampling time
time domain characteristic
frequent domain characteristic
