大型泵站水泵机组振动信号滤波方法

针对大型泵站水泵机组运行时振动信号易受多种因素影响产生随机误差、噪声干扰,进而影响机组健康状态评价结果的问题,对水泵机组振动信号滤波方法进行分析研究。以南水北调东线某泵站水泵机组运行时机泵联轴器处的振动信号为例,分别使用巴特沃斯滤波器、基于汉宁窗的FIR 滤波器以及基于谐波小波的滤波器3 种滤波方法对水泵机组振动信号进行滤波处理并进行对比分析。结果表明:巴特沃斯滤波器和基于汉宁窗的FIR 滤波结果中存在时域波形的时延和轴心轨迹的连线,滤波不完全,而基于谐波小波的滤波几乎能够完全滤除干扰;对于特定频段的滤波,基于谐波小波的滤波在频段范围比较小的时候优势更加明显。

Vibration signal filtering method of pump unit in a large pumping station

Aiming at the problem that the vibration signal of pump unit is susceptible to random error and noise interference caused by various factors, which affects the evaluation result of the health status of pump unit, the filtering methods of vibration signal of pump unit were analyzed and studied. Taking the vibration signal at the pump coupling of a pumping station from the East Route of the South-to-North Water Transfer Project as an example, filtering methods based on Batworth, Hanning window, and harmonic wavelet were used to filter the vibration signal of the pump unit. The comparative analysis results show that there are time delay and line of axis trajectory in the FIR results of filtering methods based on Batworth and Hanning window, indicating incomplete filtering, while the harmonic wavelet filtering can almost completely filter out the interference. The advantages of harmonic wavelet filtering are more obvious when the frequency range is smaller for a certain frequency band.