【作者】 梁利平； 徐科军； 朱志海； 王刚；

【Author】 Liang Liping~1 Xu Kejun~1 Zhu Zhihai~1 Wang Gang~2 (1 School of Electrical and Automation Engineering Hefei University of Technology Hefei 230009 China 2 Chongqing Sichuan Automation Co.Ltd.,Chongqing 401121)

【机构】 合肥工业大学电气与自动化工程学院； 重庆川仪自动化股份有限公司；

【摘要】 采用截止频率为F的低通滤波器对方波励磁的电磁流量计输出信号进行处理,只能去除频率高于F的噪声,频率低于F的噪声仍然存在。滤波结果存在高频波动和尖峰。为了去除这些噪声,采用Haar小波对方波励磁的电磁流量计输出信号进行处理,进行多尺度的分解、滤波和重构。用Penalization medium法计算小波滤波中的门限,对分解后各个尺度的高频系数用门限滤波。分别对12.5Hz、25Hz和50Hz单频方波励磁的输出信号和6.25Hz与75Hz的双频方波励磁的输出信号进行了处理。以25Hz单频方波励磁的输出信号为例,进行6尺度的Haar小波分解、滤波和重构。处理结果表明,这种方法能有效地去除各尺度的高频干扰,反映流量的真实情况,满足测量精度的要求。更多还原

【Abstract】 A low-pass filter with cut-off frequency f_c is used to process the output signals of electromagnetic flowmeter with square wave excitation,which only removes noise whose frequency is higher than f_c,but noise whose frequency is lower than f_c is still existed.There are high-frequency fluctuations and peaks in the filtered results.To get rid of such noises,wavelet transform with Haar wavelet bases is used to process the output signals of electromagnetic flowmeter with square wave excitation,and the multi-scale decomposition,filtering and reconstruction is performed.A filtering threshold is determined by means of Penalization medium way.The threshold is employed to filter wavelet coefficients with high frequency components at various decomposed scales. Some signals of electromagnetic flowmeter are processed by the method studied by this paper,including single frequency excitation with 12.5,25 and 50 Hz,and double frequency excitation with 6.25 and 75 Hz.Take signals of 25Hz excited electromagnetic flowmeter for example,after a 6-scale Haar wavelet filtering,the processing results show that this method can remove the disturbance of high frequency components,reflect the actual information of flow-rate and meet the requirement of measurement accuracy.更多还原