基于经验模态分解的纸机横向定量测量数据去噪

Noise Removal of Cross-directional Basis Weight Measured Data on Paper Machines Based on EMD

针对传统去噪算法不能满足更高质量横向定量控制要求,提出一种基于经验模态分解(Empirical Mode Decomposition,EMD)的去噪方法。该方法利用EMD分解时间尺度的特性和自适应性,通过去除高频分量并进行滑动平均滤波达到去噪效果。仿真结果表明,这种算法能有效地滤除噪声,同时又保留定量主要细节,有利于进行后续控制。

As conventional noise removal methods were not able to meet the requirement of higher performance cross-directional basis weight control,a new method based on EMD was proposed. The method achieved noise removal by removing the high-frequency component and moving average filter based on time scale features and adaptability of EMD. The simulation results showed that the method had good noise removal effect. The gained data retained the main details of basis weight,which was conducive to subsequent control.