由于受到外部环境和测量设备误差的影响,在线监测的油色谱数据存在数据含噪和信号振荡的问题,很难直接用于设备状态的分析。针对在线油色谱的预处理问题,提出了一种新型的小波消噪方法。分析了在线油色谱数据的特征,针对性地提出了基于小波系数概率分布的分解层数确定方法和基于离群值留存的阈值确定方法。将这种改进后的小波消噪方法应用于一台有缺陷特高压高抗的在线油色谱数据分析中,结果证明了该方法的可行性和有效性。

Due to the influence of external environment and the error of measuring equipment, on-line oil chromatogram data contains obvious noise and the signal oscillates. The monitoring data is difficult to be directly applied on analysis of the equipment state. In this paper, a novel wavelet based de-noising method is proposed for preprocessing the on-line oil chromatography data. By characteristic analysis of on-line oil chromatographic data, the method of determining the decomposition level based on the probability distribution of wavelet coefficients and the method of determining the threshold value based on outliers conservation are proposed. The improved wavelet de-noising method is applied to analyze the on-line oil chromatographic data of a defective UHV reactor. The results show that the proposed method is feasible and effective.