摘要
基于小波变换和BP神经网络理论,构造了一个应用于润滑油水分含量判断的BP神经网络分类器。通过PULSE系统对7种不同浓度润滑油加热后发生微爆效应时产生的声信号进行采集,并进行小波变换,提取能量分布特征信号,最后根据BP神经网络分类器对声信号进行分类并半定量判断润滑油中的水分含量。实验结果表明,该方法能有效地判断润滑油中水分含量是否合格,为研究润滑油水分含量的现场测量提供了新的思路和方法。
A BP neural network classifier for judging the water content in lubricating oil was constructed based on the wavelet transform and the theory of BP neural network. The micro - explosion signal of seven different lubricating oils was acquired by PULSE system and the characteristics of energy distribution was extracted after wavelet transform. Finally, the BP neural network classifier was used to judge the water content in the lubricating oil. The experimental result showed that this method is an efficient method, offering a new notion and method for researching the measurement of water content in the lubricating oil on - line.
出处
《润滑油》
CAS
2008年第2期46-50,共5页
Lubricating Oil
关键词
润滑油
水分
爆裂法
小波变换
神经网络
lubricating oil
water content
crackle test
wavelet transform
neural network
