Noise Elimination and Feature Extraction to Weak Signal of Small Leakage in Pipelines
For solving the difficult problem to distinguish weak signal of small leakage in pipelines, this paper presents a method for extracting weak signal of leakage from signals with much noise based on wavelet entropy. Not only the Signal to Noise Ratio (SNR) can be reduced and the feature is clearer, but also it is not sensitive with the form of signals. The method realizes nonlinear adaptive filtering to weak signals according to the different characteristics between the useful signal and noise. Additionally, using wavelet packet elaborate frequency division we can decompose the signal frequency bandwidth more subtly and establish the feature vectors of pipeline leakage and normal operation states based on frequency segment power, which may be used as input samples of neural network to improve detection accuracy to pipelines leakage. The experiment results showed that using this method the small leakage would be distinguished and located effectively.
pipeline leakage weak signal noise elimination wavelet entropy feature extraction
Zhigang Chen Yidong Xie Meixia Yuan
Dept.of Mechanic Engineering Beijing University of Civil Engineering and Architecture Beijing,China
国际会议
太原
英文
255-258
2011-02-26（万方平台首次上网日期，不代表论文的发表时间）