Abstract: Negative pressure wave signal denoising is one of the core issues to be solved in pipeline leakage detection method based on negative pressure wave. A comparison among traditional hard and soft threshold function, hyperbolic and exponential algorithm and improved algorithm has been made on the basis of analyzing wavelet transform theory and wavelet threshold denoising. The results of experiment show that improved algorithm used in pipeline leakage detection processing could obtain a larger SNR (signal to noise ratio) and a good denoising ability for negative pressure wave signal, which could effectively improve leakage detection accuracy.

Key words: Pipeline leakage, Wavelet transform, Threshold de-noising, Improved algorithm