Study On The Application Of Oil Pipeline Leakage Detection Based On Wavelet Transform

Pipeline transportation system is playing a more and more important role in our country's economy. It is very significant to find and locate the leakage point in time to reduce the losses and to keep the system working normally. The method of the detection and location of pipeline leakage are studied with wavelet analysis theories in this dissertation. The accomplished research work can be described as the following aspects:Firstly, the denoising method is studied with wavelet analysis. Threshold denoising based on the cross-correlations between wavelet's coefficients is proposed to denoise more effectively. The shortcoming of the rigid choice in threshold denoising is overcome with the proposed method. Secondly, according to characteristics of the leakage signal, the location arithmetic of singularity point is improved based on wavelet analysis and a new method, based on the mother wavelet and signal's cross-correlations, is proposed to choose the mother wavelet in the detection of singularity point. Thirdly, in order to overcome the shortcomings of the general location method, the velocity of the pipeline negative pressure wave is calculated accurately considering the effect of the medium temperature. The amended half-span-search method is put forward to extend the application spectrum. The process results of the actual leakage data suggest that the proposed location method could improve the location precision effectively. Lastly, a pipeline leakage monitor and detection system is primarily designed with VB and Matlab. The functions of the system include data acquisition, data communication, leakage judging, leakage alarm, leakage location and so on.