Research On Liquid Pipeline Leakage Signal Processing Based On The Wavelet Transform Combined With Blind Source Separation

With the gradually formation of our national oil and gas pipeline network,safe operation of pipeline in service has gradually become the focus of attention for the oil and gas industry.Abnormal pressure fluctuations appears on the interface of SCADA system usually characterizes condition changes in the pipeline transportation.If these changes fails to be judged timely or effectively,adverse effect may be brought on the production plan of the downstream refineries and oil depots,even the pipeline accident emergency.From the point of view of solving the practical problems of industry,the liquid pipeline negative pressure wave signal processing and leak detection are treated as the research object based on the theory of liquid pipeline transportation.In this paper,the knowledge of crossdisciplinary such as signal processing and machine learning is integrated,and the leakage condition of liquid pipelines is discussed through experiments and software simulations.First of all,numerical calculation of liquid pipeline leakage process is carried out based on the theory of transient flow,which provides a theoretical basis for the application of negative pressure wave method to pipeline leak detection and also provides a reference for the sensor selection required by the experiment.And then,the application of wavelet transform in the negative pressure wave signal preprocessing is studied,which can be used as a tool to extract the signal characteristics of pipeline leakage.Using the fusion algorithm to process the collected pressure signals and analyzing the characteristics and causes of the pressure fluctuations in different working conditions of the pipeline.Based on the pipeline physical model,an approximate mathematical model of pipeline attenuation along negative pressure is established.Using this model,the method of determining the minimum detectable leak rate and the sensitive point of the negative pressure wave method is proposed.The research shows that the magnitude of the leakage in the pipe affects the negative pressure wave amplitude,while the overall waveform has not changed.Meanwhile,the magnitude of the negative pressure wave amplitude is related to the amount of leakage.Wavelet de-noising and singularity capture are applied to the signals of abnormal pressure fluctuations,which can be used to extract the features of negative pressure wave signals in various working conditions.Loop experiments were used to simulate actual pipeline leak conditions and the actual results of wavelet denoising were verified.According to its deficiency,a new threshold function is established to improve the reconstruction accuracy of wavelet denoising.For the first time,the feasibility and the superiority of the new threshold function are analyzed from the perspective of mathematics.On the basis of improving the wavelet threshold function,the wavelet transform is fused with the blind source separation method based on maximum SNR.The applicability of this method is illustrated by the separation of known construction signals.Based on the pipeline physical model,an approximate mathematical model of exponential decay of the negative pressure wave along the pipeline is established by analyzing the dynamic process of the pipeline.Using this model,the method of determining the minimum detectable leak rate and the sensitive point of the negative pressure wave method is proposed,and the influence of each factor on the minimum detectable leak rate and the sensitive point is analyzed.