Leakage Detection Method Of Product Oil Pipeline Based On Infrasound

With the rapid development of network information technology,the advent of the era of Industry 4.0,the rapid changes in technology have driven the rapid transformation of pipeline leak detection technology in the direction of intelligent detection.As one of the main modes of transportation,pipelines have begun to suffer from various problems due to the increasing number of years in service,thus causing leaks to occur.The leakage of refined oil pipelines can lead to serious waste of national resources,loss of property,and even ecological damage due to environmental pollution if they are not precisely located and plugged in time.Although a variety of technologies have emerged to solve these problems,there has not been a unified solution due to complexity and cost.This thesis will discuss this for refined oil pipelines.First of all,infrasound pipeline leak detection methods are proposed,existing detection methods are collated and their respective advantages and disadvantages are analysed,and then the infrasound generation mechanism,propagation mechanism,interfering noise factors and localisation formulae are studied.Secondly,the infrasound noise reduction wavelet transform method will be analysed and the five wavelet bases will be compared so that a suitable mother wavelet can be selected for the infrasound signal noise reduction process.Again,the principles of the variational modal decomposition of infrasound are analysed,a suitable preset scale K is selected,and simulation experiments are carried out for the noise reduction of infrasound signals.On this basis,wavelet threshold noise reduction is fused with variable mode decomposition noise reduction,and a new noise reduction algorithm is proposed: Wavelet transform-variable mode decomposition.This algorithm first uses wavelet threshold decomposition to reduce noise,and then uses variable mode decomposition to perform secondary noise reduction on the noise reduced data,so as to accurately locate leaks.The experimental study verifies that the proposed Wavelet-VMD noise reduction algorithm has a very obvious noise reduction effect and can improve the positioning accuracy to a large extent.The study shows that the relative error can reach about 1.25% in the pressure-holding state and about 1.6% in the open pump state.The research method of infrasound-based product oil pipeline leak detection studied in this thesis can be very good for acoustic signal noise reduction and accurate positioning,which confirms the feasibility of this algorithm.