Research And Evaluation Of Acoustic Emission Technology For Defects Detection Of Tank Bottom

In order to ensure the country's oil supply,my country has built a number of strategic oil reserve bases,and the number of large vertical metal oil storage tanks has surged.As special equipment,accidents such as leakage of oil storage tanks will cause serious consequences.There are many factors that affect the safety of oil storage tanks.Among them,the most concealed are the defects caused by the long-term corrosion of the storage tank bottom plate,and the defects of the storage tank bottom plate are common.exist.Timely inspection and maintenance of storage tanks is of great significance to reducing environmental pollution and economic losses caused by unnecessary maintenance.In this paper,the acoustic emission on-line detection technology is used to study the defects of the storage tank bottom plate,and the work done is as follows:(1)In order to accurately locate the sound source location of the tank bottom,it is first necessary to accurately extract the time difference between the same sound source and the sound emission event signal received by different sensors.This paper proposes an improved quadratic correlation time delay estimation algorithm.Based on the generalized cross-correlation time delay estimation algorithm,this algorithm first combines EEMD and LMS algorithms to reduce the interference of noise to effective signals,improve the signal-to-noise ratio,and then use the weighting function performs a quadratic cross-correlation operation to estimate the time delay.This algorithm effectively improves the problems of poor anti-noise ability of time delay estimation,large amount of calculation,and low accuracy of estimation of nonlinear and nonperiodic signals in acoustic emission source positioning.(2)A triangulation algorithm based on the Newton iteration method and an overdetermined location algorithm based on the sequential quadratic programming method were used to locate the acoustic emission source of the tank floor respectively.The comparison of the positioning performance of the two algorithms under ideal and non-ideal conditions is of great significance for improving the positioning accuracy of the acoustic emission source of the tank floor in actual detection.(3)The DBSCAN clustering method is proposed to identify the area of the acoustic emission event location result.The process of DBSCAN clustering method to identify the density of acoustic emission source tank bottom is described in detail.At the same time,it is proposed to use Silhouette Coefficient,Davies-Bouldin Index and CH index as evaluation indicators to compare the clustering effect of DBSCAN under different parameter configurations,and give parameter configuration suggestions.For the entire tank bottom defect detection process,we independently developed the acoustic emission tank bottom detection data analysis and evaluation software.This article has 58 figures,6 tables and 70 references.