Pipeline Damage Identification Based On Additional Virtual Masses And Acoustic Emission

Pipeline transportation has become the fifth-largest transportation tool in contemporary society,and it has great advantages in transporting fluid media(such as oil,natural gas,water,etc.).China's pipeline projects have spread over large land areas and sea areas and have become an important lifeline for stimulating regional economic ties.However,the existing pipeline engineering is often damaged by the long service time,corrosion aging,external load impact and so on,resulting in serious accidents.In addition,the pipeline is a linear project.The conventional point detection method has the disadvantages of high operating conditions and low detection efficiency.Therefore,it is more and more important to develop fast and effective damage detection methods for complex pipeline projects.This paper proposes different damage identification methods for two different types of pipelines.For the exposed pipeline,a mobile detection method based on additional virtual mass is used.For the buried pipeline,built-in acoustic emission sensor testing technology is used for damage identification.The main research contents are summarized as follows:(1)For the exposed pipeline,based on the virtual deformation method(VDM),this paper deduces the frequency response formula after adding any virtual mass to the pipeline,and verifies the correctness of the formula with a single degree of freedom model.By combining the sensitivity matrix of the theoretical model,the damage iterative formula is deduced using the optimization of the objective function.(2)This paper first applies the method of damage identification based on additional virtual masses to the exposed pipe structure.Taking a 2m-long pipeline as an example,we firstly divide the substructure and apply the excitation at the same location to get the acceleration response,and then calculate the frequency response with the additional optimal mass.Finally,the damage degree of each substructure is iterated through the theoretical mode and the sensitivity matrix.In this paper,firstly,the numerical simulation of the pipeline is carried out by using this method,and then verified by specific tests,which proves the effectiveness of the method for the exposed pipeline damage identification.(3)For the buried pipeline,the sensor layout is inconvenient,this project will try to install the acoustic emission sensor inside the pipeline for the first time to collect the leakage signal.which verifies the effectiveness of the method and provides the basis for the subsequent built-in self capacitive acoustic emission sensors to monitor the damage of the liquid filled pipeline.First,a laboratory model for the leakage of water-filled pipelines was set up,and the four operating conditions for pipeline operation were set.The support vector machine(SVM)method is applied to classify and identify pipeline leakage conditions accurately.Combining SVM and acoustic emission technology to detect pipeline leakage,it provides reference value for realizing real-time online monitoring of pipeline operation status,and has broad application prospects.(4)The topic of using acoustic emission technology to locate the pipeline leakage source is further studied.First,the propagation speed of the acoustic emission wave on the surface of the water-filled pipe was measured by the experiment of lead breaking.Aiming at the problem that the original leakage acoustic emission signal contains too much noise,we use wavelet decomposition and empirical mode decomposition(EMD)methods to de-noise the signal,then reconstruct the signal containing more leakage sources and larger energy.Finally,the delay time is calculated by the cross-correlation function,and the pipeline leakage source is accurately located by the time delay estimation method.