Font Size: a A A

Research On The Defect Detecting Methods Of Ferromagnetic Pipes Based On Electromagnetic Eddy Current

Posted on:2019-05-15Degree:DoctorType:Dissertation
Country:ChinaCandidate:Q W LuoFull Text:PDF
GTID:1311330569487553Subject:Instrument Science and Technology
Abstract/Summary:PDF Full Text Request
Pipes are coupled by joints to maintain the lifeblood of transporting oil and gas.Their testing and maintaining have important economic and strategic significance.Non-destructive Testing(NDT)is the main technology of detecting pipes.In the NDT methods of pipes,Eddy Current Testing(ECT)is widely used because of its contactless,easy opeartions and so on.As one of the ECT methods,Remote Field Eddy Current Testing(RFECT)is becoming an advanced research hotspot of pipe testing because of its same detecting sensitivity to inner and outer wall defects.At present,there still have some problems in pipe defects detection based on RFECT,espectially in the filed of high accuracy and high reliability.Based on electromagnetic eddy current testing,this dissertation researches the defect detecting methods of ferromagnetic pipes.The dissertation combines theoritical studies,Finite Element Analysis(FEA)and physical experiments to process the practical problems of applying RFECT on pipes testing.The research of the dissertation focuses on applying differential receiving coils to remove the secondary peaks caused by the transmitter passing by defects,on applying the Wiener filter to deconvolution and to filter the practical testing noise,and on applying Multi-frequency Eddy Current Testing(MFECT)to implement the inversion of pipe physical parameters(inner diameter,conductivity and permeability).Further,the dissertation applies Eddy Current Array Testing(ECAT)to distinguish defects locations(axial and circumferential locations,inner or outer wall locations),and investigates the Pusled Remote Field Eddy Current Tesing(PRFECT)of ferromagnetic pipes.The research results in the dissertation can be used to improve the feasibility and practicality of RFECT of pipes.The works and innovations of the dissertation are as follows:(1)A method for removing the secondary peaks appeared in Remote Field Eddy Current Tesing(RFECT)signals of pipes is proposed.In the RFECT of pipes,the signal induced by transmitter penetrates the pipe wall near the transmitter and propagates along the outer pipe wall,and re-penetrates the pipe wall into pipe and is received by the receiving coils.The received signal is sensitive to both inner and outer wall defects because of its two penetrations.The twice penetration of received signal has some disadvantages,such as that it causes secondary peaks.The secondary peaks result from the transmitter passing by the defects,and they will lead to wrong assessment of defects.In order to remove the secondary peaks from testing signal,a differential testing model is applied in RFECT of pipes.By analyzing the time shift between the differential testing signals,the signals are transformed to the ones in frequency domain and the Wiener filter is used for deconvolution.Moreover,the Wiener filter is taken advantages into filtering practical testing noise.The proposed method for removing secondary peaks is validated by ANSYS simulation and experiments.Researches of this part can be applied in improving the applicability of RFECT of pipes testing to a certain extent.(2)The signal processing methods of Eddy Current Array Testing(ECAT)of pipes are researched and verified by experiments and simulations.The ECAT is employed to improve the localization and quantification accuracy of defects.The locations of defects on pipe wall can be classified into three types: axial location,circumferential location and inner or outer wall location.The localization of defects on axial and circumferential location can be achieved by setting circumferential array sensors,and the accuracy of quantifying defects can be improved by applying the proposed secondary peaks removing methods.Moreover,a high frequency is transmitted to differentiate the inner and outer wall defects based on its skin effect.Under the consideration of joint pipes in pracrice,the directional differentiation method is used to process the high frequency signals.The methods studied in this part are verified by experiments and simulations.(3)The methods for inversing the physical parameters(inner diameter,conductivity and permeability)based on Multi-frequency Eddy Current Testing(MFECT)are studied,and a new method is proposed to numerically analyzing the transfer impedance model.The purpose of nondestructive testing pipes is to achieve the localization and quantification of defects.According to the theory of RFECT,the variations of testing signals are not only caused by the defects,but also caused by the physical parameters of pipes(inner diameter,conductivity and permeability).Therefore,the physical parameters of pipes should be compensated for the aim of defects analysis,and the MFECT is used to achieve this goal.The multi-frequency signals are excited by a transmitter and received by a coil,and the inversion model is built by analyzing the transfer impedance between the exciting signals and the receiving signals.Through ANSYS simulation and numerically analyzing the transfer impedance,the inversion library is built.The Gradient Descent Algorithm(GDA)and Least Squares Support Vector Regression(LSSVR)machine are separately used to inverse the physical parameters and the results are compared.It indicates that the LSSVR has better approximation accuracy and generalization ability than the GDA.These analyses are validated by experiments.(4)The features and processing methods of Pulsed Remote Field Eddy Current Testing(PRFECT)signals are studied.PRFECT is different to RFECT because of its pulsed excitation,and the testing signals in PRFECT have broad frequency spectrum.In the PRFECT of pipes,the receiving coils are set in the transition zone to make the testing signals contain both near field and remote field information.The LSSVR is employed to extract the feature signals(such as the zero-crossing time and negative amplitude)from testing results for pipe feature analysis.As a conclusion,the secondary peaks are also found in the feature signals when the transmitter passing by the defects,and the secondary peaks in PRFECT also can be eliminated by employing the proposed methods.Researches in this part can improve the practicality of PRFECT of pipes.
Keywords/Search Tags:pipe, defect detection, electromagnetic eddy current, array testing, multi-frequency signal, pulsed signal
PDF Full Text Request
Related items