Study Of Pulsed Eddy Current Nondestructive Testing Technology Based On Finite Element Method

As an important strategic resource of China,oil is related to a country's economic development and social progress.Our country attaches great importance to the leak detection of oil pipelines.Leakage of oil will cause problems such as waste of resources and environmental pollution,and even bring huge losses to people's lives and property.Therefore,the pulsed eddy current testing is of great significance to the research of pipeline defect detection,which is conducive to breaking the monopoly of foreign pipeline inspection technology to China,protecting China's pipeline data from being leaked and maintaining national security.The main work of this paper is as follows:Firstly,this paper introduces the background of the topic and the significance of the topic,as well as a variety of methods for pipeline defect detection.Based on the theory of electromagnetic field,this paper introduces the principle of pulsed eddy current testing,deduces the related formula of pulsed eddy current testing,explaines the causes of eddy current skin effect on the tested parts and establishes the equivalent circuit of pulsed eddy current nondestructive testing model.Secondly,based on the theory of finite element modeling and electromagnetic field,the simulation platform of 2D and 3D pulse eddy current defect detection is built by ANSYS finite element simulation software.The unit type selection,grid partition,boundary condition setting,load exertion and the solution process of the model are introduced in detail.After the solution is completed,the relevant result data of the model can be analyzed and verified with the actual test to prove the correctness of the simulation.Thirdly,according to the model,the coil lift-off effect,the excitation voltage amplitude,frequency,rise time,coil size parameter,the influence of the defect size parameter on the detection result and the variation of the eddy current in the tested part are analyzed.The voltage data successfully detects the defect in the movement detection.At the same time,the magnetic field data above the defect is extracted,and the three-dimensional spatial distribution map of the magnetic field in different directions above the defect is obtained.The relationship between the defect size and the magnetic field distribution is given,which provides data support for the subsequent defect inversion study.Fourthly,based on magnetic field signal,the forward model of ANSYS finite element iterative algorithm is established.Given the initial size parameters of the predicted defects,the correct inversion results are calculated using a neural network algorithm.According to the situation that the objective function becomes larger in the inversion process,an improved algorithm combining neural network algorithm and conjugate gradient algorithm is proposed.When the objective function becomes larger,the conjugate gradient algorithm is used to get the correct inversion model.