Research On Modeling And Inversion Of Defect Magnetic Flus Detection For Ferromagnetic Materials

There are many methods of non-destructive testing of defects in ferromagnetic materials.In comparison,magnetic flux leakage detection is widely applied because of its high sensitivity,fast speed,simple operation process,low cost,and it also can detect surface and internal defects.The purpose of this paper is to reflect the shape contour and dimension parameters of defects in the workpiece under the defect leakage magnetic field information,which is the research of the inversion problem.The forward problem is the basis of the inverse problem.Therefore,to establish a fast calculation model of the defective magnetic flux leakage detection problem,is important for the reconstruction of the defect contour and the identification of the parameters.Firstly,based on the theory of magnetic flux leakage testing,this paper established a model of defect magnetic leakage field and built a three-dimensional and two-dimensional simulation model by finite element analysis software COMSOL.The characteristics of the leakage magnetic field were analyzed,and the influence of the size parameter of the defect and the detection of the lift-off height on the leakage magnetic field was simulated.The results validated the correctness of the model and provided the basis for the identification of the defect parameters,and provided guidance for the design of the magnetizing device.Secondly,combined with the advantages and disadvantages of the magnetic dipole model and the finite element numerical calculation model,this paper presented a unitary integral calculation method based on the magnetic dipole model of the material defect magnetic flux leakage detection positive solution problem,established the forward magnetic flux leakage testing model.And the algorithm program was written by using LiveLink for MATLAB interface in COMSOL,making the algorithm to be simplified and flexible.Compared the calculation results with the simulation results,the correctness of the forward model proposed in this paper was verified,which provided a way to solve the inversion problem.Finally,based on the forward model,the solution of the inversion problem was analyzed from two perspectives.One was to calculate the magnetization characteristics of the defect slab through the singular value decomposition algorithm and got the shape contour image of the defect.The second was to calculate the shape of the defect by the conjugate gradient algorithm,and directly solved the size parameters of defects,which achieved the identification of parameters.Then,taking the rectangular defects,the rectangular combination defects and the triangular defects as an example,the inversion results of two solving methods were analyzed,and the validity of the algorithm was verified.