Evolution Mechanism and High-Precision Quantitative Identification of MFL Signals from Defects Under Supersaturated...

Magnetic flux leakage (MFL) testing is a critical non-destructive testing (NDT) method for ensuring the safety of ferromagnetic storage and transportation equipment. However, existing research has predominantly focused on weak or saturated magnetization states, leaving the characteristic laws and physical mechanisms of defect signals under supersaturated magnetization conditions unclear. To address this gap, this paper systematically investigates the MFL signal evolution mechanism and develops a high-precision quantitative identification method for defects under supersaturated magnetization conditions through finite element simulation, theoretical modeling, and experimental validation. First, a three-dimensional (3D) finite element model for MFL testing is established using COMSOL Multiphysics. The regulatory effects of key parameters-sensor lift-off value, defect burial depth, length, and depth-on the peak values and distribution characteristics of axial and radial MFL signals are revealed, a signal peak characterization model for each parameter and their adjusted R2 is obtained via fitting, and the detection capability of the detector for defects with different shapes is simultaneously verified. Furthermore, actual detection is conducted on three crack defects of different sizes, and the analysis results indicate that the characterization models of each parameter obtained from the simulation exhibit high accuracy. The results show that MFL signal intensity under supersaturated magnetization conditions is significantly enhanced compared to that under saturated magnetization conditions. Furthermore, to improve defect length measurement accuracy, a signal correction method based on the midpoint of extreme values of the second derivative of axial signals is proposed. By compensating for peak offsets caused by factors like magnetic field diffusion, this method reduces the maximum defect length identification error from 14.25% (pre-correction) to below 0.3%. This study elucidates the coupling influence mechanism of multi-physical parameters on MFL signals under supersaturated magnetization conditions. The proposed high-precision signal correction method provides a novel theoretical basis and technical approach for the accurate quantification and inversion of defects in complex operating conditions.