Research on Magnetic Flux Leakage Signals Quantity Technology of Tank Floor Corrosion Defects Based on Artificial Neural Network

Author(s):  
Zhijun Yang ◽  
Guang Dai ◽  
Wei Li ◽  
Yanbiao Jiang
Author(s):  
Lynann Clapham ◽  
Vijay Babbar

The current study was designed to model the dynamic effects of detector ride and magnet liftoff on Magnetic Flux Leakage (MFL) signals from dents as well as gouges that have significant denting. The MFL tools have long been used for the detection and sizing of corrosion defects. This is comparatively straightforward for a number of reasons, one of which is that the MFL detector assembly can ride relatively smoothly along the inner pipe wall surface. This is not the case when significant denting is present, since the dent presents a perturbation in the pipe wall that can cause liftoff of the detector or magnet system. Since the tool travels at relatively high speeds down the pipe, the dent itself can cause the detector to lose contact with the trailing half of the dent. In addition, the magnet pole piece may experience partial liftoff as it traverses the dent, thus causing a change in the local flux density. In this study results from ‘static’ measurements are compared with a dynamic case in which detector liftoff is simulated through modeling and experiment. Results are discussed regarding the severity of MFL signal loss at the trailing edge of the defect as a result of detector liftoff. The effect of partial liftoff of the magnet as it passes over the dent is also examined. Magnet liftoff is found to increase the local magnetic flux near the liftoff region, causing the MFL signal from the dent wall to increase rather than decrease in the vicinity of magnet liftoff region.


Sign in / Sign up

Export Citation Format

Share Document