Data Fusion for MFL Signal Characterization
Author:
- 6 December 2010
- journal article
- Published by Trans Tech Publications, Ltd. in Applied Mechanics and Materials
- Vol. 44-47, 3519-3523
Abstract
The objective of data fusion is to be able to draw inferences that may not be feasible with data from a single sensor alone. In this paper, data from three sets of sensors are fused to estimate the defect profile from magnetic flux leakage (MFL) inspection data. The three sensors measure the axial, radial and tangential components of the MFL field. Data is fused at the feature level. Examples of signal features are amplitude, width, etc. A radial basis function network (RBFN) is then employed to map the fused features appropriately to obtain the geometric profile of the defect. The feasibility of the approach is evaluated using the data obtained from the MFL inspection of oil pipes. The results obtained by fusing the axial, radial and tangential components appear to be better than those obtained using the axial and radial component alone.Keywords
This publication has 6 references indexed in Scilit:
- MFL signals and artificial neural networks applied to detection and classification of pipe weld defectsNDT & E International, 2006
- Data dimensionality reduction with application to simplifying rbf network structure and improving classification performanceIEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2003
- Effects of alignment of nearby corrosion pits on MFLNDT & E International, 2003
- Advanced signal processing of magnetic flux leakage data obtained from seamless gas pipelineNDT & E International, 2002
- Correlation between magnetic flux leakage and magnetic Barkhausen noise: Stress dependence in pipeline steelJournal of Magnetism and Magnetic Materials, 1997
- A Resource-Allocating Network for Function InterpolationNeural Computation, 1991