Answers are grounded in this paper and cite their sources.
A nonlinear three-dimensional finite element model was developed in order to determine the length, width and depth of several surface-breaking flaws on steel plates using magnetic flux leakage signals. The flaw geometries were reconstructed using the model as part of an inverse problem procedure put together with a commercially available finite element tool and a standard optimization algorithm. A total of 128 experimental signals were used to validate the finite element model. Our results show good agreement between the actual and reconstructed geometries, demonstrating the model strength over variations in flaw length, width, depth and different levels of the excitation field.
No takes yet. Share an insight, caveat, or question.
Schifini et al. (2005) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: