Inverse problem in nondestructive testing using arrayed eddy current sensors

Area of Science:

• Materials Science and Engineering
• Nondestructive Testing
• Computational Electromagnetics

Background:

• Nondestructive testing (NDT) is crucial for assessing material integrity without causing damage.
• Eddy current testing (ECT) is a widely used NDT method for detecting surface and near-surface flaws.
• Accurate crack profile reconstruction is essential for reliable material assessment.

Purpose of the Study:

• To develop a fast and accurate model for reconstituting crack profiles using arrayed eddy current sensors.
• To implement an efficient inverse problem-solving approach for crack characterization.
• To enhance the speed and reliability of eddy current-based nondestructive testing.

Main Methods:

• Development of a direct problem model using 2D axisymmetric finite element analysis and superposition.
• Application of the reciprocity principle with dyadic Green's functions to calculate impedance variations.
• Utilizing genetic algorithms to iteratively solve the inverse problem by subdividing the crack surface into cells.

Main Results:

• A fast crack profile reconstitution model was successfully developed.
• The model accurately determines crack depths based on impedance variations.
• The computational efficiency was significantly improved by evaluating the dyadic Green's function matrix independently.

Conclusions:

• The developed model offers a fast and efficient method for crack profile reconstitution in NDT.
• Arrayed eddy current sensors combined with genetic algorithms provide a powerful tool for material defect analysis.
• This approach enhances the capabilities of eddy current testing for critical structural health monitoring applications.