Eddy Current Classification of Sheet Ferromagnetic Materials Based on Material Property Profiles

Area of Science:

• Materials Science
• Electromagnetism
• Non-Destructive Testing

Background:

• Accurate classification of ferromagnetic sheet materials with similar electromagnetic properties is challenging.
• Existing methods often struggle with signal interference and variations from uncontrollable factors.

Purpose of the Study:

• To develop a novel strategy for fine-grade classification of sheet ferromagnetic materials.
• To enhance the accuracy and reliability of material identification using the eddy current method.

Main Methods:

• Integration of three components: suppression of uncontrollable factors, reconstruction of material property profiles, and classification in a reduced-dimensional latent space.
• Utilizing a Taguchi-optimized probe for direct signal acquisition without prior processing.
• Application of Generative Adversarial Network (GAN) latent space for classification.

Main Results:

• The proposed strategy effectively identifies sheet samples using the eddy current method.
• The optimally designed probe achieved a high signal-to-noise ratio.
• Significant reduction in output-signal variations caused by uncontrollable factors was observed.

Conclusions:

• The novel strategy provides an innovative framework for effective material grade classification.
• The method demonstrates robustness against signal interference and environmental variations.
• This approach advances non-destructive testing capabilities for ferromagnetic materials.