Miniaturized Design of Electromagnetic Acoustic Transducer and Its Application in Defect Detection

To address the limitations of bulky size and low energy conversion efficiency in conventional electromagnetic acoustic transducers (EMATs), this study proposes a miniaturized EMAT optimized for surface defect detection in thick steel plates. A two-dimensional multiphysics simulation model is developed in COMSOL to analyze the transduction process. Key design parameters including magnet geometry, coil configuration, and lift-off distance are optimized using orthogonal experimental design, resulting in a 52.4% volume reduction and a 4.9% increase in signal amplitude. To mitigate the performance degradation caused by miniaturization, a complementary ensemble empirical mode decomposition algorithm is employed, improving the signal-to-noise ratio (SNR) from 10 dB to 28.73dB. Experimental validation demonstrates strong beam directivity with a 12.93° half-angle and a sidelobe suppression ratio of 13.78dB. The surface wave velocity deviates by only 1.25% from the theoretical value, and the defect localization error remains within 1 mm. The proposed EMAT offers a compact, non-contact solution with high precision and robustness, well suited for robotic nondestructive testing (NDT) in complex industrial environments.