Deep learning-based corrosion-like defect localization technique for high-voltage cable aluminum sheaths using guided waves

Abstract

High-voltage cable accessories are important parts of a power system and are needed to ensure the reliability of electrical connections. However, their aluminum sheaths are prone to corrosion under complex working conditions, which has a detrimental effect on the normal operation of the electric system. Ultrasonic guided wave detection is a promising non-destructive testing method suited to detecting aluminum sheath corrosion in complex high-voltage cable accessory structures. However, current ultrasonic guided wave detection methods still require manual extraction of the signal features and have a high reliance on professional knowledge. This paper proposes a deep learning-based corrosion-like defect localization technique for high-voltage cable aluminum sheaths using guided waves. First, the original ultrasonic guided wave signals of corrosion defects at different locations are obtained using an ultrasonic guided wave detection platform. Then, the original signals are input into a variable auto-encoder (VAE) network to obtain a low-dimensional representation for automatic feature extraction. Finally, the low-dimensional representation is input into a gated recurrent unit (GRU) based recurrent network for corrosion defect localization. In the feature extraction stage, the VAE can automatically extract the effective features and avoid the interference of noisy signals. In the defect localization stage, the GRU can accurately identify the location of corrosion defects. The experimental results indicate that the VAE-GRU method is capable of accurately identifying corrosion defects based on the original signals.