Online Monitoring of Overhead Ground Wire Corrosion via Ultrasonic Guided Wave

Corrosion of overhead ground wires (OGWs) in power transmission systems poses significant challenges to structural integrity and operational safety. Traditional corrosion monitoring methods, such as the coupon weight loss method and electrochemical analysis, are time-consuming and lack real-time detection capabilities. To address these limitations, this paper proposes a self-powered wireless monitoring system based on ultrasonic guided waves (UGWs) and deep learning for real-time corrosion assessment of OGWs. The system integrates ultrasonic excitation and reception modules, a self-powered energy harvesting mechanism, and a wireless communication unit for remote data transmission. Ultrasonic guided wave signals are processed using Fourier Transform (FT) and Wavelet Transform (WT) to extract key features such as time-of-flight (ToF), amplitude attenuation, and frequency dispersion. A convolutional neural network (CNN) is employed to analyze the signal patterns and assess corrosion severity. Experimental results show that the proposed system accurately detects and quantifies corrosion defects, achieving a coefficient of determination (R2) of 0.83. The combination of ultrasonic guided wave technology, deep learning-based analysis, and self-powered wireless communication enables efficient, autonomous, and real-time corrosion monitoring of OGWs, providing a reliable solution for power grid maintenance and safety enhancement.