Wavelet Transform-Based Overvoltage Feature Extraction for High-Speed Moving Trains

Overvoltage will be generated when the train lifts the bow, over-phase and off-line of the bow network, resulting in a rise of the car body potential, affecting the on-board electrical equipment, endangering the safety of equipment and personnel, and posing a hidden danger to train safety. Based on wavelet transform and multi-scale energy statistics, this paper analyses transient high-frequency overvoltage signals in the time-frequency domain, calculates the energy values of overvoltage signals at different scales, extracts characteristic quantities, quickly identifies overvoltage types, realizes digital and intelligent train operation, and provides support for transient fault diagnosis and troubleshooting.