Transient Signal Sensing Device Based On Time-frequency Domain And Deep Neural Network

Transient signal sensing devices are primarily used for fault detection and power quality analysis in distribution networks. To enable more accurate and comprehensive evaluation of fault data in distribution networks, we have developed a transient signal analysis method based on time-frequency domain analysis and deep learning. This method leverages a model that combines time-frequency domain signal analysis with deep neural networks, enabling the device to capture fault characteristics and perform fault identification. The model not only focuses on individual characteristic parameters but also conducts comprehensive analysis and mining of multiple characteristic parameters, thereby determining the root cause of faults and their specific occurrence locations. This approach facilitates a more thorough and in-depth fault analysis, contributing to effective fault handling and prevention in power grids.