HAI Zhichen, LIU Yongjun, SHAO Kexin, et al. A Traveling Wave Time-Frequency Spectrogram and Improved CNN Based Method for Ground Fault Location in Substation DC Systems[J]. 2025, 53(7): 100-107.
HAI Zhichen, LIU Yongjun, SHAO Kexin, et al. A Traveling Wave Time-Frequency Spectrogram and Improved CNN Based Method for Ground Fault Location in Substation DC Systems[J]. 2025, 53(7): 100-107. DOI: 10.20204/j.sp.2025.07013.
A Traveling Wave Time-Frequency Spectrogram and Improved CNN Based Method for Ground Fault Location in Substation DC Systems
This paper proposes a novel method for ground fault location in substation DC systems using traveling wave time-frequency spectrograms and an improved convolutional neural network (CNN). First, fault-induced traveling wave signals undergo frame blocking and windowing processing, with time-frequency spectrograms extracted via short-time Fourier transform . These spectrograms are then input as samples to construct a CNN-based fault location model. The sparrow search algorithm is further employed to optimize model parameters, enhancing localization accuracy. Finally, case studies are conducted to evaluate the performance of the proposed method. The results demonstrate superior convergence and higher accuracy compared to conventional fault location methods.
ISLAM A ZENHOM,MOSTAFA F SHAABAN,W A OMRAN,et al. Grid interactive charging of evs in pv-powered parking lots considering uncertainties[J].IEEE Access.2023,11:111292-111301.