Power distribution network fault section positioning method

The invention provides a power distribution network fault section positioning method. The method comprises the following steps: collecting zero-sequence current at two ends of each section after a power distribution network has a fault; preprocessing the zero-sequence current, and then carrying out short-time Fourier transform to obtain a time-frequency diagram; and performing feature extraction and classification on the time-frequency graph according to ConvNeXt introduced with transfer learning to obtain a fault section positioning result of the power distribution network. The method is high in high resistance tolerance, has a certain anti-noise capability, and can still carry out section positioning under the condition that part of data is lost. 本发明提供了一种配电网故障区段定位方法，包括：采集配电网发生故障后的各区段两端的零序电流；对所述零序电流进行预处理后，再进行短时傅里叶变换，获取时频图；根据引入迁移学习的ConvNeXt，对所述时频图进行特征提取与分类，获取配电网故障区段定位结果。本发明耐受高阻能力强，具有一定的抗噪声的能力，且在部分数据丢失的情况下依然能进行区段定位。