Research on abnormal data detection in distribution networks using a parallel network integrating GraphSAGE and CNN-GRU

As distribution networks advance toward digitalization and intelligence, data quality has become a critical factor for ensuring their safe and stable operation. To address the challenge of anomaly detection in distribution network data, this paper proposes a detection method based on a parallel network architecture integrating GraphSAGE and CNN-GRU. GraphSAGE identifies abnormal nodes at the network level, transforms measurement data into time-frequency spectrograms, and then classifies anomalies using the CNN-GRU model, ultimately achieving precise anomaly classification and detection.