Intelligent Anomaly Detection Method of Gateway Electrical Energy Metering Devices using Deep Learning

Accurate anomaly detection of gateway electrical energy metering device is important for maintenance and operations in the power systems. Traditionally, was typically performed manually through analysis collected information. However, manual process time-consuming labor-intensive. In this condition, paper proposes a hybrid deep-learning model, which integrates Stacked Autoencoder (SAE) Long Short-Term Memory (LSTM), intelligently detecting abnormal events device. The proposed model named SAE-LSTM first uses SAE to extract deep latent features three-phase voltage data from device, then adopts LSTM separating based on extracted features. can effectively highlight temporal information data, thereby enhancing accuracy detection. simulation experiments verify advantages under different signal-to-noise ratios. experimental results real datasets demonstrate that it suitable devices practical scenarios.