A Real-time Anomaly Detection Using Convolutional Autoencoder with Dynamic Threshold

arXiv:2404.04311v1 Announce Type: new
Abstract: The majority of modern consumer-level energy is generated by real-time smart metering systems. These frequently contain anomalies, which prevent reliable estimates of the series' evolution. This work introduces a hybrid modeling approach combining statistics and a Convolutional Autoencoder with a dynamic threshold. The threshold is determined based on Mahalanobis distance and moving averages. It has been tested using real-life energy consumption data collected from smart metering systems. The solution includes a real-time, meter-level anomaly detection …

abstract anomaly anomaly detection arxiv autoencoder consumer cs.ai cs.lg detection dynamic energy evolution generated hybrid modeling modern real-time series smart statistics systems threshold type work