Generative Adversarial Network with Soft-Dynamic Time Warping and Parallel Reconstruction for Energy Time Series Anomaly Detection
This work addresses identifying anomalous energy consumption in buildings, presenting an incremental improvement with a novel hybrid method for faster detection.
The paper tackles anomaly detection in energy time series by using a 1D DCGAN with Soft-DTW as a differentiable reconstruction loss, achieving faster detection through parallel computation. It demonstrates effectiveness on hourly data from 15 buildings, showing Soft-DTW outperforms Euclidean distance.
In this paper, we employ a 1D deep convolutional generative adversarial network (DCGAN) for sequential anomaly detection in energy time series data. Anomaly detection involves gradient descent to reconstruct energy sub-sequences, identifying the noise vector that closely generates them through the generator network. Soft-DTW is used as a differentiable alternative for the reconstruction loss and is found to be superior to Euclidean distance. Combining reconstruction loss and the latent space's prior probability distribution serves as the anomaly score. Our novel method accelerates detection by parallel computation of reconstruction of multiple points and shows promise in identifying anomalous energy consumption in buildings, as evidenced by performing experiments on hourly energy time series from 15 buildings.