Abstract: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.

What problem does this paper attempt to address?

The paper primarily addresses the issue of anomaly detection in building energy time series data. Specifically, the research team developed a method based on one-dimensional deep convolutional generative adversarial networks (1D-DCGAN) to detect anomalies in energy time series data. The core problems that this paper attempts to solve are as follows: 1. **Energy Waste Identification**: By effectively identifying anomalous events in building energy consumption, it aims to reduce energy waste and thereby lower operational costs. 2. **Complex Anomaly Detection**: Traditional anomaly detection methods mainly focus on instantaneous anomalies (point anomalies), whereas the method proposed in this paper aims to identify continuous anomalous events (sequence anomalies), which is a more challenging task. 3. **Utilizing Generative Adversarial Networks (GANs)**: Leveraging the powerful capabilities of GANs to capture complex patterns and dependencies in normal data and generate synthetic data points similar to the original data, thereby achieving effective detection of complex anomalies. 4. **Optimizing GAN Training**: Employing the training process of Wasserstein GAN (W-GAN) to improve the stability and performance of the GAN model, addressing common issues in traditional GAN training such as gradient vanishing and mode collapse. 5. **Improving Reconstruction Loss Function**: Proposing Soft-DTW as a differentiable reconstruction loss function to replace Euclidean distance, used to measure the difference between two time series subsequences, thereby improving the accuracy of anomaly detection. 6. **Parallel Computing Acceleration**: Significantly reducing testing time and improving overall efficiency by parallelizing the reconstruction process of multiple data points. 7. **Sequence-Level Anomaly Scoring**: Combining reconstruction loss and anomaly scoring methods based on prior probability distribution in the latent space to identify anomalous time series segments. In summary, the paper proposes an innovative GAN-based method that not only effectively detects anomalies in energy time series but also significantly improves detection accuracy and efficiency through the introduction of the Soft-DTW loss function, parallel computing, and W-GAN techniques. This method is of great significance for improving energy management efficiency and reducing energy waste.

Generative Adversarial Network with Soft-Dynamic Time Warping and Parallel Reconstruction for Energy Time Series Anomaly Detection

Data anomaly detection for structural health monitoring using a combination network of GANomaly and CNN

DEGAN: Time Series Anomaly Detection using Generative Adversarial Network Discriminators and Density Estimation

DTAAD: Dual Tcn-Attention Networks for Anomaly Detection in Multivariate Time Series Data

Research on DUAL-ADGAN Model for Anomaly Detection Method in Time-Series Data

TSI-GAN: Unsupervised Time Series Anomaly Detection using Convolutional Cycle-Consistent Generative Adversarial Networks

Deep Generative model with Hierarchical Latent Factors for Time Series Anomaly Detection

Segmentation-Based Adversarial Denoising Auto-Encoder for Anomaly Detection in Multivariate Time Series Data

MTS-DVGAN: Anomaly Detection in Cyber-Physical Systems using a Dual Variational Generative Adversarial Network

Time Series Anomaly Detection with Adversarial Reconstruction Networks

A Deep Learning Approach to Anomaly Sequence Detection for High-Resolution Monitoring of Power Systems

Anomaly Detection in Power Generation Plants with Generative Adversarial Networks

Hybrid Anomaly Detection via Multihead Dynamic Graph Attention Networks for Multivariate Time Series

Time Series Anomaly Detection in Smart Homes: A Deep Learning Approach

A Novel Multi-dimensional Time-series Data Anomaly Detection Model Based on Generative Adversarial Network Aided Autoencoder

Temporal Convolutional Networks for Anomaly Detection in Time Series

Dual-encoder Bidirectional Generative Adversarial Networks for Anomaly Detection

Disentangled Dynamic Deviation Transformer Networks for Multivariate Time Series Anomaly Detection

Reconstruct Anomaly to Normal: Adversarial Learned and Latent Vector-constrained Autoencoder for Time-series Anomaly Detection

Convolutional Transformer based Dual Discriminator Generative Adversarial Networks for Video Anomaly Detection