The field of anomaly detection and cyber-physical system security is rapidly evolving, with a focus on developing innovative methods to detect and prevent potential threats. Recent studies have explored the use of deep learning techniques, such as autoencoders and generative adversarial networks, to improve anomaly detection in various domains, including power systems and industrial control systems. Additionally, researchers have investigated the application of reinforcement learning and meta-learning to enhance the robustness and adaptability of anomaly detection models. Notable papers in this area include the proposal of a pioneering AI-based control framework for robust short-term voltage stability assessment under cyber-attacks, and the development of a novel approach for log-based anomaly detection using learned adaptive filters. Furthermore, the use of topology-preserving loss functions and multi-view anomaly detection methods has shown promise in improving the accuracy and effectiveness of anomaly detection systems. Overall, these advancements have the potential to significantly enhance the security and resilience of cyber-physical systems, and are expected to continue shaping the direction of research in this field. Noteworthy papers include: AI-Enhanced Resilience in Power Systems, Improving log-based anomaly detection through learned adaptive filter, and Anomaly Detection in Time Series Data Using Reinforcement Learning, Variational Autoencoder, and Active Learning.
Advancements in Anomaly Detection and Cyber-Physical System Security
Sources
AI-Enhanced Resilience in Power Systems: Adversarial Deep Learning for Robust Short-Term Voltage Stability Assessment under Cyber-Attacks
Improving log-based anomaly detection through learned adaptive filter
Anomaly Detection in Time Series Data Using Reinforcement Learning, Variational Autoencoder, and Active Learning
Moving Target Defense Against Adversarial False Data Injection Attacks In Power Grids
A Robust Method for Fault Detection and Severity Estimation in Mechanical Vibration Data
Multi-Flow: Multi-View-Enriched Normalizing Flows for Industrial Anomaly Detection
Pyramid-based Mamba Multi-class Unsupervised Anomaly Detection
Quantifying Robustness: A Benchmarking Framework for Deep Learning Forecasting in Cyber-Physical Systems
Diagnostic Method for Hydropower Plant Condition-based Maintenance combining Autoencoder with Clustering Algorithms
Deep-Learning-Directed Preventive Dynamic Security Control via Coordinated Demand Response
Directional Sign Loss: A Topology-Preserving Loss Function that Approximates the Sign of Finite Differences
Loss Functions in Deep Learning: A Comprehensive Review
iADCPS: Time Series Anomaly Detection for Evolving Cyber-physical Systems via Incremental Meta-learning
Attention-Based Multi-Scale Temporal Fusion Network for Uncertain-Mode Fault Diagnosis in Multimode Processes
AnomalousNet: A Hybrid Approach with Attention U-Nets and Change Point Detection for Accurate Characterization of Anomalous Diffusion in Video Data
Enhanced Entropy-Based Metric for Characterization of Delayed Voltage Recovery
Novel Data-Driven Indices for Early Detection and Quantification of Short-Term Voltage Instability from Voltage Trajectories
Reconstruction-Free Anomaly Detection with Diffusion Models via Direct Latent Likelihood Evaluation
Autoencoder-Based Detection of Anomalous Stokes V Spectra in the Flare-Producing Active Region 13663 Using Hinode/SP Observations
Hybrid Temporal Differential Consistency Autoencoder for Efficient and Sustainable Anomaly Detection in Cyber-Physical Systems
Restoring Feasibility in Power Grid Optimization: A Counterfactual ML Approach
AMAD: AutoMasked Attention for Unsupervised Multivariate Time Series Anomaly Detection
ASRL:A robust loss function with potential for development
LCL Resonance Analysis and Damping in Single-Loop Grid-Forming Wind Turbines