Files
Download Full Text (273 KB)
Description
Oil pipelines, water plant systems, and other critical infrastructure are managed and operated by industrial control systems (ICS). These systems safeguard the operations of critical infrastructures, requiring minimal disruption from cyberattacks or malfunctions. The use of anomaly detection methods in control systems (ICS) can reduce system interruptions. However, anomaly detection methods often require annotated data, which may not be available for the control system. Additionally, the datasets used for the control systems do not include sensor outputs and environmental data, resulting in a restricted view of the system. This research investigates how SSL models can be applied to different control system data streams. The study will develop a framework for fusing sensor data, network data, and environmental data. The approach consists of two phases. The first phase includes training SSL models on single data types (e.g., sensor dataset) and an approach for merging multiple data streams. The second phase will combine the SSL model with the aggregated data for anomaly detection. The contributions of this research effort include data fusion framework for control systems and an SSL model trained on the combined dataset without labels for anomaly detection.
Publication Date
3-2025
Department
Computer Science
City
Mobile
Disciplines
Artificial Intelligence and Robotics | Databases and Information Systems | Information Security
Recommended Citation
Gladney, Barbara, "Learning Without Labels: A Self-Supervised Learning Approach for Anomaly Detection in Control Systmes" (2025). Shelby Hall Graduate Research Forum Posters. 10.
https://jagworks.southalabama.edu/southalabama-shgrf-posters/10
Included in
Artificial Intelligence and Robotics Commons, Databases and Information Systems Commons, Information Security Commons
