Controller Cyber-Attack Detection and Isolation.

Area of Science:

• Engineering
• Computer Science
• Cyber Security

Background:

• Industrial control systems (ICS) are vulnerable to cyber-attacks and process faults.
• Existing methods for fault detection and isolation (FDI) and control loop performance assessment have limitations in integrated cyber security.
• Cybernetic faults can disrupt ICS operations, necessitating robust diagnostic tools.

Purpose of the Study:

• To develop and evaluate an integrated approach for detecting and isolating cyber-attacks in industrial control systems.
• To enhance the cyber security of critical infrastructure by diagnosing anomalies affecting control loop performance.
• To validate the proposed method using real-world operational data from a power unit boiler.

Main Methods:

• Integration of model-based control algorithm checking with control loop performance indicator tracking.
• Utilizing a binary diagnostic matrix for anomaly isolation.
• Employing standard operational data: process variable (PV), setpoint (SP), and control signal (CV).

Main Results:

• The proposed integrated approach effectively detects and isolates anomalies, including cyber-attacks.
• The method demonstrated applicability and effectiveness on a superheater control system in a power unit boiler.
• Standard operational data proved sufficient for diagnosing complex cybernetic faults.

Conclusions:

• The integrated cyber security approach offers a promising solution for protecting industrial control systems.
• The method is effective in diagnosing both process faults and cyber-attacks using readily available data.
• Further research is needed to explore the full applicability and limitations of the approach in diverse ICS environments.