A Zoning-Based Secure Control Approach Against Actuator Attacks in Industrial Cyber-Physical Systems

The integration of cyber and physical domains improves the flexibility and efficiency of industrial systems. However, it also introduces security threats into the physical processes of the systems. More seriously, most of field devices, such as sensors and actuators, in industrial systems are vulnerable to security threats because they are usually not deployed with security countermeasures. Therefore, how to secure the physical processes of these industrial cyber-physical systems becomes an emerging problem. To address this problem, a zoning-based approach is presented in this article for identification and mitigation of actuator attacks in multiple-input multiple-output processes governed by linear time-invariant systems. It partitions the set of actuators into multiple zones that ensures the controllability of the surviving zones when a zone is compromised. When a zone is intruded, the system states of the compromised zone and other zones cooperate to confirm that the detected attack is against actuators. Then, to mitigate the impact of the actuator attack on the system functions, the attacked actuators are isolated. After that, the residual actuators in the surviving zones are used to reconfigure the control loops of the systems. Case studies are conducted to illustrate the effectiveness of the presented approach.