Enhancing Cybersecurity in Process Control Systems using Data-Driven Approaches

The robustness of process control systems (PCS) to cyberattacks has become an increasingly important topic, especially in light of the increased dependence of process operations on networked control system architectures wherein dedicated sensor-controller and controller-actuator links are replaced by real-time, wired or wireless, shared communication networks. While these attacks have been traditionally solved using Information Technology (IT)-based solutions, recent cases with increased severity and increased frequency have shown the inadequacy of such methods. Moreover, in cyber-physical systems such as PCS, these breaches have physical implications on the process which results in poor quality products, higher operating costs, or even loss of lives (ref Fig. 1). This has revealed the need for addressing these issues using Operational Technology (OT)-based solutions against such attacks. OT-based solutions require the design of efficient cyberattack detection, estimation, and mitigation methods for enhancing cyber resilience of PCS. This thesis aims to develop methods for resilience of PCS against cyberattacks using operational data-driven approaches.