Cyber Attack Detection in Nonlinear Binary Distillation Column

Abstract

This thesis focuses on addressing the issue of cyber security for Industrial Control Systems (ICS). Evolution in computing and internet technology has encouraged increasing number of Industrial Control Systems (ICS) to be linked to cyber world giving rise to a new class of systems called Cyber Physical system (CPS) making them more vulnerable to cyber threats. The effect of cyber-attacks differs in cyber physical critical ICS compared to traditional ICT systems as they can cause damage to physical infrastructure posing threats to human health and environment. The four main objectives of this thesis are as follows: i. Mathematically model various malicious adversary attacks and intrusions. ii. Mathematically model dynamics of a crude oil distillation column (DC) and design its control system. iii. Analyze the effect of cyber-attacks on dynamic performance of the DC. iv. Design attack detection techniques and validate their performance and effectiveness on the model of a DC.