Cyber attack resilient monitoring and control framework from transmission to active distribution power networks

Abstract

In today’s world, our power grids are becoming smarter, due to the rapid and wide spread integration of digital sensors, computers, Internet and Communication Technologies (ICTs) etc. The remarkable advancements of such modernized technologies and far-reaching use of various sophisticated remotely control devices have transformed our age old traditional energy sector from purely physical system to somewhat complex Cyber Physical Systems (CPS). While this CPS infrastructure has effectively prevented various disastrous scenarios like blackouts, uncontrolled shutdowns, unwanted frequency and voltage fluctuations, power loss and grid instability, nevertheless the close integration of power system’s physical operation with that of unsecured cyber networks brings a new risk of cyber threats via unauthorized control access to the communication channels, exploitation of networking protocols, forced equipment outage and damage, manipulation of sensing and control signals and any kind of other sabotaging activities that jeopardize the normal monitoring and control functionality of power grid ranging from power transmission to power distribution. Thus this thesis aims to provide an overall comprehensive security solution towards developing a Cyber Attack Resilient Monitoring and Control (CARMC) framework by unveiling vulnerabilities across transmission (T-System) to distribution (D-System) power networks. The research begins by identifying and addressing the key vulnerabilities introduced in the T-System networks. Following this, a comprehensive attack resilient framework is developed based on strategic placement of Phasor Measurement Units (PMUs) at such optimal locations that safeguard a minimal sets of measurements in order to make the system resilient against any kind of False Data Injection Attacks (FDIAs) on those selected vulnerable lines. After securing a set of critical meters by developing a secured metering infrastructure, the next research study of T-System focused on the detection and control technique of another simple but impactful attack named, Replay Attack (RA) that targets one of the core power system monitoring application of energy management system i.e Power System State Estimation (PSSE). In order to safeguard the PSSE against RA, the proposed technique leverages the secured phasor measurements obtained from the optimal PMU locations through a hybrid state estimator (HYB-SE) to correct the manipulated conventional meter readings. The later half of the thesis focuses on detecting and mitigating vulnerabilities associated to the D-System networks, specifically Microgrids (MGs), where the Distributed Energy Resource (DER) controller’s and its communication links are being targeted by the attacker to cause voltage and frequency instability to the grid. To this end, for the detection, classification and localization of cyber attacks, a statistical two-sample hypothesis test, called as Maximum Mean Discrepancy (MMD) index and a rule based algorithm coupled with XGBoost classifier is utilized respectively. After the attack being detected and classified successfully, the next study aims to develop a cyber-attack resilient control framework for the MG system based on designing Unknown Input Observer’s (UIO) states and Back-stepping Integrated Sliding Mode Controller (BSMC) to mitigate the overall effect of injected attack into the DER’s secondary controller. Finally, with the aim of having a secured monitoring infrastructure in D-System, the last research study addressed the problem of accurate detection of islanding event in the presence of cyber-attacks. The effectiveness of the proposed CARMC framework is validated through extensive offline simulation performed in MATLAB, PSCAD, RSCAD software and real-time testing incorporating various hardware platforms such as Real-Time Digital Simulator (RTDS) and dSPACE 1104 Research & Development controller board. The results demonstrate the ability of the CARMC framework to bolster the resilience of transmission and active distribution networks against diverse cyber threats.