Master Thesis Defense - May 15, 2017: Towards a Smarter Power Grid: Vulnerability Assessment and Security Metric Deployment

Parisa Akaber

Monday, May 15, 2017 at 3:00 p.m.
Room EV001.162

You are invited to attend the following M.A.Sc. (Information Systems Security) thesis examination.

Examining Committee

Dr. A. Youssef, Chair
Dr. C. Assi, Supervisor
Dr. M. Debbabi, Supervisor
Dr. L. Wang, CIISE Examiner
Dr. A. Agarwal, External Examiner (ECE)

Abstract

Smart Grid is considered as one of the most critical cyber-physical infrastructure; leveraging the advanced coupled communication infrastructure, it is designed to address the limitations and drawbacks of the current power grid and offer a more available, reliable, and efficient power delivery system. Despite its promised advantages, coupling a cyber system with the power grid would increase the grid attack surface by adding known cyber vulnerabilities and threats. Furthermore, security solutions proposed for the traditional power system may not be applicable for the smart grid since they do not consider all smart grid added characteristics (e.g., synchronization). Therefore, it is crucial to lay out a study for the smart grid vulnerabilities and propose corresponding security evaluation and mitigation techniques.

In this thesis, our objective is to model the smart grid as a cyber-physical network considering all the characteristics of power and communication networks as well as the interdependencies among their component. We first propose a contingency analysis security evaluation framework for the smart grid considering concurrent failures resulting from malicious compromises. The proposed framework enables the utility to quantify and monitor the criticality level of the system under study from the security perspective, and decide on proper mitigation/protection actions to avoid catastrophic power outages.

Then, we investigated the critical link (power or communication) identification problem in the smart grid. We highlight the importance of considering the interdependencies among the power and communication network components by showing how a single failure in one side of the grid (cyber or physical) could cascade through both sides and disrupt the power delivery service for a large area immediately. We study the minimum number of links whose removal would have the largest impact on the system in terms of unserved load. The result of this study is beneficial for efficient and optimal resource allocation while designing protection mechanisms for the grid.

Finally, we address the power service restoration problem through network reconfiguration in the presence of distributed energy storage systems. Service restoration is a mandatory procedure which should be performed after any failure occurrence in order to increase the consumer satisfaction and decrease the penalty paid by utility. In this chapter, an optimal restoration approach is devised which is a combination of minimizing the restoration time, unserved load, and energy storage usage cost.