Title | A dynamic defense-attack game scheme with incomplete information for vulnerability analysis in a cyber-physical power infrastructure |
Publication Type | Conference Paper |
Year of Publication | 2019 |
Authors | Gao, Boyo, Shi, Libao, Ni, Yixin |
Conference Name | 8th Renewable Power Generation Conference (RPG 2019) |
Date Published | oct |
Keywords | composability, cyber-physical power systems, GAME THEORY CYBER-PHYSICAL ATTACKS, Metrics, particle swarm optimization, Power Grid Vulnerability Assessment, pubcrawl, resilience, Resiliency, vulnerability analysis |
Abstract | The modern power system is experiencing rapid development towards a smarter cyber-physical power grid. How to comprehensively and effectively identify the vulnerable components under various cyber attacks has attracted widespread interest and attention around the world. In this paper, a game-theoretical scheme is developed to analyze the vulnerabilities of transmission lines and cyber elements under locally coordinated cyber-physical attacks in a cyber-physical power infrastructure. A two-step scenario including resources allocation made by system defender in advance and subsequent coordinated cyber-physical attacks are designed elaborately. The designed scenario is modeled as a game of incomplete information, which is then converted into a bi-level mathematical programming problem. In the lower level model, the attacker aims at maximizing system losses by attacking some transmission lines. While in the higher level model, the defender allocates defensive resources, trying to maximally reduce the losses considering the possible attacks. The payoffs of the game are calculated by leveraging a strategy of searching accident chains caused by cascading failure analyzed in this paper. A particle swarm optimization algorithm is applied to solve the proposed nonlinear bi-level programming model, and the case studies on a 34-bus system are conducted to verify the effectiveness of the proposed scheme. |
DOI | 10.1049/cp.2019.0285 |
Citation Key | gao_dynamic_2019 |