Resilience of Smart Power Grids to False Pricing Attacks in the Social Network

Abstract

Communication through social media can help engage end users to improve the efficiency of demand-side management in smart power grids. However, this opens a channel between the social network and the power grid through which malicious attackers can publish false information that can actually cause problems to the power grid. In this paper, we analyze this new problem by modeling a social network-coupled smart grid and investigating its vulnerability to false pricing attacks in the social network. The energy consumption profile based on social information is modeled as a consumption rescheduling problem, which aims to maximize the benefit of demand-side management. The false price spreading process is described by a multi-level influence propagation model, which takes into account the personalities of the end users. Different attack strategies are considered and the power operator's response is modeled. The residual ampacity of distribution lines and the expected energy not supplied are adopted to quantify the impacts of the attacks on the power system. To account for the stochastic characteristics of the influence propagation process, Monte Carlo simulation is utilized. The proposed modeling and analysis framework is applied on a modified IEEE 13 nodes test feeder and a notional social network. The vulnerability to attack is analyzed at both component and system levels.