A Genetic Algorithm Approach for the Analysis of Electric Grid Interdiction with Line Switching

Abstract

This paper addresses the vulnerability analysis of the electric grid under terrorist threat. This problem is formulated as a mixed-integer nonlinear bilevel program. In the upper-level optimization, the terrorist agent maximizes the damage caused in the power system, which is measured in terms of the level of system load shed. On the other hand, in the lower-level optimization, the system operator minimizes the damage by means of an optimal operation of the power system. The distinctive modeling feature introduced in this paper is that, among the different corrective actions available, the system operator has the capability to modify the network topology. Due to its nonconvexity and nonlinearity, exact solution techniques are not currently available. This paper proposes a novel genetic algorithm approach to achieve near optimal solutions in moderate computing times. Some numerical results obtained by the proposed algorithm are provided and compared with those published, based on the IEEE Reliability Test System.