Multi-period vulnerability analysis of power grids under multiple outages: An AC-based bilevel optimization approach

Abstract

This paper describes a methodology for the N-k contingency analysis of bulk power systems. The method encompasses the evaluation of contingencies’ effects on the power system over a range of system demand levels. The proposed model is inherently a multi-period bilevel optimization problem. Unlike the conventional bilevel optimization problems for the N-k contingency analysis, the proposed model considers the effects of reactive power dispatch, losses, and voltage profile. In doing so, the problem is formulated as a multi-period AC-based bilevel mixed-integer nonlinear programming (MINLP) problem. To guarantee the global optimality of the solution, this paper linearizes and then transforms it into a one-level mixed-integer linear programming (MILP) problem using different linearization techniques and the duality theory. The simulation results on the annual load profile of the IEEE Reliability Test System (RTS) verify the effectiveness of the proposed model.