Security-constrained economic dispatch using semidefinite programming

Abstract

This paper models the security-constrained economic dispatch problem with post-contingency corrective actions. In the nonlinear AC setting, the model is a large-scale nonconvex problem and is difficult to solve. A novel solution approach is proposed to deal with the scale and nonconvexity issues separately and effectively. The key point is to approximate the nonconvex AC feasibility problem with its semidefinite programming (SDP) relaxation and use these SDP models as a convex subproblem within a Benders' decomposition framework. Numerical experiments demonstrate the superior solution quality of this approach and its tractability for IEEE test cases, while suggesting future work for computational improvements on large-scale problems.