Power System Flexibility Region Under Uncertainty With Respect to Congestion and Voltage Constraints

This paper evaluates, for hours-ahead forecasted states, how flexible is a power system operation to face uncertainty due to renewable generation. Focusing on static operation constraints (circuit congestion and bus voltage magnitude), the paper proposes to determine the system robust flexibility region under operation uncertainty. This robust flexibility region is defined as the maximal range of uncertainty that the system can accommodate with the support of available control actions (e.g. generation active/reactive power re-dispatch) without breaching static operation constraints. This goal is formulated as a non-standard non-convex, nonlinear robust optimization problem. The problem is solved through a two-stage decomposition algorithm, which loops between an upper level, that computes the robust flexibility region, and a lower level, that assesses whether it is robust with respect to any realization of uncertainty. The lower level relies on specific optimal power flow (OPF) formulations as non-convex nonlinear programs (NLPs) while the upper level resorts to quadratic programming (QP). The proposed approach is illustrated on the Nordic32 system.