A new method for solving preventive security-constrained optimal power flow based on linear network compression

In a security-constrained optimal power flow (SCOPF) problem, control actions are so calculated that not only the objective function is minimized in base case (no contingency), but also the constraints are satisfied for post-contingency states. In a large-scale power system, if all contingency states are considered, the dimensions of optimization problem becomes so large that it may fail to converge to a solution. In this paper, to reduce the dimensions of problem, first for each contingency the network is partitioned into three regions of internal, external and boundary. The internal region is the one that its variables are more affected by the contingency. The rest of the network is called the external region. The boundary region consists of some buses at the external one connecting external region to internal one. After partitioning, the equations of external and boundary regions are linearized and some of external region variables and equations are omitted. This work reduces the network equations and consequently reduces the problem dimensions. In addition, two commonly used contingency filtering methods are described and compared. The simulations are done on the IEEE 39 and 118-bus test systems.