Robust tie-line scheduling for multi-area power systems with finite-step convergence

The problem of robust tie-line scheduling is considered, which premeditates boundary phase angles such that the overall cost in real time is minimum under the worst scenario of uncertainty. The proposed method alternatively solves a deterministic tie-line scheduling under a specific realization of uncertainty and a worst scenario searching problem for the resulting boundary state. In particular, the deterministic tie-line scheduling is efficiently solved via critical region projection with sub-gradient. The proposed method has finite-step convergence and typically converges rapidly with outstanding computation and communication efficiencies.