Receding horizon load restoration for coupled transmission and distribution system considering load-source uncertainty

Abstract

This paper proposes a conditional value-at-risk (CVaR) based two-stage model predictive control (MPC) method for efficient dynamic load restoration decision-making in the coupled transmission and distribution (TS-DS) system with renewable energy. The CVaR values are employed to describe uncertainties of the load and source sides. It benefits on-line load restoration with uncertainties by fast uncertainty management and prediction error correction. In order to improve the computation of the multi-step load restoration optimization in the coupled TS-DS system, a two-stage load restoration model is constructed with the first stage relaxed multi-step optimization and the second stage single-step tracing optimization. By solving linear programming (LP), mixed integer linear programming (MILP) and mixed integer quadratic programming (MIQP) problems, the proposed CVaR based two-stage MPC method achieves on-line receding horizon load restoration of the coupled TS-DS system facing with load-source uncertainty. The effectiveness of the proposed method is validated using the IEEE-118 and IEEE-33 test systems, and a real-world coupled TS-DS system.