Two-Stage Robust Economic Scheduling of Power System Based on LHS

The uncertainty of renewable energy generation and load demand is an important factor that affects the optimal scheduling of power systems. Robust schedule model regards these uncertainties as intervals and considers the upper or lower boundary values of those intervals as worst-case scenarios, even if the probability of such situations is very low. The model may result in higher operation costs for the power system and excessive conservatism. In order to reduce the conservatism of robust optimal scheduling and improve the economic efficiency of system operation, a two-stage robust economic scheduling of power system based on Latin hypercube sampling is proposed in this paper. The model uses Latin hypercube sampling to re-evaluate the uncertainty of wind power, photovoltaic power and load power, and to obtain new estimated interval of renewable energy generation and load. In addition, the shiftable load is also considered in the paper, which effectively reduces the peak-valley difference and decrease the difficulty of peak regulation in the power system. The Benders decomposition algorithm is used to decompose the model into the main problem and subproblem. The max-min problem in the subproblem is decoupled by the dual theory. A case that includes wind power, photovoltaic power and energy storage is simulated to test the performance of the proposed model. And the results show that the proposed model can decrease the conservatism, improve the economic efficiency of system operation, and reduce the difficulty of peak regulation in the power system by adjusting the shiftable load.