Day-ahead economic dispatch of power system with high proportion renewables based on distributionally robust joint chance constraints

As the proportion of wind power in the power system continues to increase, accurately understanding the characteristics of wind power forecast errors is crucial for enhancing grid safety and economic efficiency. First, this paper proposes a wind power forecast error uncertainty model considering the correlation in wind power output predictions by utilizing the Dirichlet process Gaussian mixture model, thereby effectively enhancing the accuracy of the wind power uncertainty model. On this basis, a day-ahead scheduling model considering wind power uncertainty is established. The model incorporates reserves and power flow constraints associated with wind power forecast errors and reformulates them into distributionally robust joint chance constrains. Subsequently, an affine adjustment strategy for conventional unit generation is introduced, effectively addressing the deviations between actual and scheduled wind power, to maintain system power balance. In addition, the study incorporates the wind curtailment planning as a decision variable in the optimization process, and develops a day-ahead scheduling model that considers wind power risk costs. Finally, the effectiveness of the proposed model is validated through simulations on the modified IEEE 30-bus test system, demonstrating significant improvements in economic efficiency and computational performance.