With the increase of wind power installed capacity

the stochastic behaviours of wind energy pose a great challenge to the day-ahead generation scheduling. Thus a large amount of reserve capacity is needed

which causes economic deterioration of the power system. An improved two-stage robust optimization with recourse method considering the rolling trading of inner-day electricity market is proposed based on the analysis of time-varying prediction uncertainty of wind power. In the first stage of day-ahead optimization process

reserve capacity of day-ahead generation scheduling is reduced considering expected power deviation of inner-day rolling trading. In the second stage

prediction model with high precision is used to calculate the inner-day regulated power to improve the economy of the power system. Correctness and effectiveness of the proposed method are verified by contrastive analysis cases.