Abstract: Demand response (DR) has been an important mean for peaking adjustment， operation economy raising and wind curtailment reduction. Considering the time characteristics of wind power and demand-side resources， as well as the stackelberg game between power grid dispatch and load aggregators (LAs)， this paper establishes a two-stage hierarchical dispatch model. In this paper， a fuzzy chance constraint programming is used to deal with the uncertainties of wind power， loads and demand-side resources. Aiming at hierarchical distributed intra-day model with multiple load aggregators， this paper adopts analytical target cascading (ATC) in model decoupling， and the parallel solution of each part is realized. The test results show that the costs of power system has been decreased to 95.5% and 94% by dispatching DR and LAs， respectively， while each LA earnings have been increased more than 8%， which means the proposed strategy can enhance economy of power system and take all load aggregators earnings into account， and promote wind power accommodation as well.

Key words: demand response (DR), load aggregator (LA), stackelberg game, analytical target cascading (ATC)