The modeling method for uncertainty sets in the robust interval economic dispatch is discussed.Firstly，a robust interval economic dispatch model is developed with probabilistic boundaries of uncertainty variables.Considering various wind power prediction information provided by the system control centers，different methods are proposed for modeling of uncertainty sets with assigned confidence levels，in which wind power correlation among wind farms is taken into consideration.The proposed approach enables system operators to adjust the conservativeness of robust dispatch schedules flexibly according to the tolerance of risks.Finally，Monte Carlos simulations are carried out on a modified IEEE 24 bus system to verify effectiveness of the uncertainty set modeling methods，and to compare the performance of these schemes.