Wind power has been promoted to mitigate the energy crisis worldwide. However, owing to the chaotic nature of atmospheric movement, wind power generation always exhibits nonlinear and non-stationary uncertainties. Designing an energy storage system for dispatch of a wind farm is an effective integration solution. This paper presents a framework for integrating the stochastic unit commitment, energy storage, dispatch formulations that account for wind power uncertainty. Generation cost and the fluctuation of load are proposed as multi-objective functions with the constraint functions. Numerical results on three cases including thermal generators, wind farm and energy storage and comparisons with results obtained using state-of-art optimization software, show that can be solved most efficiently and the energy storage device improves the flexibility of system operation.