Large-scale wind power integration brings about many challenges to power system. Energy storage is regarded as an effective solution to the problems and some demonstration projects have been constructed. As the output power of wind turbines is a stochastic variable which is determined by wind velocity, it is very difficult to choose the operation strategy of energy storage. Probability method can be a wise selection of this kind of problem. Mathematical model for seeking the energy storage operation strategy is developed based on chance-constrained programming. Genetic algorithm is used to solve the optimization problem, and Monte-Carlo method is applied to deal with the chance-constrained question.