With an analysis and estimation of the automatic generation control（AGC）capacity demand on intermittent power，wind power scenario is generated by Latin hypercube sampling and reduced later.And also，the uncertainty probability of system N-1 fault is forecasted by Markov chain.On this basis，a flexible stochastic optimal scheduling model considering intermittent power and energy storage is developed，while the unit commitment，unit power distribution and hierarchical coordinated and optimal configuration of the AGC and reserve capacity are taken into account.The proposed approach and model take into account the uncertainty probability and severity of the system N-1 fault and intermittent power，the cost of AGC and reserve capacity ancillary services，the remaining charge and discharge ability of energy storage devices，as well as the network constraints.This complicated model is linearized and solved by commercial mixed-integer linear programming solver CPLEX.The numerical cases of IEEE 6-bus and IEEE 39-bus system have illustrate the validity of the approach proposed.