Reactive power management is very important in power systems for the secure transmission of active power, especially when a part of system generation is provided by stochastic sources like wind energy. This paper presents a new algorithm for reactive power management in the presence of wind generators and considering the stochastic nature of these sources and load simultaneously .In this regard, the proposed probabilistic algorithm, minimizes the overall cost function of the system considering the cost of each of the reactive power sources including wind generators. Besides economic issues, the voltage stability margin, having sufficient reactive power reserve in each area of voltage control and considering transmission congestion probability as technical aspects of the planning, have been investigated .Another advantage of this method compared to the previous one, is using of doubly-fed induction generator (DFIG) and its capability in providing reactive power considering the constraints of grid side and rotor side converters. The proposed optimization algorithm uses a multi objective function with different weighting coefficients. This algorithm is applied to minimize total reactive power, cost and losses and maximize voltage stability margin and reactive power reserve, simultaneously, meanwhile the probabilistic nature of wind and load forecasting inaccuracy is considered in this algorithm. The proposed method is implemented on the IEEE 30-bus test system and the simulation results demonstrate the effectiveness of proposed algorithm in real conditions for PMSMs against internal faults, especially inter-turn faults.