Vanadium redox battery (VRB) system is a new type of energy storage system, which can be used to shave power grid peak, improve power quality, and smooth fluctuations of power and Voltage in photovoltaic and wind power systems. The state of charge (SOC) of VRB is an indication of how much energy is stored in the battery. Estimating the VRB SOC accurately in real time is very important when the battery is used to power system. In consideration of deficiencies of existing methods, a new approach is introduced to use Extended Kalman Filter (EKF) method. In this paper, state space model of VRB based on its equivalent circuit model is proposed, and SOC Estimation Using EKF Method for VRB is described in detail. By using Matlab software, it can be proved that the EKF method can estimate real-time SOC of VRB and is more accurate than existing methods.