Abstract: Prediction of the state of charge (SOC) of MH/Ni battery is very important for battery management system of electric vehicles.Through discussing the electrochemistry of MH/Ni system, a model based on radial basis function (RBF) neural network was employed to predict the state of charge of MH/Ni battery.The model was used to predict the state of charge at a certain state of discharging process.The proposed model had high prediction speed.The predicted SOC closely resembled the measured value.Artificial neural network technique is simple and understandable.It is a powerful tool to estimate the SOC of MH/Ni battery.

摘要： 电动车电池管理系统的核心任务是对电池荷电状态（SOC）进行预测.在分析了MH/Ni电池充放电反应机理的基础上，应用径向基函数（RBF）神经网络建立了预测MH/Ni电池荷电状态的模型，并且应用该模型对电池放电过程中某一状态下的荷电状态进行预测.该模型预测速度快，并且预测值与试验值吻合.人工神经网络建模技术简单直观，是预测MH/Ni电池SOC有力工具.