基于蚁群神经网络算法的电池健康状态估计

电池健康状态(SOH)是进行电池健康监管和维护的重要依据。以某种车用磷酸铁锂单体电池为实验对象,提出了一种蚁群算法优化后的神经网络算法,以电池直流内阻定义SOH,并将该算法应用到电池健康状态估计模型。结果表明所提出的模型和方法预测电池最大直流内阻误差为0.1 mΩ,平均误差为0.049 mΩ,表明该方法能较为准确地预测电池直流内阻,实时反映电池的健康状态。

State of health estimation for lithium-ion battery based on ant colony neural network

The state of health (SOH) of the battery is an important basis for the battery health monitoring and maintenance.One kind of the lithium-iron phosphate monomer battery for the vehicle was chosen as the experiment object.A new neural network algorithm optimized by the ant colony algorithm was proposed.DC internal resistance of the battery was defined as the SOH.The algorithm was used in the battery health status estimation model.The results show that the maximum and mean DC internal resistance error forecasted by the proposed algorithm are 0.1 and 0.049 mΩ,respectively.It shows that the method can accurately predict the DC internal resistance of the battery and constantly reflect the SOH.