Abstract: This paper proposes a method and system based on Kalman filtering and deep neural networks to more accurately assess the state of health (SOH) of power batteries in new energy vehicles, and thus evaluating battery performance and lifespan. By combining Kalman filtering and deep neural networks, an innovative SOH estimation framework is established to enhance the precision and robustness of the estimation results. Additionally, a flexible and scalable SOH estimation system is designed to adapt to various operating conditions. To validate the superiority and effectiveness of the proposed method, the experimental analysis are conducted, and the test results convincingly demonstrate its superiority, providing a viable technical solution for the battery management system and battery echelon utilization in new energy vehicles. By more accurately

Key words: New energy vehicles; Battery management system; Battery echelon utilization; Kalman filter; Deep neural network; SOH estimation

摘要： 文章提出了一种基于卡尔曼滤波和深度神经网络的方法和系统，以更精确地评估新能 源汽车动力电池的健康状态（SOH），从而评估电池的性能和寿命。通过融合卡尔曼滤波和深 度神经网络，建立了一个创新 SOH 估算框架，以提高估算结果的精度性和鲁棒性，此外，还 设计了一种适应不同工况且具备灵活性和可扩展性的 SOH 估算系统。为了验证提出方法的卓 越性和有效性进行了试验分析，试验结果充分证明了该方法的优越性，并为新能源汽车的电 池管理系统和电池梯次利用提供了可行的技术解决方案。通过比较精确地评估动力电池的健 康状态，能够更好地管理电池性能，并有效延长其使用寿命。

关键词: 新能源汽车；电池管理系统；电池梯次利用；卡尔曼滤波；深度神经网络；SOH 估 算