采用极限学习机实现锂离子电池健康状态在线估算

摘要

Battery's state of health(SOH) estimation is one of the core technologies in battery management system of electric vehicles. In order to accurately estimate lithium-ion battery's SOH online,health indicator(HI), characterizing battery degradation,is proposed and constructed under dynamic conditions,and the extreme learning machine (ELM) is introduced to offline train the ELM degradation model for the whole life cycle of battery and ful-fill the online estimation of SOH. The results of experiments show that the method adopted can accurately estimate the SOH of lithium-ion battery with an estimation error less than 2%.%电池健康状态(SOH)估算是电动汽车电池管理系统核心技术之一.为准确在线估算锂离子电池SOH,提出在动态工况下构建表征电池衰退的健康指标(HI),并引入极限学习机(ELM)离线训练电池全生命周期的ELM衰退模型,实现SOH在线估算.实验结果表明,该方法能准确在线估算锂离子电池SOH,估算误差不超过2%.