基于DP模型的锂离子电池能量状态估算  

Estimation of state of energy of lithium-ion batteries based on DP model

作　　者：尹乐乐 靳成杰 康健强 谭祖宪 秦鹏 YIN Le-le;JIN Cheng-jie;KANG Jian-qiang;TAN Zu-xian;QIN Peng(School of Automotive Engineering,Wuhan University of Technology,Wuhan Hubei 430070,China;Shenzhen Pengcheng New Energy Technology Co.,Ltd.,Shenzhen Guangdong 518102,China)

机构地区：[1]武汉理工大学汽车工程学院,湖北武汉430070 [2]深圳市鹏诚新能源科技有限公司,广东深圳518102

出　　处：《电源技术》2019年第10期1619-1622,共4页Chinese Journal of Power Sources

摘　　要：估算能量状态是电池管理系统的主要功能之一,因为对于电动汽车而言能量状态是预测续航里程、能量管理分配和优化以及实现电池组均衡的的重要参数。传统的功率积分方法,其准确性依赖于较高精度的电压、电流传感器,因而成本高。因此,基于改进的戴维南电路模型,将扩展卡尔曼滤波法(EKF)用来估算电池的剩余能量状态和荷电状态,且使用遗忘递推最小二乘法在线实时辨识模型参数。结果表明,此方法具有较好的估算精度,在复杂动态电流测试工况估算误差可以保持在2%以内,而且能量状态(SOE)比荷电状态(SOC)更适合反映能量的变化。Estimation of state of energy is one of the main functions of battery management system,because for electric vehicles,state of energy(SOE)is an important parameter to predict the cruising range,allocate and optimize energy management and achieve battery pack equalization.The accuracy of the conventional power integral method mainly relies on a higher precision of voltage and current sensors,thereby increasing the cost.In this study,based on the improved Thevenin circuit model,the EKF method was used to estimate the residual energy state and state of charge(SOC)of the battery,whereas the model parameters were simultaneously identified online by using the forgetting recursive least square method(FRLS).The results show that this method has good estimation accuracy with estimation error less than 2%under complex dynamic current test conditions.In addition,it is found that SOE is more suitable for reflecting energy changes than SOC.

关 键 词：锂离子电池 卡尔曼滤波 在线辨识 电池模型 能量状态 荷电状态 

分 类 号：TM912[电气工程—电力电子与电力传动]