Fisher identifiability analysis for a periodically-excited equivalent-circuit lithium-ion battery model
Fisher identifiability analysis for a periodically-excited equivalent-circuit lithium-ion battery model
复制标题
周期性激励等效电路锂离子电池模型的Fisher辨识性分析
DOI：
10.1109/acc.2014.6859360
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
H. Fathy
中科院分区：
文献类型：
--
作者：
Aabhas Sharma;H. Fathy
本文利用 Fisher 信息量化了锂离子电池周期性循环的一阶非线性等效电路模型的内阻和充电容量的可识别性。该论文推导了方差的解析 Cramér-Rao 界，最大似然估计器可以在存在白噪声和高斯电压测量噪声的情况下确定这些参数。该数学分析表明，即使对于简单得多的等效电路模型，电池参数可识别性的挑战也从根本上存在，这一挑战在高阶电化学电池模型的文献中已经得到认可。该分析还量化了电池开路电压相对于充电状态的敏感性影响参数可识别性的程度。该论文对通过定期循环测试确定两个电池健康状态指标（即功率和容量衰减）的准确性进行了初步分析。
This paper uses Fisher information to quantify the identifiability of internal resistance and charge capacity for a first-order nonlinear equivalent-circuit model of a lithium-ion battery undergoing periodic cycling. The paper derives analytic Cramér-Rao bounds on the variances with which a maximum-likelihood estimator can determine these parameters in the presence of white and Gaussian voltage measurement noise. This mathematical analysis shows that the challenge of battery parameter identifiability, already recognized in the literature for higher-order electrochemical battery models, is fundamentally present even for much simpler equivalent circuit models. The analysis also quantifies the degree to which the sensitivity of battery open-circuit voltage with respect to state of charge affects parameter identifiability. The paper serves as a first-cut analysis of the accuracy with which one can determine two battery state-of-health metrics - namely, power and capacity fade - from periodic cycling tests.
影响因子：
--
作者：
JACQUEZ, JA;PERRY, T
通讯作者：
PERRY, T