光纤布拉格光栅
电池(电)
估计员
卡尔曼滤波器
荷电状态
人工神经网络
拉伤
结构健康监测
扩展卡尔曼滤波器
锂离子电池
计算机科学
电子工程
工程类
光纤
人工智能
电气工程
电信
统计
数学
物理
功率(物理)
内科学
医学
量子力学
作者
Jun Peng,Shuhai Jia,Shuming Yang,Xilong Kang,Hongqiang Yu,Yaowen Yang
标识
DOI:10.1016/j.est.2022.104950
摘要
Multisensory and artificial intelligence approaches are key tools to achieve intelligent management of future battery systems. Strain monitoring using optical fiber sensors is an important role of multi-sensing in batteries. In this paper, the strain of batteries is monitored by fiber Bragg grating sensors, and the strain data are used to estimate the state of charge (SoC) and state of health (SoH) of batteries. A Kalman filtering (KF) model is proposed for SoC estimation based on strain signal of cells. Moreover, this work employs an artificial neural network (NN) for SoC estimation based on the strain data. The experimental data are acquired from commercial lithium-ion cells under two operating conditions. The KF model is established based on multiple regression between strain and SoC, which shows good performance in estimation for the static cycles. For NN estimators, input variables with strain parameter can enhance the accuracy of SoC estimation. A KF model based on the peak strain is developed to estimate the capacity degradation of battery, and the results show that strain can be used as an indicator to estimate SoH. The results present an encouraging outcome that SoC estimation can be achieved using non-electrical parameters solely, and the strain signal can also be used as an auxiliary parameter to improve the accuracy of SoC estimation. This new exploration provides a basis for multi-parameter cooperative estimation of battery state in the future battery system with a multisensory approach.
科研通智能强力驱动
Strongly Powered by AbleSci AI