TY - JOUR
T1 - Data driven battery modeling and management method with aging phenomenon considered
AU - Li, Shuangqi
AU - He, Hongwen
AU - Su, Chang
AU - Zhao, Pengfei
N1 - Publisher Copyright:
© 2020
PY - 2020/10/1
Y1 - 2020/10/1
N2 - The battery is one of the most important parts of electric vehicles (EVs), and the establishment of an accurate battery state estimation model is of great significance to improve the management strategy of EVs. However, the battery degrades with the operation of EVs, which brings great difficulties for the battery modeling issue. This paper proposes a novel aging phenomenon considered vehicle battery modeling method by utilizing the cloud battery data. First of all, based on the Rain-flow cycle counting (RCC) algorithm, a battery aging trajectory extraction method is developed to quantify the battery degradation phenomenon and generate the aging index for the cloud battery data. Then, the deep learning algorithm is employed to mine the aging features of the battery, and based on the mined aging features, an aging phenomenon considered battery model is established. The actual operation data of electric buses in Zhengzhou is used to validate the practical performance of the proposed methodologies. The results show that the proposed modeling method can simulate the characteristic of the battery accurately. The terminal voltage and SoC estimation error can be limited within 2.17% and 1.08%, respectively.
AB - The battery is one of the most important parts of electric vehicles (EVs), and the establishment of an accurate battery state estimation model is of great significance to improve the management strategy of EVs. However, the battery degrades with the operation of EVs, which brings great difficulties for the battery modeling issue. This paper proposes a novel aging phenomenon considered vehicle battery modeling method by utilizing the cloud battery data. First of all, based on the Rain-flow cycle counting (RCC) algorithm, a battery aging trajectory extraction method is developed to quantify the battery degradation phenomenon and generate the aging index for the cloud battery data. Then, the deep learning algorithm is employed to mine the aging features of the battery, and based on the mined aging features, an aging phenomenon considered battery model is established. The actual operation data of electric buses in Zhengzhou is used to validate the practical performance of the proposed methodologies. The results show that the proposed modeling method can simulate the characteristic of the battery accurately. The terminal voltage and SoC estimation error can be limited within 2.17% and 1.08%, respectively.
KW - Aging-considered battery model
KW - Battery degradation quantification
KW - Battery energy storage
KW - Battery management system
KW - Deep learning
KW - Electric vehicles
UR - http://www.scopus.com/inward/record.url?scp=85086585158&partnerID=8YFLogxK
U2 - 10.1016/j.apenergy.2020.115340
DO - 10.1016/j.apenergy.2020.115340
M3 - Article
AN - SCOPUS:85086585158
SN - 0306-2619
VL - 275
JO - Applied Energy
JF - Applied Energy
M1 - 115340
ER -
