State of health estimation of lithium-ion batteries under variable load profile
| Affiliation auteurs | !!!! Error affiliation !!!! |
| Titre | State of health estimation of lithium-ion batteries under variable load profile |
| Type de publication | Conference Paper |
| Year of Publication | 2017 |
| Auteurs | Li H, Ravey A, N'Diaye A, Djerdir A |
| Conference Name | IECON 2017 - 43RD ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY |
| Publisher | IEEE Ind Eect Soc; Inst Elect & Elect Engineers; Chinese Assoc Automat; Syst Engn Soc China; Chinese Power Supply Soc; Natl Nat Sci Fdn China; Chinese Acad Sci; Chinese Electrotechn Soc; Beihang Univ Sch, Reliabil & Syst Engn; RMIT Univ; Beijing JiaoTong |
| Conference Location | 345 E 47TH ST, NEW YORK, NY 10017 USA |
| ISBN Number | 978-1-5386-1127-2 |
| Résumé | The accurate estimation of state of health (SOH) of Lithium-ion battery is critical to guarantee safety and reliability of whole system. Large number of researches on SOH have many strict constrained assumptions, such as repeated same load, which differ from real situation. In this paper, the study focus on variable load profile. An artificial neural network is designed to determine the battery SOH using operation times for 12 different currents and the beginning capacity for every cycle. Two kinds of validation samples are chosen to test the neural network. Simulation results show that for randomly chosen 20 validation samples from whole dataset, each sample errors don't exceed 4.37% and root mean squared error is 1.33%, for 20 sequential validation samples from one battery, maximum estimation error is 4.96% and root mean squared error is 1.25%, all below 5%, proving the good behavior of the designed neural network SOH estimation. |