Battery Voltage Estimation Using NARX Recurrent Neural Network Model
Adrian Chmielewski , Jakub Możaryn , Piotr Piórkowski , Krzysztof Jakub Bogdziński
AbstractThis work presents a prediction of battery terminal voltage in subsequent charging/discharging cycles. To estimate chosen signals the NARX (Auto-Regressive with eXogenous input) model based on Recurrent Neural Net-work has been employed. A training and testing data were gathered at the la-boratory test stand with the Lithium Iron Phosphate (LiFePO4) battery in dif-ferent working conditions. Test stand research was conducted for 40 charg-ing/discharging cycles. Furthermore, the paper presents the results of the identification of double RC model parameters for a specified state of charge level. As a result, the analysis of the proposed methodology has been dis-cussed.
|Publication size in sheets||0.65|
|Book||Szewczyk Roman, Zieliński Cezary, Kaliczyńska Małgorzata (eds.): Automation 2019: Progress in Automation, Robotics and Measurement Techniques, Advances in Intelligent Systems and Computing, vol. 920, 2020, Cham, Switzerland, Springer, ISBN 978-3-030-13272-9, [978-3-030-13273-6], 727 p., DOI:10.1007/978-3-030-13273-6|
|Keywords in English||Artificial Neural Network, LiFePO4 battery, experimental research, Recur-rent Neural Networks.|
|License||Publisher website (books and chapters only); author's final; ; after publication|
|Score||= 20.0, 07-07-2020, ChapterFromConference|
|Publication indicators||= 1|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.