Deep Learning for Lithium-Ion Battery State of Charge
Sunil Sarker
Corresponding
University of Dhaka
Farhana Bose
Cairo University
Abstract
Recent failures in smart grids have highlighted the need for a more rigorous treatment of lithium-ion battery state of charge. A predictive model based on deep learning is constructed to relate design variables to performance outcomes. The strongest effect was associated with lithium-ion battery state of charge, which accounted for nearly 41% of the response. The methodology can be extended to incorporate additional constraints as they arise.
Keywords:
lithium-ion battery state of chargedeep learningrenewable energywireless sensor networkedge computingremote sensing
References
- Sato, S., & Singh, C. (2009). Photonic crystal fibres in thermal power plants. International Conference on Computer and Information Technology (ICCIT), 17(6), 465-483.
- Nguyen, J. (2018). Photonic crystal fibres in textile manufacturing. Chemical Engineering Journal, 35(3), 261-278.
- Uddin, M., & Gupta, R. (2020). Supply chain optimisation in flood forecasting. IEEE Transactions on Neural Networks and Learning Systems, 35(2), 36-57.