Dynamic Battery Type Detection Using Neural Networks

Dynamically detecting battery chemistries, including LiFePO4, Ni-MH, and Lead Acid, is explored through extensive simulations. Utilizing discharge curves as training data, three neural network architectures-Single Hidden Layer, Double Hidden Layer, and Radial Basis Transfer Function-are employed for...

Full description

Saved in:
Bibliographic Details
Published in2023 IEEE 20th International Conference on Smart Communities: Improving Quality of Life using AI, Robotics and IoT (HONET) pp. 1 - 5
Main Authors Lopez, Hector K., Zilouchian, Ali, Abtahi, Amir
Format Conference Proceeding
LanguageEnglish
Published IEEE 04.12.2023
Subjects
Adaptation models
Artificial neural networks
Chemistry
Discharges (electric)
Energy Storage
Lead
Neural Networks
Technological innovation
Training data
Vehicle-to-Grid Battery Modeling
Online AccessGet full text

Cover

More Information
Summary:Dynamically detecting battery chemistries, including LiFePO4, Ni-MH, and Lead Acid, is explored through extensive simulations. Utilizing discharge curves as training data, three neural network architectures-Single Hidden Layer, Double Hidden Layer, and Radial Basis Transfer Function-are employed for pattern recognition across diverse discharge profiles. The objective is to enable the identification of connected battery types and optimize charging control. This research holds significance in real-time Electric Vehicle (EV) charging optimization, offering the capability to discern various battery chemistries. Additionally, in Peer-to-Peer (P2P) energy markets, the dynamic contribution of batteries to the grid requires safe and efficient charging for interoperable systems. The findings presented in this study introduce adaptable systems, fostering innovation in sustainable energy practices.
ISSN:1949-4106
DOI:10.1109/HONET59747.2023.10374630