LSTM for State of Charge Estimation of Lithium Polymer Battery on Jetson Nano

Hantono, Bimo Sunarfri and Cahyadi, Adha Imam and Putu Pratama, Gilang Nugraha (2021) LSTM for State of Charge Estimation of Lithium Polymer Battery on Jetson Nano. In: International Conference on Information Technology and Electrical Engineering (ICITEE).

[thumbnail of LSTM_for_State_of_Charge_Estimation_of_Lithium_Polymer_Battery_on_Jetson_Nano.pdf] Text
LSTM_for_State_of_Charge_Estimation_of_Lithium_Polymer_Battery_on_Jetson_Nano.pdf
Restricted to Registered users only

Download (778kB)

Abstract

Battery Management System (BMS) is an important element for batteries. It can protect the batteries from operating in dangerous conditions. BMS consists of many items. One of them is State of Charge (SoC), which indicates the batteries' charge level. Currently, we are unable to measure the internal states of the battery directly. Therefore, in order to obtain the information of SoC, we need to estimate it. This paper proposes Long Short-Term Memory (LSTM) to estimate SoC of lithium polymer battery, where NVIDIA Jetson Nano handles the computation. Based on the experiments, we have RMSE scores of 1.797 and 1.976, respectively, for training and testing. Those results show that it is possible to employ LSTM on NVIDIA Jetson Nano for estimating the SoC of lithium polymer battery. © 2021 IEEE.

Item Type: Conference or Workshop Item (Paper)
Additional Information: Cited by: 4
Uncontrolled Keywords: Battery management systems; Brain; Charging (batteries); Lithium; Lithium-ion batteries; Battery M anagement system; Condition; Internal state; Jetson nano; Lithium polymer; State-of-charge estimation; States of charges; Training and testing; Long short-term memory
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Faculty of Engineering > Electrical and Information Technology Department
Depositing User: Sri JUNANDI
Date Deposited: 25 Oct 2024 06:32
Last Modified: 25 Oct 2024 06:32
URI: https://ir.lib.ugm.ac.id/id/eprint/8652

Actions (login required)

View Item
View Item