A Temporal Convolution Network Approach to State-of-Charge Estimation in Li-ion Batteries

TL;DR

This paper introduces a novel application of Temporal Convolution Networks for accurately estimating the State of Charge in Lithium-ion batteries used in electric vehicles, achieving high accuracy across various drive cycles.

Contribution

First implementation of TCNs for SOC estimation, demonstrating superior accuracy and robustness in EV battery management.

Findings

Achieved 99.1% accuracy in SOC estimation.

Effective across multiple drive cycles and conditions.

First use of TCNs in this domain.

Abstract

Electric Vehicle (EV) fleets have dramatically expanded over the past several years. There has been significant increase in interest to electrify all modes of transportation. EVs are primarily powered by Energy Storage Systems such as Lithium-ion Battery packs. Total battery pack capacity translates to the available range in an EV. State of Charge (SOC) is the ratio of available battery capacity to total capacity and is expressed in percentages. It is crucial to accurately estimate SOC to determine the available range in an EV while it is in use. In this paper, a Temporal Convolution Network (TCN) approach is taken to estimate SOC. This is the first implementation of TCNs for the SOC estimation task. Estimation is carried out on various drive cycles such as HWFET, LA92, UDDS and US06 drive cycles at 1 C and 25 {\deg}Celsius. It was found that TCN architecture achieved an accuracy of 99.1%.