GAETS: A Graph Autoencoder Time Series Approach Towards Battery Parameter Estimation

TL;DR

GAETS employs a novel graph autoencoder approach with a non-linear NOTEARS-based structure learning to improve battery parameter estimation accuracy over existing methods.

Contribution

Introduces GAETS, a graph autoencoder time series method utilizing non-linear NOTEARS for structure learning in battery parameter estimation.

Findings

Outperforms state-of-the-art GTS architectures in battery parameter estimation.

Leverages graph neural networks to model variable dependencies effectively.

Uses gradient descent for structure learning, enhancing efficiency.

Abstract

Lithium-ion batteries are powering the ongoing transportation electrification revolution. Lithium-ion batteries possess higher energy density and favourable electrochemical properties which make it a preferable energy source for electric vehicles. Precise estimation of battery parameters (Charge capacity, voltage etc) is vital to estimate the available range in an electric vehicle. Graph-based estimation techniques enable us to understand the variable dependencies underpinning them to improve estimates. In this paper we employ Graph Neural Networks for battery parameter estimation, we introduce a unique graph autoencoder time series estimation approach. Variables in battery measurements are known to have an underlying relationship with each other in a certain correlation within variables of interest. We use graph autoencoder based on a non-linear version of NOTEARS as this allowed us to perform gradient-descent in learning the structure (instead of treating it as a combinatorial optimisation problem). The proposed architecture outperforms the state-of-the-art Graph Time Series (GTS) architecture for battery parameter estimation. We call our method GAETS (Graph AutoEncoder Time Series).