Sodium-ion batteries (SIBs) constitute an outstanding development in energy storage technology due to their affordability and abundant sodium availability. To enhance the modelling of these batteries, the problem of mass transport in solid-phase spherical electrode particles is considered which is based on Fick’s second law and is represented as a partial differential equation (PDE). The modelled PDE is transformed into a dimensionless form by applying appropriate dimensionless variables. The obtained dimensionless PDE is solved using Laplace based Hermite Collocation Method (LT-HCM) and the results are compared with finite difference method (FDM). A dataset has been generated using the LT-HCM for parameters of batteries. Furthermore, time series analysis is conducted to predict the state of charge (SOC) using long short-term memory (LSTM) networks and other machine learning (ML) models. The predictions are validated against the LT-HCM, demonstrating the efficacy of LSTM in accurately forecasting SOC. Additionally, the effect of dimensionless physical parameters has been examined and discussed in detail. Therefore, this work enhances the accuracy of SOC estimation by integrating mathematical and data driven based approach while offering efficient techniques for optimizing and modelling electrochemical energy conversion and storage devices.
Shivaranjini et al. (Fri,) studied this question.