Digital Twin Simulation for Electric Vehicle Battery Management

This technical report presents a digital twin approach for managing lithium-ion batteries in electric vehicles. It outlines the integration of complex multiphysics models with reduced order models (ROM) to enhance battery management systems (BMS). The report details the development of a digital twin simulation utilizing a dataset generated from COMSOL® Multiphysics, focusing on the Cahn-Hilliard equation for a single particle model of lithium iron phosphate (LiFePO4) cathodes. The methodology includes machine learning techniques to derive latent variables and establish statistical relationships between the simulation data and battery performance metrics such as voltage and temperature. The report discusses the validation of the ROM as an on-board BMS, capable of real-time operation under varying ambient conditions. It also highlights the necessity of reduced-order models to manage the complexities of battery dynamics while ensuring efficient data processing for real-time applications. The findings contribute to the ongoing research aimed at improving battery management for electric and hybrid vehicles.