Battery Ageing Prediction and Optimization in a Fleet of Electric Autonomous Vehicles

Forskningsprojekt, 2021 – 2023

Electric vehicles (EVs) with large battery packs are predicted to constitute more than a half of the total passenger fleet by 2050.  This market evolution towards the large-scale introduction of EVs sets outstanding requirements for the battery life (to be equivalent to vehicle life) and the battery costs (to be at the same level as conventional propulsion systems). Performance degradation and battery aging are the biggest uncertainties of the total lifetime for traction batteries. Thus, it is imperative to  prolong battery service life for promoting EV applications. Moreover, the advent of connected and autonomous vehicles (AVs) can make possible the real-time eco-driving control of EVs, which is believed to be a game changer in extending battery life and increasing energy efficiency.

 Identifying and predicting battery ageing remains challenging as the internal mechanisms are interconnected and further influenced by environmental conditions (e.g. climate) and the utilization modes (e.g. successive accelerations/decelerations and constant velocities). With capacity fade, the available energy decreases hence impacting the electric driving range; with internal resistance increases, the available power decreases and so do the capabilities for accelerating and energy recuperation (e.g. during regenerative braking).

 In the present traffic system studies, battery models have been overly simplified, particularly for the aging part, rendering premature battery aging and low energy efficiency. This is mainly due of the computational difficulty in consolidating fleet speed/trajectory control in real-time (often with significant stochasticity) and battery systems (highly nonlinear in nature). 

The research activities are divided into two main parts. The first part concerns the development of control-oriented battery system models that accurately capture the battery's aging behaviour and are suitable for battery management systems in AEVs. The second part will focus on the design of intelligent control strategies for an AEV fleet.