Skip to main content Skip to secondary navigation

Second-Life EV Batteries: Predicting State of Health of Retired EV Batteries using Physics-Informed Machine Learning methods

Main content start

In this research project, a novel battery modeling framework based on the enhanced single particle model (ESPM) to account for degradation mechanisms of retired electric vehicle batteries. While accounting for the transport and electrochemical phenomena in the battery solid and electrolyte phases, the dominant anode-related aging mechanisms, namely, solid electrolyte interphase (SEI) layer growth and lithium plating, are modeled. For the first time, the loss of active material (LAM), which describes the tendency of anode and cathode, over time, to reduce the electrode material available for intercalation and deintercalation, is introduced in the ESPM (Figure 1).

Model identification results show the effectiveness of the proposed modeling strategy for both fresh cell scenarios and aged conditions (Figure 2).  This model is a first step that will allow to tackle the state of health evaluation and monitoring in retired batteries. Moreover, the coupling of the aging mechanisms with the LAM dynamics provides a comprehensive means for the prediction of both linear and non-linear capacity fade trajectories, crucial to assess the health of batteries that are considered for second-life applications.

Figure 1



Figure 1. Schematic representation of the ESPM. Starting from a fresh particle, mechanical stresses due to lithium intercalation and deintercalation can induce the formation of cracks. After prolonged cycling, isolation of active material can also occur, together with SEI growth and lithium plating.

Figure 2




Figure 2.1



Figure 2. Comparison between experimental voltage (top) and SOC profiles (bottom) with ESPM predicted voltage and SOC. Fresh (#0) and aged cells (at #1000 and #3300 cycles) are considered.