Open AccessLithium-Ion Diagnostics: The First Quantitative In-Operando Technique for Diagnosing Lithium Ion Battery Degradation Modes under Load with Realistic Thermal Boundary Conditions
Author(s) -
Ryan Prosser,
Gregory J. Offer,
Yatish Patel
Publication year - 2021
Publication title -
journal of the electrochemical society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.258
H-Index - 271
eISSN - 1945-7111
pISSN - 0013-4651
DOI - 10.1149/1945-7111/abed28
Subject(s) - coolant , nuclear engineering , battery (electricity) , materials science , electrode , voltage , lithium (medication) , constant current , capacity loss , heat generation , lithium ion battery , degradation (telecommunications) , mechanics , open circuit voltage , thermal , offset (computer science) , state of charge , ion , work (physics) , current (fluid) , chemistry , electrochemistry , thermodynamics , electrical engineering , computer science , engineering , physics , medicine , power (physics) , organic chemistry , programming language , endocrinology
A diagnostic technique capable of quantitatively estimating degradation modes in-operando, including loss of lithium inventory and loss of active material, which operates under charge and discharge loads with realistic thermal boundary conditions is presented for the first time. The technique uses a zero-dimensional heat generation model with only three parameters requiring estimation, a simple heat transfer model, and requires just three temperature measurements per cell, voltage and current. The technique has been demonstrated to work for pouch cells with tab cooling and a constant coolant temperature and for charge and discharge rates of C/2, 1C and 2C. Compared to state-of-the-art open circuit voltage (OCV) model methods, the technique predicts electrode capacities and offset of a fresh cell with accuracies of 3% and 6% respectively. Further the technique has been shown to predict loss of lithium and loss of active material in the positive and negative electrodes with accuracies of 0.18%, 0.22% and 1.99% respectively. The technique can therefore provide information of the same quality as the current state-of-the-art techniques but works under application relevant conditions and due to its simplicity is suitable for implementation on-line in a battery management system (BMS).