Although mass manufacture has made lithium-ion batteries cheaper, cost and durability remain obstacles to the widespread adoption of battery electrical vehicles. The lifetime of the batteries falls well below the consumer expectation for long-term applications such as transport. The automotive industry wants to better understand the causes and mechanisms of degradation to enable improved control and prediction of the state of health of battery systems.
Degradation mechanisms can occur on length-scales from the nano to the macroscopic, and timescales from seconds up to years. A full understanding of the causes and effects of degradation of lithium-ion batteries for automotive applications therefore requires synergistic investigation across these length and time scales and with the combination of many experimental techniques.
This project has created a cross-disciplinary consortium of researchers and industry partners to develop a comprehensive mechanistic understanding of the relationship between external stimuli (such as temperature and cycling rate) and the physical and chemical processes occurring inside the battery that lead to degradation.
Forecasting the state of health and remaining useful life of Li-ion batteries is an unsolved challenge that limits technologies such as consumer electronics and electric vehicles.
The Degradation team has built an accurate battery forecasting system by combining electrochemical impedance spectroscopy (EIS)—a real-time, non-invasive, and information-rich measurement that is previously underused in battery diagnosis —with Gaussian process machine learning.
Over 20,000 EIS spectra of commercial Li-ion batteries were collected at different states of health, states of charge, and temperatures —the largest dataset to our knowledge of its kind.
The Gaussian process model takes the entire spectrum as input, without further feature engineering, and automatically determines which spectral features predict degradation.
This model accurately predicts the remaining useful life, even without complete knowledge of past operating conditions of the battery.
The results demonstrate the value of EIS signals in battery management systems. This has multiple potential benefits, such as improving the economics of grid-based storage, assessing the length of second life batteries, and measuring the wear and tear on electric vehicles.
1 March 2018 – 30 June 2021
Professor Clare Grey, University of Cambridge
Dr Rhodri Jervis, University College London
Dr David Hall, University of Cambridge
University of Cambridge (Lead)
University College London
Imperial College London
University of Liverpool
University of Manchester
University of Oxford
University of Sheffield
University of Southampton
University of Warwick
Research Organisations, Facilities and Institutes
National Physical Laboratory (NPL)
+ 8 Industrial Partners