Project Details

Description

The SuMaT-accelerator will build a state of health (SoH) and remaining useful life (RUL) model for lithium-ion batteries (LiBs) based on electrochemical impedance spectroscopy (EIS) data making use of machine learning (ML)/artificial intelligence (AI) approaches . A key stumbling block to advance LiB technologies is the unpredictability of battery degradation: accurate prediction of SoH and RUL is needed to inform the user whether a battery should be replaced and avoid unexpected capacity fade. Moreover, battery prognosis is crucial to expanding the recycling sector, enabling facilities to decide whether a battery should be recycled as scrap metal or used for less demanding “second-life” applications.

EIS is known to be a real-time, non-invasive technique, that contains rich information on all materials properties. However, deploying EIS to predictive battery diagnosis is hampered by the complex data interpretation. With ML, the possibility arises to find hidden non-trivial patterns, harvest information etc. from large amounts of data beyond the scope of the human capacities. The implementation of ML based on EIS data will allow us to identify battery parameters without pre-defined models for accurate SoH and RUL estimation, even without complete knowledge of past operating conditions of the battery.

AcronymIOFACC11
StatusActive
Effective start/end date1/10/2230/09/24

Keywords

  • lithium-ion batteries
  • state of health
  • remaining useful life
  • second life
  • machine learning

Flemish discipline codes in use since 2023

  • Battery technology

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.