| About this Abstract |
| Meeting |
2022 TMS Annual Meeting & Exhibition
|
| Symposium
|
Advanced Materials for Energy Conversion and Storage 2022
|
| Presentation Title |
Use of Machine Learning Methods to Predict Remaining Useful Life of Lithium-ion Batteries after Experiencing Non-catastrophic Nail Puncture |
| Author(s) |
Casey M. Jones, Meghana Sudarshan, Alex Serov, Vikas Tomar |
| On-Site Speaker (Planned) |
Casey M. Jones |
| Abstract Scope |
The purpose of this work was to simulate the operation of a Lithium-ion cell in an abusive environment, such as those found in electric vehicles, aerospace applications, etc., and use the collected data to predict the rate of degradation and remaining useful life with different machine learning methods. A nail puncture test was performed with a depth approximately halfway through each cell during cycling at a rate of 1C, and the cells were allowed to continue cycling afterwards. The penetrations caused a rapid spike in temperature, resulting in decomposition of the electrolyte and solid electrolyte interface. Combined with the physical damage to the electrodes, this generated an accelerated rate of aging in the cells under test. The resulting data on cycle number, capacity, temperature, and other factors was used to develop machine learning algorithms that were implemented in order to predict the remaining useful life of the cells after puncture. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Energy Conversion and Storage, Other, Other |