Health indicator selection for state of health estimation of second-life lithium-ion batteries under extended ageing
Nowadays, the economic viability of second-life (SL) Li-ion batteries from electric vehicles is still uncertain. Degradation assessment optimization is key to reduce costs in SL market not only at the repurposing stage, but also during SL lifetime. As an indicator of the ageing condition of the batt...
| Autores: | , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2022 |
| País: | España |
| Institución: | Universidad Pública de Navarra |
| Repositorio: | Academica-e. Repositorio Institucional de la Universidad Pública de Navarra |
| OAI Identifier: | oai:academica-e.unavarra.es:2454/44660 |
| Acceso en línea: | https://hdl.handle.net/2454/44660 |
| Access Level: | acceso abierto |
| Palabra clave: | Ageing Health indicator Lithium-ion batteries Second-life batteries State of health estimation |
| Sumario: | Nowadays, the economic viability of second-life (SL) Li-ion batteries from electric vehicles is still uncertain. Degradation assessment optimization is key to reduce costs in SL market not only at the repurposing stage, but also during SL lifetime. As an indicator of the ageing condition of the batteries, state of health (SOH) is currently a major research topic, and its estimation has emerged as an alternative to traditional characterization tests. In an initial stage, all SOH estimation methods require the extraction of health indicators (HIs), which influence algorithm complexity and on-board implementation. Nevertheless, a literature gap has been identified in the assessment of HIs for reused Li-ion batteries. This contribution targets this issue by analysing 58 HIs obtained from incremental capacity analysis, partial charging, constant current and constant voltage stage, and internal resistance. Six Nissan Leaf SL modules were aged under extended cycling testing, covering a SOH range from 71.2 % to 24.4 %. Results show that the best HI at the repurposing stage was obtained through incremental capacity analysis, with 0.2 % of RMSE. During all SL use, partial charge is found to be the best method, with less than 2.0 % of RMSE. SOH is also estimated using the best HI and different algorithms. Linear regression is found to overcome more complex options with similar estimation accuracy and significantly lower computation times. Hence, the importance of analysing and selecting a good SL HI is highlighted, given that this made it possible to obtain accurate SOH estimation results with a simple algorithm. |
|---|