Single value decomposition for sparse temperature sensing and state observation in multicell battery packs
[EN] Adequate cell temperature estimation in lithium-ion batteries becomes crucial for state of charge (SOC) observation and safety purposes. The diagnosis of individual cell temperature in multicell battery packs depends on the number of temperature sensors available and the thermal dynamics of the...
| Autores: | , , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2024 |
| País: | España |
| Institución: | Universitat Politècnica de València (UPV) |
| Repositorio: | RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia |
| Idioma: | inglés |
| OAI Identifier: | oai:riunet.upv.es:10251/228835 |
| Acceso en línea: | https://riunet.upv.es/handle/10251/228835 |
| Access Level: | acceso embargado |
| Palabra clave: | SVD Thermal distribution Battery packs Kalman filter |
| Sumario: | [EN] Adequate cell temperature estimation in lithium-ion batteries becomes crucial for state of charge (SOC) observation and safety purposes. The diagnosis of individual cell temperature in multicell battery packs depends on the number of temperature sensors available and the thermal dynamics of the system. This paper explores the potential of single value decomposition (SVD) of thermal distribution on battery packs in order to retain the critical information and minimize the number of states, but also the number of sensors required. The algorithm proposes a thermal lumped model to identify the thermal dynamics of the pack under different control actions and atmospheric conditions, and uses a Kalman filter to update the model states with temperature readings. Experimental tests were carried out to in a prototype with 20 cylindrical 21700 Li-Ion cell, equipped with several thermocouples and recording thermal images every 5 s. Results show that combining SVD with dynamic models and observers, the temperature distribution of the pack can be predicted with negligible errors. |
|---|