Refine
Year of publication
- 2022 (1)
Document Type
- Conference Proceeding (1) (remove)
Conference Type
- Konferenzartikel (1)
Language
- English (1)
Has Fulltext
- yes (1)
Is part of the Bibliography
- yes (1)
Keywords
- grey-box model (1) (remove)
Institute
Open Access
- Diamond (1)
- Open Access (1)
Lithium-ion batteries show strongly nonlinear behaviour regarding the battery current and state of charge. Therefore, the modelling of lithium-ion batteries is complex. Combining physical and data-driven models in a grey-box model can simplify the modelling. Our focus is on using neural networks, especially neural ordinary differential equations, for grey-box modelling of lithium-ion batteries. A simple equivalent circuit model serves as a basis for the grey-box model. Unknown parameters and dependencies are then replaced by learnable parameters and neural networks. We use experimental full-cycle data and data from pulse tests of a lithium iron phosphate cell to train the model. Finally, we test the model against two dynamic load profiles: one consisting of half cycles and one dynamic load profile representing a home-storage system. The dynamic response of the battery is well captured by the model.