Optimizing Cycle Life Prediction of Lithium-ion Batteries via a Physics-Informed Model
Abstract: Accurately measuring the cycle lifetime of commercial lithium-ion batteries is crucial for performance and technology development. We introduce a novel hybrid approach combining a physics-based equation with a self-attention model to predict the cycle lifetimes of commercial lithium iron phosphate graphite cells via early-cycle data. After fitting capacity loss curves to this physics-based equation, we then use a self-attention layer to reconstruct entire battery capacity loss curves. Our model exhibits comparable performances to existing models while predicting more information: the entire capacity loss curve instead of cycle life. This provides more robustness and interpretability: our model does not need to be retrained for a different notion of end-of-life and is backed by physical intuition.
Submission Length: Regular submission (no more than 12 pages of main content)
Changes Since Last Submission: We have included, in now Appendix A, a complete list of fitted vs true capacity loss curves in response to Reviewer FBYN’s comments that there might be biases in the fitted capacity loss curves. In addition, we have also included a complete distribution of the $R^2$s for each fitted curve to provide evidence that supports the claim that the capacity fade curve is recovered. Thanks again all for detailed feedback.
Code: https://github.com/nathan99sun/HybridPred
Assigned Action Editor: ~Cedric_Archambeau1
Submission Number: 3034
Loading