Degradation Mode Analysis (DMA) is increasing in popularity as a tool for diagnosing the state-of-health (SOH) of a cell in more detail than capacity-based methods. Assuming degradation modes can define the cell state, error quantification for DMA methods is usually done with datasets generated from known degradation mode combinations [1, 2].

This work instead uses a PyBaMM [3] model under different conditions to generate a synthetic dataset on which the error in degradation modes can be quantified by comparison with internal states of the model. This approach allows the performance of different DMA methods, implemented in PyProBE [4], to be quantified at different cell SOH, reached as the result of different dominant degradation mechanisms and using different reference performance test (RPT) procedures. Unlike the previously mentioned validation data for DMA, the degradation modes are not an input to the model. Furthermore, using a physics-based model allows the effects of degradation on the dynamic and kinetic processes in the cell to influence the synthetic dataset, so that their impact on the accuracy of DMA can be quantified. This work therefore provides an insight into the magnitude and sources of error in DMA to researchers and industry who might use these methods on real cells where the true values are impossible to quantify in-situ.

References:
1. M. Dubarry, D. Beck, Journal of Power Sources 479 (2020), 228806
2. C.R. Birkl, M.R. Roberts, E. McTurk, P.G. Bruce, D.A. Howey. Journal of Power Sources 341 (2017), 373–386
3. V. Sulzer, S.G. Marquis, R. Timms, M. Robinson, S.J. Chapman, Journal of Open Research Software 9 (2021), 1, 14-21
4. T. Holland, D. Cummins, M. Marinescu. PyProBE: Python Processing for Battery Experiments. Journal of Open Source Software 10 (2025), 106