To assure optimal operations of the traction battery at all times and further provide reliable prediction of the remaining range, the exact determination of the state of charge (SoC) is key. The increased incorporation of cells with lithium iron phosphate (LFP) cathodes poses critical challenges for conventional SoC estimation within the battery management unit (BMU) due to the inherently pronounced voltage plateaus in combination with the open-circuit voltage (OCV) hysteresis.

The origins of the latter are manifold and can be traced back to thermodynamic-entropic effects, mechanical stress and the two-phase transition mechanism of the electrodes. Since physics-based modelling is complex and rather strenuous for online-application, this work aims to efficiently model the cell’s behavior phenomenologically based on equivalent circuit models (ECM) incorporating selected hysteresis models.

The parametrization is based on three cell tests: A galvanostatic intermittent titration technique (GITT) as well as two low-current hysteresis transition profiles. All tests are each carried out on three pristine 18650 LFP cells at 25°C. Temperature and degradation induced influences. Ultimately, the presented models aim to provide accurate predictions of the cell’s dynamic and open circuit behavior.

While all investigated hysteresis models can adequately imitate simple hysteresis transitions, the greatest accuracy and plausibility in a highly dynamic environment can be achieved with the one-state-hysteresis (OSH) model utilizing a SoC-dependent degradation parameter.