Operando impedance-based battery cell internal temperature estimation under non-stationarity and non-linearity conditions

Electrochemical impedance spectroscopy, a method for battery diagnostics, is used to estimate the internal temperature of a lithium-ion battery cell during highly dynamic load profiles. For the first time, a recurrent neural network is trained and evaluated with operando impedance data for temperature estimation. Furthermore, an approach is considered that guides the training process of the neural network by incorporating physical constraints. The model’s development based on an extensive series of measurements with different load profiles, tested under realistic conditions on large-format lithium-ion cells. The estimation accuracy of the data-driven approach is evaluated and compared against model-based methods, including the extended Kalman filter. An impedance correction model is proposed, which leads to a significant enhancement of the model-based estimation. The recurrent neural network under consideration achieves a mean square error of 1.07 ° C for the investigated testing profiles in the temperature range up to 60 ° C . • Application of recurrent neural network with operando impedance data. • Impedance-based cell internal temperature estimation based on extended Kalman filter. • Comparison of model-based and data-based approaches. • 406 measurements during dynamic load profiles for model training and validation. • Evaluation of a physics-based approach for neural network training process guidance.