Fig. 8: SOH estimation results from the linear regression model (LRM) using charge and discharge energies, charging impedance, and resistance as features versus the aging cycle number (Cycle). | Communications Engineering

Fig. 8: SOH estimation results from the linear regression model (LRM) using charge and discharge energies, charging impedance, and resistance as features versus the aging cycle number (Cycle).

From: Domain knowledge-guided machine learning framework for state of health estimation in Lithium-ion batteries

Fig. 8

The capacity loss and estimation error profiles for cells V4 (a), W5 (b), W7 (c), and W9 (d) are shown. Augmented Capacity points (obtained as discussed in Sec. Data augmentation approach) are shown in red. Three scenarios are displayed: blue represents the LRM output trained solely with energy during charging and energy during discharging (\({{{\rm{LRM}}}}_{{E}_{{{\rm{ch}}}},{E}_{{{\rm{dis}}}}}\)); the light purple represents the LRM output trained with energy during charging, energy during discharging along with charging impedance (\({{{\rm{LRM}}}}_{{E}_{{{\rm{ch}}}},{E}_{{{\rm{dis}}}},{Z}_{{{\rm{CHG}}}}}\)); the dark purple represents the LRM output trained with energy during charging, energy during discharging, charging impedance, and resistance (\({{{\rm{LRM}}}}_{{E}_{{{\rm{ch}}}},{E}_{{{\rm{dis}}}},{Z}_{{{\rm{CHG}}}},R}\)). All models are trained exclusively using data from cell W8.

Back to article page