Relaxation time effect on NMC811 equivalent circuit models for marine duty cycle simulation

Pulse experiments such as the galvanostatic intermittent titration technique (GITT) are commonly used to parameterize lithium-ion batteries (LIBs) in physics-based models, but have recently been adapted to parameterize equivalent circuit models (ECMs). This study investigates the impact of GITT relaxation time on the accuracy of ECMs for LIBs in marine duty cycles, using nickel-rich LiNi 0.8 Mn 0.1 Co 0.1 O 2 (NMC811) with Li x W y O z coating (LWO-NMC811) as a cathode material. GITT was performed with varying relaxation times to collect state-of-charge vs open-circuit voltage (SOC-OCV) curves and ECM parameters. The results show that longer relaxation times improve the accuracy of SOC-OCV curves and ECM parameter fit. However, shorter relaxation times may better replicate real-world battery discharge cycles, so subsequent model validation was performed. Validation experiments using constant current discharge and ship cycle profiles demonstrate that ECMs built with longer relaxation times have lower root mean square error, indicating better accuracy. However, 1-hour relaxation time experiments provide a balanced approach, with an average error difference of 7.54 mV and 80% less experiment time versus 5-hour relaxation time experiments for cells cycled within 90% to 10% SOC. This study suggests potential for optimizing relaxation times based on specific application needs and highlights the importance of supplementing continuous efforts to improve NMC811 cycling stability in high-voltage windows with parallel modeling studies. • Collection of LWO-NMC811 ECM parameters in high-voltage window, 1% SOC intervals. • 5 galvanostatic intermittent titration technique (GITT) relaxation times tested. • GITT relaxation time sublinear effect on ECM voltage prediction accuracy. • ECMs validated using both constant current and ship cycle profiles.