Representative battery load profiles require dynamic modeling beyond electrical power

During the development of battery electric vehicles, engineers subject battery systems to dynamic power input benchmark profiles to estimate their behavior within varying boundary conditions. However, existing approaches typically neglect the coupled dynamics of other critical stressors, particularly state-of-charge and temperature. As these variables strongly influence battery degradation, current benchmark profiles lack the ability to faithfully reproduce realistic system behavior and degradation responses. To address this limitation, this study proposes a methodology for the synthesis and evaluation of multivariate load profiles that jointly incorporate dynamic power, state-of-charge, and temperature characteristics. We assess their representativity by quantifying the replication of downstream battery reactions using industry-level electrothermal system response and cell degradation simulation models. The results show that accurate representation of voltage-related states is critical for reproducing degradation behavior. The proposed profiles achieve high-fidelity replication of statistical system states and enable up to 100 % accuracy in full cycle equivalents, SoH R , or service life at SoH C = 70 % within the boundaries of simulation. These findings highlight the importance of multivariate, dynamically coupled benchmark profiles and provide a systematic framework for developing representative test procedures with improved real-world validity.