Review on Equivalent Circuit and Data driven approaches for Electric Vehicle Battery Models

Batteries play a crucial role in electric vehicle development (EV).In EV batteries real-time state estimation is challenging and hence is approximated indirectly. In this paper, a comparative review of state estimation techniques and battery models is conducted. The Equivalent Circuit Model (ECM) is found more appropriate since it achieves a balance between computing efficiency and accuracy. While chemistry-based models are more accurate than ECMs, however, for practical uses a trade-off is essential. In this paper the data-driven state estimation approaches are also reviewed. In depth analysis of dynamical equations are not required in data driven models, but are better at capturing non-linearities, are adaptable, and can be generalized to different case scenarios. Data-driven techniques are also less sensitive to parameter uncertainty and are capable of learning from data. This paper concludes by emphasizing the need for trade-off between real-time applicability and accuracy by focusing on the ECMs and data driven models.