Advanced Lithium-ion battery models for electric vehicles: Assessing the optimal order reduction technique

Physics-based models could greatly improve lithium-ion battery management system’s ability to estimate cell states. Integrating these models into battery embedded hardware can potentially extend the performance and lifespan of battery cells. Nevertheless, it is crucial to simplify this model to lessen the computational workload for a successful implementation in embedded microprocessors. This article uses the analytic hierarchy process to select the most suitable reduced-order technique for advanced models running in microcontrollers. The model’s implementation feasibility is assessed based on weighted criteria significant for control-oriented applications, such as accuracy, execution time, required memory, observability, and controllability. The application of the process and the corresponding boundary criteria are exemplified in different evaluation scenarios, representative of the electrical vehicle market. The successful results of the proposed process validate its scalability and reproducibility in similar decision-making scenarios. This approach not only advances the integration of advanced physics-based models into lithium-ion battery management systems but also broadens the applicability of decision-making tools in control field. • Primarily validate that reduced-order electrochemical models have the potential to fit into electrical vehicle embedded battery hardware. • Demonstrate that decision-making techniques significantly assist in the complex integration electrochemical models into battery management systems.