Mitigation of intra-cycle mechano-chemical degradation-based capacity fade in Lithium-ion batteries: Application of a model predictive controller

Abstract

Electric vehicles (EVs) require a sophistical battery management system (BMS) to track system temperature, stress, and battery health. Also, control of these batteries is important for minimizing the capacity fade. Unfortunately, previously developed battery models like single-particle model (SPM), and pseudo-two-dimensional (P2D) have certain disadvantages making it difficult to employ them in BMS. For example, they show deviations from the experimental observations, and are not computationally efficient. To address this problem, a new battery model is constructed by integrating an existing SPM with the various mechano-chemical degradation mechanisms to accurately predict dynamic intra-cycle capacity fade. Further, to manipulate the applied current to minimize the capacity fade during the charging of a battery, the developed model is combined with a model predictive control (MPC) framework. In conclusion, the developed model can track capacity fade, temperature, and mechanical stress in Lithium-ion batteries (LIBs), and has been experimentally validated. Finally, the MPC shows a superior performance in battery health regulation than the existing constant current-constant voltage (CC-CV) charging protocol.