Module-Based Active Equalization for Battery Packs: A Two-Layer Model Predictive Control Strategy

High-performance and safe operation of a serially connected lithium-ion battery pack in the electric vehicle necessitates effective cell equalization to maintain the state-of-charge of each cell at the same level. In this work, an improved module-based cell-to-pack-to-cell (CPC) equalization system is developed, where the module-level (ML)/cell-to-module-to-cell (CMC) equalizers are utilized for equalization of the battery modules/cells in each module. Compared with the conventional CPC balancing system, it has the advantages of simple modular structure and convenient maintainability. Then, a two-layer model predictive control (MPC) strategy is proposed, in which the ML equalizers are controlled by the top-layer MPC and the controlled CMC equalizing currents in each module are designed by the bottom-layer MPC algorithms in parallel. Its computational complexity is much less than the centralized MPC, which makes it more feasible for real-time cell equalization implementation in practical applications. A rigorous mathematical convergence proof of the proposed equalization control strategy is provided based on the Lyapunov stability theorem. Finally, extensive results are provided to verify the proposed improved module-based CPC equalization system and the two-layer MPC-based equalization control approach with excellent equalization performance being demonstrated.