Nonlinear Model-Predictive Optimal Control of an Active Cell-to-Cell Lithium-Ion Battery Pack Balancing Circuit

This paper presents a model-predictive framework for the optimal control of active cell-to-cell balancing in a lithium-ion battery pack. The framework addresses the tradeoff between minimizing (i) the state of charge (SOC) imbalance between battery cells and (ii) the energy dissipated by balancing. There is a rich existing literature on the design and control of pack balancing circuits. However, to the best of the authors’ knowledge, the use of model-predictive optimal control for multi-objective active balancing remains relatively unexplored. We solve the balancing problem using nonlinear model predictive control (NMPC), and exploit the differential flatness of battery cell dynamics to implement NMPC efficiently using pseudo-spectral optimization. A simulation study shows the effectiveness of the proposed approach for a two-cell balancing problem, but the approach can be generalized to longer battery strings.