Leveraging Adaptive Model Predictive Controller for Active Cell Balancing in Li-ion Battery

TL;DR

This paper introduces an adaptive model predictive control method for active cell balancing in Li-ion batteries, improving voltage uniformity and potentially extending battery life in electric vehicles.

Contribution

It proposes a novel adaptive MPC approach utilizing RLS for predictive balancing of battery cells, enhancing charge equalization in series-connected Li-ion batteries.

Findings

Voltage uniformity among cells improved in simulations

The proposed method achieves more efficient balancing

Battery lifespan could be extended through better charge management

Abstract

Automotive industry is moving toward fully electric and hybrid electric vehicles. Accordingly, energy storage unit is one of the most important blocks in these electric drives. Battery stacks which contain a number of cells are being used for supplying the vehicles' energy. Charge equalization for series connected battery strings has a significant effect on battery life. In this paper, an adaptive model predictive controller (AMPC) is proposed to manage the cell equalizing process. The series connected cells' voltages and currents are collected, then leveraging Recursive Least Square (RLS) method, the future voltage samples for all of the cells are predicted. MPC controller specifies a sequence which results in the optimum balancing performance of the proposed circuit. Simulation results prove that using the suggested algorithm, the voltage set of the series cells has moved more uniformly.