Development of an Active Equalizer for Lithium-Ion Batteries

• Published in: Electronics (Basel) Vol. 11; no. 14; p. 2219
• Main Author: Yang, Zong-Zhen
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 15.07.2022
• Subjects: Balancing; Control methods; Converters; Design; Efficiency; Electric potential; Electric vehicles; Energy storage; Equalizers; Equilibrium; Lithium; Lithium-ion batteries; Rechargeable batteries; State of charge; Switching; Voltage
• DOI: 10.3390/electronics11142219
• ISSN: 2079-9292

In this paper, a bi-directional-buck-boost-converter-based active equalizer is developed. The energy between adjacent cells can be transferred bi-directionally by manipulating the balancing current to solve the unbalanced problem in a battery module. It is noted that the conduction time of the main switch in the conventional buck-boost equalizer is fixed. Thus, the balancing current will diminish as the voltage difference of the adjacent cells decreases, which results in a prolonged equilibrium period. This paper has proposed two methods, namely, the varied-on-time (VOT) method and the voltage ratio modulation (VRM) method, to shorten the equilibrium period. In the VOT method, the conduction time of the main switch is determined according to high state-of-charge (SOC) cell voltage. In this way, the balancing current is able to be kept at the desired level rather than reduced during the balancing process. On the other hand, the VRM method computes the proportion of the conduction time and the cut-off time in a switching cycle based on the voltages of adjacent cells. Hence, the equalizer can deliver the maximum energy in a switching period and shorten the equilibrium period. The simulation platform and experiments with four batteries connected in serial are carried out to verify the proposed control methods. According to the experimental results, the VOT method saves 10.3%, 11.7%, and 16% of the equilibrium time compared with the fixed duty cycle (FDC) method. The VRM method can shorten 35.9%, 36.6%, and 37.3% of the equilibrium time compared with the FDC method.