A Fast Equalization Method for Series Batteries based on Cuk

Converters

Qinghai Meng and Zhou Zhang

North China University of Technology, Shijingshan, Beijing, China

Keywords: Cuk converters, SOC, Series batteries, Equalization

Abstract: In series battery packs, the accumulated differences among every single battery could significantly decrease

the performances of the packs. Consequently, equalization circuits are very necessary for series batteries.

Based on Cuk converters, this paper introduces a fast equalization method considering SOC (State of

Charge) differences of series batteries. The method increases equalization current as large as possible, and

simultaneously, achieves fast equalization and low consumption. The experimental platform is constructed,

and the results verify the validity of the proposed fast equalization method.

1 INTRODUCTION

With the development of technology, the energy

storage system composed of batteries is widely

applied in the fields of microgrids and new energy

vehicles(Zhang and Wang, 2019). The production

and working environment may lead to the

differences among serial batteries, which could

significantly decrease the performance of the whole

pack(Liu and Zou, 2018). It is of great significance

to apply equalization methods to eliminate the

differences, so as to ensure the high-efficiency of

utilization of energy in the battery pack(Zhiliang and

Xiang, 2018).

The Cuk converter topology is widely used to

perform equalization function(Rui and Lizhi, 2015).

Literature(Yan and Cheng, 2015) surveys the battery

equalization using a fuzzy controller.

Literature(Ouyang and Chen, 2018) researches

quasi-sliding mode control based on Cuk converters

to balance the batteries. Literature (Samadi and Saif,

2014) discusses the predictive control for cell

balancing in Li-ion battery packs. Unfortunately, the

control methods mentioned above are too

complicated to perform an equalization function, and

the balancing speed may not be so fast. This paper

introduces a fast equalization method considering

SOC (State of Charge) differences of series batteries

based on Cuk converters, and simultaneously,

achieves fast equalization and low consumption. The

experimental platform is constructed, and the results

verify the validity of the proposed fast equalization

method.

2 TOPOLOGY ANALYSIS

This paper adopts the bidirectional DC/DC circuit

based on the Cuk converters, as shown in figure 1.

The equalization circuit is divided into two parts, the

left part is a double-layer bridge, mainly used for

selecting the single cell needed to be balanced, and

the right half performs an equalization function,

controlling energy flowing in both directions.

Figure 1. Topology based on the Cuk converter.

3 EQUALIZATION BASIS AND

ITS ESTIMATION

The battery’s state of charge (SOC) is selected as the

equalization basis, which is of more significant

Meng, Q. and Zhang, Z.

A Fast Equalization Method for Series Batteries based on Cuk Converters.

DOI: 10.5220/0011104800003355

In Proceedings of the 1st International Joint Conference on Energy and Environmental Engineering (CoEEE 2021), pages 18-21

ISBN: 978-989-758-599-9

advantages than the terminal voltage of batteries.

Because the relationship between SOC and the

terminal voltage is nonlinear. SOC can directly

reflect a battery’s state, which can help to obtain a

more accurate result.

After the selection of the equalization basis, the

more important part is to estimate SOC in a proper

way. This paper is based on extended Kalman filter

(EKF) to complete this. The core of EKF is the use

of Taylor expansion for partial linearization. This

paper builds an equation of second-order RC battery

model, as shown in figure 2. The state variables are

the SOC、electrochemical polarization voltage U

and concentration polarization voltage U

controlling variable I. Hence the equation can be

derived as formula (1).

Figure 2. The second-core RC battery model.

()

kWI

ηT

SOC

+×

























−×













−×+





































e12

e11

e00

0e0

001

(1)

where η is the Cullen Coefficient, which can be

obtained from battery discharging experiment.

Generally, when the battery is charging, η equals to

1, and less than 1 when discharging.

The observation equation is expressed as:

()

k0kkOCVk

21 VIRUUSOCUU ++++=

(2)

The Kalman Filter is initialized as:

()()

[]



×=

）（

vvEv

wwEw

x-xx-xEx

xEx

00000

ˆˆ

)(

(3)

and recursived as:

()

[]

()





 

−

−−+

−

−−

−

−=

−+=

kkkk

kkkkkk

kkk

kkkk

kkk

xCLEx

vxgyLxx

vCxCCxL

wAxAx

uxfx

ˆˆˆ

(

(4)

According to the Hybrid Pulse Power

Characteristic (HPPC) test presented in the Freedom

Car manual(FreedomCar, 2003) and existing data

obtained before, the time constant

、

and

polarization resistance R

、R

can be received from

the MATLAB curve fitting toolbox using the

parameter identification means.

4 CONTROL STRATEGY

In this paper, constant current method is used for

equalization, and the loss can be expressed as

RtIQ

(5)

where I

is the equalizing current, R is the equivalent

resistance during the process, and t is the time

needed for equalizing. Another equalization method

is to use a rectangular wave with a duty cycle α. If

the two types of equalizing currents start in the same

time, the amplitude of the rectangular wave current

needs to be increased. Therefore, the loss of the

rectangular wave current balanced in one switching

cycle can be derived as

tRIQ

(6)

where I

is the rectangular wave current. The duty

cycle varies from 0 to 100%, so the conclusion is

QQ >

(7)

Only when the SOC imbalance degree of the battery

group reaches a certain degree, the equilibrium

control begins, and the balance is judged by the

variance of the SOC to open the equilibrium.The

control policy flowchart is shown in figure 3.

A Fast Equalization Method for Series Batteries based on Cuk Converters

SOC variance

>threshold

Turn on

equalization

control

Decide the

number of the

battery with

the largest

deviation

Turn on the

Correspondin

g MOSFETs

Closed-loop

control the

equalization

current

SOC－SOC median

＜threshold

YES

Figure 3. The control strategy flow chart.

SIMULATION AND

EXPERIMENT

5.1 Simulation Verification

First, the rated capacity of each single battery in the

simulation is 60Ah, and the rated voltage is 3.3V. In

order to get the results faster, the initial SOCs are set

to 65.502%, 65.503%, and 65.500% respectively.

Figure 4 shows the simulation result using the

proposed control strategy to equalize three batteries.

The SOC of the first battery was set as the median.

The initial SOC deviation of the third battery was

bigger than that of the second one. Hence the third

battery needs to be charging equalized first, where

sets the charging current as 5A. Then the second

battery needs to be completed discharging

equalization, where sets the discharging current as

10A. Figure 5 shows the current of serial battery

packs and the assisting battery respectively.

Figure 4. Simulation result of the equalization process.

Figure 5. The current wave of both sides.

5.2 Experimental Verification

An experimental platform is built for the battery

equalization, as shown in figure 6. Firstly, the

battery pack is discharging equalized, and the

reference values of the discharging currents are 2A,

4A, 6A, 8A, and 10A, respectively. The wave when

the current value is 10A are shown in figure 7. Then

the battery pack is charging equalized, and the

reference values of the charging currents are 2A, 3A,

4A, and 5A, respectively. All the experiments are

successfully realized quantitative control of the

currents. The waves when current value is 5A are

shown in figure 7. And the results verify that method

could achieve fast equalization and low assumption.

Figure 6. The experimental platform.

Figure 7. Discharging and charging current wave.

CoEEE 2021 - International Joint Conference on Energy and Environmental Engineering

6 CONCLUSION

This paper introduces a fast equalization method

considering SOC (State of Charge) differences of

series batteries based on Cuk converters. The

method increases equalization current as large as

possible, and simultaneously, achieves fast

equalization and low consumption. The

experimental platform is constructed, and the results

verify the validity of the proposed fast equalization

method.

ACKNOWLEDGMENTS

I’d like to thank all the teachers and students who

helped me to finish this paper. Without their help,

the work can not be completed successfully.

REFERENCES

H. Zhang, Y. Wang, H. Qi and J. Zhang, Active battery

equalization method based on redundant battery for

electric vehicles[J], IEEE Transactions on Vehicular

Technology, 2019, 68(8), p 7531-7543.

X. Liu and Y. Zou, The proportional current control

strategy for equalization circuits of series battery

packs, ICEMS, Jeju, 2018, p 846-849.

Z. Zhang, X. Cheng, Z. Lu and D. Gu, SOC estimation of

lithium-ion battery pack considering balancing current,

IEEE Transactions on Power Electronics, 2018, 33(3),

p 2216-2226

L. Rui, W. Lizhi, H. Xueli, D. Qiang and Z. Jie, A review

of equalization topologies for lithium-ion battery

packs, CCC, Hangzhou, 2015, p 7922-7927.

J. Yan, Z. Cheng, G. Xu, H. Qian and Y. Xu, Fuzzy

control for battery equalization based on state of

charge, IEEE 72nd Vehicular Technology Conference

- Fall, Ottawa, ON, 2010, p 1-7.

Q. Ouyang, J. Chen, J. Zheng and Y. Hong, SOC

estimation-based quasi-sliding mode control for cell

balancing in lithium-ion battery packs, IEEE

Transactions on Industrial Electronics, 2018, 65(4), p

3427-3436.

M. F. Samadi and M. Saif, Nonlinear model predictive

control for cell balancing in Li-ion battery packs, 2014

American Control Conference, Portland, OR, 2014, p

2924-2929.

Freedom Car Battery Test Manual For Power-Assist

Hybrid Electric Vehicles. 2003: p 13-16.

A Fast Equalization Method for Series Batteries based on Cuk Converters