International Journal of Power Electronics and Drive Systems (IJPEDS)

Vol. 15, No. 4, December 2024, pp. 2168~2179

ISSN: 2088-8694, DOI: 10.11591/ijpeds.v15.i4.pp2168-2179  2168

Active balancing charging using ANFIS to reach longest lifetime

for lithium ion

Farid Dwi Murdianto, Indhana Sudiharto, Azzahra Farah Andraeni

Department of Electrical Engineering, Politeknik Elektronika Negeri Surabaya, Surabaya, Indonesia

Article Info ABSTRACT

Article history: This research aims to address challenges in efficient energy storage in electric
vehicles, particularly those using lithium-ion batteries. The research focuses
Received Jan 16, 2024 on voltage imbalances in battery packs with cells connected in series, which
Revised May 27, 2024 can reduce the battery pack's lifespan. We investigated active balancing
Accepted Jun 13, 2024 charging methods using bidirectional fly-back converters in both pack-to-cell
and cell-to-pack modes. Current charging methods face limitations such as
prolonged balancing times, circuit complexity, and low efficiency. To
Keywords: overcome these challenges, this research proposes integrating an artificial
intelligence approach using adaptive neuro-fuzzy inference system (ANFIS)
Active balancing control. This approach allows for dynamic adjustment of duty cycles,
ANFIS expediting the balancing process. Experimental results show that employing
Bidirectional fly back converter ANFIS control can significantly reduce balancing times. In pack-to-cell mode,
Cell to pack balancing time decreased from 660 seconds to 600 seconds, while in cell-to-
Lithium ion battery pack mode, it decreased from 660 seconds to 580 seconds. This research
Pack to cell makes a significant contribution to the development of more efficient charging
technology for lithium-ion batteries in electric vehicles. Its potential
implications include improved battery lifespan and electric vehicle
performance in the future.
This is an open access article under the CC BY-SA license.

Corresponding Author:
Farid Dwi Murdianto
Department of Electrical Engineering, Politeknik Elektronika Negeri Surabaya
Raya ITS St. – PENS Campus Sukolilo, Surabaya 60111, Indonesia
Email: [email protected]

1. INTRODUCTION
Currently, the growth of electric vehicles has been rapid, making efficiency in electrical energy
storage a crucial aspect [1], [2]. Lithium-ion batteries are the preferred choice due to their high working voltage,
high energy density, lack of memory effect, low self-discharge rate, and long lifespan [3]. To meet the required
voltage and power capacity, batteries with cells connected in series are used [4], [5]. However, differences in
the output voltage of each battery cell or voltage imbalance may occur due to variations in manufacturing
processes, environmental conditions, and different charging and discharging cycles [6], [7]. When recharging
series-connected battery packs, some battery cells may require additional charging even after one cell reaches
full capacity [8], [9]. Battery imbalance can lead to overcharging or undercharging, significantly reducing
battery life and affecting battery capacity [1], [10]. In extreme cases, battery imbalance can pose a risk to
devices or even humans. Therefore, it is important to have an effective battery balancing system [11], [12].
In recent years, several battery balancing methods have been developed, with passive balancing and
active balancing being the main methods [13]. Passive balancing involves the use of resistors and switches
connected in parallel on each battery cell. Excess cells will dissipate energy through the resistors when the
system is operational. This method not only causes energy wastage but also presents challenges in thermal
management. Meanwhile, active balancing can be classified into five categories: adjacent cell to cell (AC2C),

Journal homepage: http://ijpeds.iaescore.com

Int J Pow Elec & Dri Syst ISSN: 2088-8694  2169

direct cell to cell (DC2C), pack to cell (P2C), cell to pack (C2P), and any cell to any cell (AC2AC). The active
balancing process involves transferring energy from high-voltage batteries to low-voltage batteries. Among
these categories, P2C and C2P utilize bidirectional flyback converters [14]. Beside their high efficiency, these
converters also offer ease of isolation, simple control, and small size [15], [16].
Previous research has applied this approach; however, these methods still have limitations such as
long balancing times, circuit complexity, and low efficiency. In this context, the use of artificial intelligence
techniques, particularly adaptive neuro-fuzzy inference system (ANFIS) control, offers a more adaptive and
intelligent solution compared to traditional approaches like fuzzy logic. While fuzzy logic methods can also
address voltage imbalances, ANFIS provides advantages in adaptability and higher intelligence. By combining
artificial neural networks and fuzzy logic, ANFIS dynamically adjusts the balancing process based on real-
time data and environmental conditions. This not only enhances efficiency but also reduces the risk of
overcharging or undercharging, thus extending battery life and improving overall safety. The results of the tests
and comparisons will be presented in this paper.

2. METHOD
In this paper, we delve into the design and implementation of an active voltage balancing system. This
system operates between cells and battery packs utilizing a bidirectional flyback converter. Furthermore,
ANFIS control is integrated to regulate the balancing process effectively.
Figure 1 shows that there are four battery cells connected in series. Each battery cell is paralleled with
a voltage sensor that provides feedback to the microcontroller [13]. This feedback is used to control the duty
cycle in the bidirectional flyback converter circuit. Each battery cell is connected to four relays, or two pairs
of positive and negative relays. One pair of relays is connected to the left side of the converter as a battery
pack, and the other pair is connected to the right side of the converter as a battery cell. The relays act as switches
to regulate the battery cell as a source or load, or as a battery pack or cell. The bidirectional flyback converter
is used to achieve the balancing function with a simpler structure.

Figure 1. Block diagram system

2.1. Balancing system

The system's balancing process is divided into two modes: from the battery pack to the cell and from
the cell to the battery pack. When operating in pack-to-cell mode, or conversely, cell-to-pack mode, this battery
pack, which consists of four series-connected battery cells, channels power from three cells to one other cell,
or vice versa. This configuration is illustrated in Figure 2.

Active balancing charging using ANFIS to reach longest lifetime for lithium ion (Farid Dwi Murdianto)
2170  ISSN: 2088-8694

Figure 2. displays the flowchart for the balancing system. Initially, voltage readings are taken from
each sensor in every battery cell. The average voltage value of the entire battery is then calculated. A
comparison is made to determine if any of the battery cells in batteries 1, 2, 3, or 4 have a voltage value greater
than the average. If so, that battery will be used as a source. Meanwhile, if any of the battery cells 1, 2, 3, or 4
has a voltage value lower than the average voltage value, it will be used as a load. In other words, if only one
battery cell voltage is lower than the average voltage, then the pack-to-cell mode will operate, and vice versa,
if only one battery cell voltage is higher than the average voltage, then the cell-to-pack mode will operate.
These two modes will alternate [11].

Figure 2. Flowchart of the balancing system

2.2. Sequential switch

Table 1 shows the sequential switches to be used in this system. S'1 to S'4 and Q'1 to Q'4 will connect
the balancing circuit with the battery pack on the left side of the converter. S1 to S4 and Q1 to Q4 will connect
the balancing circuit with the battery cells on the right side of the converter. In Table 1 for information number
1 if the battery is an active source, 0 if the battery is an active load, and tick if switch network 1 and switch
network 2 are connected.

Int J Pow Elec & Dri Syst, Vol. 15, No. 4, December 2024: 2168-2179
Int J Pow Elec & Dri Syst ISSN: 2088-8694  2171

Table 1. Sequential switch

Left side battery Right side battery Switch network 1 Switch network 2
B B B B B B B B S’ S’ S’ S’ Q’ Q’ Q’ Q’ S S S S Q Q Q Q
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
1 1 1 - - - - 0 √ - - - - - √ - - - - √ - - - √
- 1 1 1 0 - - - - √ - - - - - √ √ - - - √ - - -
0 0 0 - - - - 1 √ - - - - - √ - - - - √ - - - √
- 0 0 0 1 - - - - √ - - - - - √ √ - - - √ - - -

2.3. Bidirectional flyback converter modeling

To achieve bidirectional balancing, the implemented system replaces the rectifier diodes in
conventional flyback converters with synchronous rectifier MOSFETs [17] and reduces the losses caused by
the output rectifier diodes [11]. This enhancement significantly improves the efficiency of the system. The
complete converter circuit demonstrating these modifications is depicted as in Figure 3.
The planned balancing circuit consists of a transformer and two MOSFETs [18]. This converter
operates in both pack-to-cell and cell-to-pack modes. For instance, in pack-to-cell mode, the balancer enters
this mode of operation since the pack has a higher capacity. As illustrated in Figure 3. MOS1 is turned on while
MOS2 is turned off, and the pack energy is stored in the magnetizing inductor Lm. Then, MOS2 is turned on
while MOS1 is turned off. Currently, the energy stored in the magnetizing inductor is transferred to the
secondary winding to charge the battery cell. The opposite occurs in cell to pack mode [19]. Tables 2 and 3
provide the values for each component parameter used in this converter.

Figure 3. Bidirectional flyback converter circuit

Table 2. Flyback converter parameter for pack-to- Table 3. Flyback converter parameter in cell-to-
cell mode pack mode
Parameter Value Unit Parameter Value Unit
Vs (input voltage) (max) 12.6 Volt Vs (input voltage) (max) 4.2 Volt
Vs (min) 10.8 Volt Vs (min) 3.6 Volt
Vo (output voltage) 4.2 Volt Vo (output voltage) 12.6 Volt
Io (output current) 2.8 Ampere Io (output current) 2.8 Ampere
Po (output power) 11.76 Watt Po (output power) 35.28 Watt
Dmax (maximum duty cycle) 50 % Dmax (maximum duty cycle) 90 %
Fs (switching frequency) 100 Hertz Fs (switching frequency) 100 Hertz
Lm (magnetizing inductance) 22 µHenry Lm (magnetizing inductance) 2.7 µHenry
Np:Ns (transformer turns ratio) 4:2 winding Np:Ns (transformer turns ratio) 4:2 winding
Co (output capacitance) 4700 µFarad Co (output capacitance) 2200 µFarad

2.4. ANFIS control modeling

ANFIS control combines the learning capabilities of neural networks with the decision-making of
fuzzy control [20], [21]. The purpose of this combination is to optimize the advantages of each control and
reduce their shortcomings, resulting in ANFIS being a control system that belongs to the adaptive network
category [22]. The control system can adjust or adapt to the data and conditions present in the system [23].
ANFIS uses a set of data derived from the fuzzy control inputs and outputs generated by the ongoing
system simulation. This data is then used as a training dataset to achieve the desired error rate. When designing
ANFIS control, it is possible to adjust the number and type of membership functions as needed [24]. In this
case, 7x7 membership functions are used because a greater number of membership functions results in a smaller
error. Additionally, the training process uses a hybrid method that combines the least squares estimator (LSE)
and error back propagation (EBP) method [25], which is carried out over a thousand iterations. It is expected
that the error value will decrease and the resulting training data will improve with each iteration. Test results
are shown in Figures 4(a)-4(c).

Active balancing charging using ANFIS to reach longest lifetime for lithium ion (Farid Dwi Murdianto)
2172  ISSN: 2088-8694

(a)

(b)

(c)

Figure 4. ANFIS control modeling: (a) ANFIS structure, (b) variable duty cycle output, and (c) ANFIS
structure viewer

2.5. Open loop simulation modeling

The simulation's parameter values are adjusted to match the calculation results. PSIM 2022 software
is used for this simulation. Figure 5 displays the system simulation circuit in both pack-to-cell and cell-to-pack
modes. In this simulation, the duty cycle value is manually set. Specifically, the duty cycle value is set
at 43.03% for pack-to-cell mode and 63.12% for cell-to-pack mode, as determined by the system
requirements. These values are crucial for determining the performance of the balancing system under different
operational conditions.

Int J Pow Elec & Dri Syst, Vol. 15, No. 4, December 2024: 2168-2179
Int J Pow Elec & Dri Syst ISSN: 2088-8694  2173

Figure 5. Open loop simulation circuit

2.6. Simulation modeling of closed-loop integration

Closed-loop integration simulation (closed circuit) is a simulation of the entire system using ANFIS
control. The parameter values used for the simulation are adjusted to the calculation results. The simulation
was conducted using PSIM 2022 software. Figure 6 shows the system simulation circuit in pack-to-cell and
cell-to-pack modes.
This simulation utilizes ANFIS to provide feedback in the form of duty cycle. The duty cycle is
obtained by integrating real-time Vo readings with the set point voltage value, which is set to produce the
desired output. The aim is to improve upon the open loop simulation (without control).

Active balancing charging using ANFIS to reach longest lifetime for lithium ion (Farid Dwi Murdianto)
2174  ISSN: 2088-8694

Figure 6. Close loop integration simulation circuit

3. RESULTS AND DISCUSSION

The simulations were conducted using PSIM 2022 software in both pack-to-cell and cell-to-pack
modes. After performing open-loop simulation and closed-loop integration, the system's performance was
evaluated. The obtained results provide insights into the effectiveness of the proposed approach.

3.1. Open loop simulation testing

Following open loop simulation testing, the test results are presented as the voltage of each cell,
namely VB1, VB2, VB3, and VB4, before and after balancing. These results are crucial for evaluating the
performance of the balancing system. For further details and comprehensive analysis, as in Tables 4 and 5.

Int J Pow Elec & Dri Syst, Vol. 15, No. 4, December 2024: 2168-2179
Int J Pow Elec & Dri Syst ISSN: 2088-8694  2175

Table 4 shows the results of open loop simulation testing in pack to cell mode. Initially, the voltage
of batteries 1, 2, and 3 was higher than the average voltage, and therefore they were used as a source, forming
a series-connected battery pack. Regarding the battery voltage, 4 is lower than the average voltage and is
therefore used as a load. This can be considered balanced because the voltage of each battery was the same for
a balancing time of 660 seconds. The simulation test results for each battery voltage can be seen in Figure 7.
Table 5 shows the results of open loop simulation testing in cell to pack mode. Initially, batteries 1,
2, and 3 had a voltage lower than the average, so they were used as a load, forming a series-connected battery
pack. Battery 4 had a voltage higher than the average, so it was used as a source. This can be considered
balanced because the voltage of each battery was the same for a balancing time of 660 seconds. Figure 8 shows
the simulation test results for each battery voltage.

Table 4. Test results for open loop simulation of pack to cell mode
Parameter Before balancing After balancing Unit
VB1 4.222 4.149 Volt
VB2 4.222 4.149 Volt
VB3 4.222 4.149 Volt
VB4 3.942 4.149 Volt

Table 5. Test results for open loop simulation of cell to pack mode
Parameter Before balancing After balancing Unit
VB1 3.942 4.008 Volt
VB2 3.942 4.008 Volt
VB3 3.942 4.008 Volt
VB4 4.222 4.008 Volt

Figure 7. Open loop simulation test results of pack to cell mode graph
(horizontal axis: time (s), vertical axis: battery voltage (V))

3.2. Simulation testing of closed loop integration

Following the close loop integration simulation testing, test results were obtained in the form of the
voltage of each cell, namely VB1, VB2, VB3, and VB4 before and after balancing. The results are presented in
Tables 6 and 7. These tables provide a detailed overview of the voltage levels and the effectiveness of the
balancing process. Table 6 shows the results of the close loop integration simulation testing in pack to cell mode.
Initially, the voltage of batteries 1, 2, and 3 was higher than the average voltage, making them a suitable source,
namely a series-connected battery pack. Regarding the battery voltage, 4 is lower than the average voltage, so it
is utilized as a load. This can be considered balanced because the voltage of each battery was the same for a
balancing time of 600 seconds. The simulation test results for each battery voltage can be observed in Figure 9.
Active balancing charging using ANFIS to reach longest lifetime for lithium ion (Farid Dwi Murdianto)
2176  ISSN: 2088-8694

Table 7 shows the results of the closed-loop integration simulation testing in cell-to-pack mode.
Initially, batteries 1, 2, and 3 had a voltage lower than the average voltage, so they were used as a load, forming
a series-connected battery pack. Battery 4 had a voltage higher than the average, so it was used as a source.
This can be considered balanced because the voltage of each battery was the same for a balancing time of 580
seconds. Figure 10 shows the simulation test results for each battery voltage.

Table 6. Simulation testing results of closed-loop integration of pack-to-cell mode

Parameter Before balancing After balancing Unit
VB1 4.126 4.078 Volt
VB2 4.126 4.078 Volt
VB3 4.126 4.078 Volt
VB4 3.937 4.078 Volt

Figure 8. Open loop simulation test results graph for cell-to-pack mode
(horizontal axis: time (s), vertical axis: battery voltage (V))

Figure 9. Graph of simulation test results for closed-loop integration of pack-to-cell mode
(horizontal axis: time (s), vertical axis: battery voltage (V))

Int J Pow Elec & Dri Syst, Vol. 15, No. 4, December 2024: 2168-2179
Int J Pow Elec & Dri Syst ISSN: 2088-8694  2177

Table 7. Simulation testing results of closed loop integration of cell to pack mode
Parameter Before balancing After balancing Unit
VB1 3.942 3.994 Volt
VB2 3.942 3.994 Volt
VB3 3.942 3.994 Volt
VB4 4.158 3.994 Volt

Figure 10. Graph of simulation test results for closed-loop integration from cell to pack mode
(horizontal axis: time (s), vertical axis: battery voltage (V))

3.3. Comparison of open loop simulation and closed loop integration test results
The test results for the open loop simulation without control and the closed loop integration simulation
using ANFIS control are shown above. The balancing times for each mode are presented. A comparison was
made between simulations with and without control, and the results are shown in Table 8.
After conducting simulations and closed-loop integration using PSIM 2022 software, the obtained
results indicate that the use of adaptive neuro-fuzzy inference system (ANFIS) control has a significant impact
on the battery voltage balancing time. In the pack to cell mode simulation, the balancing time with ANFIS
control is 600 seconds, whereas without ANFIS control, it requires 660 seconds. Similarly, in the cell to pack
mode simulation, the balancing time with ANFIS control is 580 seconds, whereas without control, it requires
the same time as the previous mode, which is 660 seconds.
The interpretation of these results indicates that the use of ANFIS control significantly accelerates the
battery voltage balancing process, reducing the time required to achieve the desired level of balance.
This has important implications in practical applications, especially in electric vehicles where battery efficiency
and performance are crucial. By speeding up the balancing time, ANFIS control can enhance battery
charging efficiency, extend battery life, and ultimately improve the overall performance and durability of
electric vehicles.

Table 8. Comparison of simulation test results

Mode Control Balancing time (s)
Pack to cell Without control 660
With ANFIS 600
Cell to pack Without control 660
With ANFIS 580

4. CONCLUSION
Overall, this study addresses challenges associated with efficient energy storage in electric vehicles,
with a specific focus on lithium-ion batteries. The primary investigation revolves around voltage imbalances

Active balancing charging using ANFIS to reach longest lifetime for lithium ion (Farid Dwi Murdianto)
2178  ISSN: 2088-8694

in series-connected battery packs, a factor known to compromise the overall battery pack lifespan. Active
balancing methods are explored, utilizing bidirectional flyback converters in configurations from both pack to
cell and cell to pack. These methods aim to overcome common limitations in existing charging techniques,
including lengthy balancing times, circuit complexity, and suboptimal efficiency.
To address these challenges, the study proposes the integration of an artificial intelligence approach
through adaptive neuro-fuzzy inference system (ANFIS) control. This innovative methodology facilitates
dynamic adjustment of duty cycles, thus significantly expediting the balancing process. Experimental findings
affirm the effectiveness of ANFIS control, demonstrating a significant reduction in balancing time.
Specifically, in pack-to-cell mode, the balancing time decreased from 660 seconds to 600 seconds, while in
cell-to-pack mode, it decreased from 660 seconds to 580 seconds.
Essentially, this Packs”, in research presents a simulation model for a balancing system aimed at
enhancing battery performance and reliability. By leveraging active balancing techniques and integrating
ANFIS control, the proposed system offers promising prospects for advancing lithium-ion battery charging
technology in electric vehicles. The results are expected not only to improve battery lifespan but also to enhance
electric vehicle performance in the future.

REFERENCES
[1] Oriti G. and Julian A.L, Norgaard P, “Battery Management System with Cell Equalizer for Multi-Cell Battery Pack,” 2014 IEEE
Energy Conversion Congress and Exposition, ECCE 2014, 2014, doi: 10.1109/ECCE.2014.6953493.
[2] J. C. A. Pelayo, S. J. V. Tan, P. S. D. Yu, G. V. Magwili, and F. R. G. Cruz, “Simulated Solar Assisted Battery Management System
with Fuzzy Temperature Control, Flyback Converter Active Cell Balancing Circuit and Coulomb Counting SoC Estimation Method
using MATLAB Simulink,” International Conference on Humanoid, Nanotechnology, Information Technology, Communication
and Control, Environment, and Management, HNICEM 2020, Institute of Electrical and Electronics Engineers Inc., Dec. 2020. doi:
10.1109/HNICEM51456.2020.9400158.
[3] H. Xiong, D. Song, F. Shi, Y. Wei, and L. Jinzhen, “Novel voltage equalization circuit of the lithium battery pack based on
bidirectional flyback converter,” IET Power Electronics, vol. 13, no. 11, pp. 2194–2200, Aug. 2020, doi: 10.1049/iet-pel.2019.1620.
[4] P. Ramesh, A. Patra, D. Kastha, “A Transformer Coupled Modular Multi-Active Half Bridge Charge Balancing Architecture in a
Lithium-ion Battery Pack for Electric Vehicles ,” IEEE International Conference on Power Electronics, Drives and Energy Systems
(PEDES), 2020, doi : 10.1109/PEDES49360.2020.9379807.
[5] M. Raber, D. Hink, A. Heinzelmann, D. Ould Abdeslam, “A Novel Non-Isolated Active Charge Balancing Architecture for Lithium-
Ion Batteries,” International Symposium on Industrial Electronics (ISIE), 2018, doi 10.1109/ISIE.2018.8433669.
[6] F. Shi and D. Song, “A novel high-efficiency double-input bidirectional DC/DC converter for battery cell-voltage equalizer with
flyback transformer,” Electronics (Switzerland), vol. 8, no. 12, Dec. 2019, doi: 10.3390/electronics8121426.
[7] P. Ramesh, J. Sahoo, A. Patra, D. Kastha, “Development of Parallel Charge Transfer Path Using Non-Isolated Active Half-Bridge
Charge Balancing Architecture,” IEEE International Conference on Power Electronics, Smart Grid, and Renewable Energy
(PESGRE), 2023, doi 10.1109/PESGRE58662.2023.10404967.
[8] C. Yao Wu, Y. Ying Huang, C. Huan Ku, “Development of an active and passive balancing strategy for a LiFePO4 battery pack,”
IEEE International Conference on Applied System Invention (ICASI), 2018, doi : 10.1109/ICASI.2018.8394356.
[9] G. L. K. Moganti, V. N. S. Praneeth, S. R. K. Vanjani, “An Implantable Bipolar Active Charge Balancing Circuit with Six
Adjustment Current levels for Facial Paralysis Patients,” 2021 11th IEEE International Conference on Intelligent Data Acquisition
and Advanced Computing Systems: Technology and Applications (IDAACS), 2021, doi: 10.1109/IDAACS53288.2021.9660895.
[10] L. Yao, P. Li, M. Je, “A pulse-width-adaptive active charge balancing circuit with pulse-insertion based residual charge
compensation and quantization for electrical stimulation applications,” IEEE Asian Solid-State Circuits Conference (A-SSCC),
2015, doi : 10.1109/ASSCC.2015.7387478.
[11] Y. L. Lee, C. H. Lin, and S. J. Yang, “Power loss analysis and a control strategy of an active cell balancing system based on a
bidirectional flyback converter,” Applied Sciences (Switzerland), vol. 10, no. 12, Jun. 2020, doi: 10.3390/app10124380.
[12] Y. Shang, C. Zhang, N. Cui, and J. M. Guerrero, “A cell-to-cell battery equalizer with zero-current switching and zero-voltage gap
based on quasi-resonant lc converter and boost converter,” IEEE Trans Power Electron, vol. 30, no. 7, pp. 3731–3747, Jul. 2015,
doi: 10.1109/TPEL.2014.2345672.
[13] S. L. Wu, H. C. Chen, and C. H. Chien, “A novel active cell balancing circuit and charging strategy in lithium battery pack,”
Energies (Basel), vol. 12, no. 23, 2019, doi: 10.3390/en12234473.
[14] K. Yu, Y. Shang, X. Wang, N. Wang,, B. Duan, and C. Zhang, “A Multi-Cell-to-Multi-Cell Equalizer for Series-Connected Batteries
Based on Flyback Conversion,” Conference on Vehicle Control and Intelligence (CVCI), 2019, doi:
10.1109/CVCI47823.2019.8951748
[15] S. T. Wu, Y. N. Chang, C. Y. Chang, and Y. T. Cheng, “A fast charging balancing circuit for LiFePO4 battery,” Electronics
(Switzerland), vol. 8, no. 10, Oct. 2019, doi: 10.3390/electronics8101144.
[16] A. Ahmad, V. Singh, A. V. R. Teja, S. Payami, “Parallel Active Charge Balance Technique of Li-Ion Batteries Using Combined
Phase Shifted and Boost DC-DC Converter,” Annual Conference of the IEEE Industrial Electronics Society, 2023, doi:
10.1109/IECON51785.2023.10311979.
[17] P. Thummala, H. Schneider, Z. Zhang, and M. A. E. Andersen, “Bidirectional flyback converter with multiple series connected
outputs for high voltage capacitive charge and discharge applications,” in Conference Proceedings - IEEE Applied Power
Electronics Conference and Exposition - APEC, Institute of Electrical and Electronics Engineers Inc., May 2015, pp. 50–56. doi:
10.1109/APEC.2015.7104331.
[18] R. Rakhmawati, S. Suhariningsih, and S. Husniatuzzahro’, “Design flyback converter for charging battery using fuzzy logic
controller,” in Proceedings - 2020 International Seminar on Application for Technology of Information and Communication: IT
Challenges for Sustainability, Scalability, and Security in the Age of Digital Disruption, iSemantic 2020, Institute of Electrical and
Electronics Engineers Inc., Sep. 2020, pp. 396–402. doi: 10.1109/iSemantic50169.2020.9234195.

Int J Pow Elec & Dri Syst, Vol. 15, No. 4, December 2024: 2168-2179
Int J Pow Elec & Dri Syst ISSN: 2088-8694  2179

[19] Yang D., Li S., Qi G., “A Bidirectional Flyback Cell Equalizer for Series-connected Lithium Iron Phosphate Batteries,”
International Conference on Power Electronics Systems and Applications (PESA), 2015, doi : 10.1109/PESA.2015.7398904.
[20] I. Sudiharto, F. D. Murdianto, and L. Isnani, “Implementation smart charging technique using particle swarm optimization to achieve
best performance charger,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 14, no. 3, pp. 1604-1614.
2023, doi: 10.11591/ijpeds.v14.i3.pp1604-1614.
[21] A. M. Noman, K. E. Addoweesh, and A. I. Alolah, “Simulation and Practical Implementation of ANFIS-Based MPPT Method for
PV Applications Using Isolated Ćuk Converter,” International Journal of Photoenergy, 2017 doi: 10.1155/2017/3106734.
[22] R. Erfani, F. Marefat, S. Mandal, P. Mohseni, “A 1.3mA biphasic current stimulator IC with active charge balancing for nerve
interfacing applications,” IEEE International Symposium on Circuits and Systems (ISCAS), 2016, doi:
10.1109/ISCAS.2016.7527434.
[23] E. Joelianto, B. Rahmat, “Adaptive neuro fuzzy inference system (ANFIS) with error backpropagation algorithm using mapping
function,” 2008. International Journal of Artificial Intelligence, Vol. 1, number A08.
[24] G. H. Min, J. I. Ha, “Active cell balancing algorithm for serially connected li-ion batteries based on power to energy ratio,” IEEE
Energy Conversion Congress and Exposition (ECCE), 2017, doi: 10.1109/ECCE.2017.8096514.
[25] S. Shabaan, M. I. Abu El-Sebah, and P. Bekhit, “Maximum power point tracking for photovoltaic solar pump based on ANFIS
tuning system,” Journal of Electrical Systems and Information Technology, vol. 5, no. 1, pp. 11–22, May 2018, doi:
10.1016/j.jesit.2018.02.002.

BIOGRAPHIES OF AUTHORS

Farid Dwi Murdianto is a lecturer at Electronic Engineering Polytechnic

Institute of Surabaya (EEPIS). He has a bachelor's degree in Applied Engineering in
Industrial Electrical deriving out of EEPIS and a master's degree in Engineering in Power
Systems deriving out of ITS. Renewable energy and power electronics are his research fields
of interest. As a lecturer, he has taught subjects such as renewable energy, power quality, and
microcontrollers. He has been teaching for more than six years. He can be contacted at email:
[email protected].

Indhana Sudiharto is a lecturer at Electronic Engineering Polytechnic Institute

of Surabaya (EEPIS). He has a bachelor of engineering degree in Power Systems and a master
of engineering degree in Power Systems from ITS. Power systems, renewable energy, and
power electronics are his research fields of interest. He has more than 27 years of teaching
experience and has taught subjects such as electrical measurement and instrumentation,
electrical power system protection, and power quality. He can be contacted at email:
[email protected].

Azzahra Farah Andraeni is a student who is currently pursuing a bachelor's

degree in Applied Engineering at Electronic Engineering Polytechnic Institute of Surabaya
(EEPIS). Currently, she is doing the final project research in power electronics, specifically
balancing systems. Her areas of interest include power electronics, renewable energy, and
power quality. She can be contacted at email: [email protected].

Active balancing charging using ANFIS to reach longest lifetime for lithium ion (Farid Dwi Murdianto)