Materials Today: Proceedings xxx (xxxx) xxx

Contents lists available at ScienceDirect

Materials Today: Proceedings

journal homepage: www.elsevier.com/locate/matpr

Design, simulation analysis of universal battery management system for

EV applications
R. Sreedhar ⇑, K. Karunanithi
Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Avadi, Chennai, India

a r t i c l e i n f o a b s t r a c t

Article history: Now a days, Electric Vehicles (EVs) are familiar with abundant types of battery chemistries with different
Received 20 November 2020 specifications. Battery Management System (BMS) is an essential and important device of an EV for
Accepted 2 December 2020 smooth operation and long life. In present scenario, a separate BMS has been used for the particular bat-
Available online xxxx
tery. So, there is a need of a common BMS for all type of EVs. This paper proposes a simulation design of
common BMS for four different batteries namely Nickel-Metal-Hydride, Nickel-Cadmium, Lithium-Ion
Keywords: and Lead-Acid. A common simulation and mathematical modelling of aforementioned four types of bat-
Battery Management System (BMS)
tery is executed here. The proposed system is designed to manipulate and control the instantaneous val-
Electrical Vehicles (EVs)
Extended Kalman Filter (EKF)
ues of the battery parameters like Battery voltage, Current, Power consumption and State of Charge (SoC).
State of Charge (SoC) This proposed BMS continuously monitors the battery parameters with high accuracy, especially in SoC
measurement. The working of BMS is controlled and monitored by using Extended Kalman Filter (EKF).
The entire mathematical modelling and simulation work of BMS is analysed by using MATLAB/Simulink.
Ó 2021 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the Emerging Trends in
Materials Science, Technology and Engineering.

1. Introduction also in small power applications as Uninterrupted Power Supplies

(UPS), inverter systems [4,5].
Batteries are one of the most commonly used energy storage Battery Monitoring means continuously observing the battery
devices in several locations like EVs Smart Grids, and Renewable functional constraints during its operating period. A good BMS
Power Plants. EVs are subjected to highly varying loads due to has to monitor the parameters like operating voltage, current,
many reasons like the change in speed of vehicle, road condition, power consumption, SoC, SoH, battery aging, internal impedance
traffic and variable load conditions. Due to this the capacity of bat- and ambient temperature during charging and discharging cycles.
tery is drastically affected. So, it is need to monitor the working of The study of simulation characteristics, modelling of Li-ion and
battery in all aspects when the battery is subjected to a continu- lead-Acid battery packs are presented in [6–8]. The health progno-
ously fluctuating load. sis and safety monitoring of Li-Ion battery is clearly discussed and
A BMS can manage a secondary battery cell or pack, by safe- the ways to maintain the health of the battery is clearly explained
guarding the battery to run within its safe limits and by keeping in [9,10].
an eye on battery characteristics like State of charge (SoC), State The reliability of EVs are subjected to the exact value of SoC and
of health (SoH), Thermal management and battery aging. Different SoH monitored by the BMS implemented in a particular EV. There
types of BMS with battery modelling and equivalent circuit models are lot of conventional and advanced methods are available to
are explained clearly in [1]. The design and implementation of a monitor SoC and SoH of a battery. Different SoH estimation meth-
smart and advanced BMS using MATLAB software is proposed in ods based on applications are reviewed in [11,12]. The proposed
[2,3]. BMS has lot of applications in Smart Grids, EV’s, Solar and BMS is validated by using the real-time physical constraints and
the predefined parameters of the four different batteries.
This paper proposes a new design & simulation of Universal Bat-
tery Management System which monitors the battery parameters
⇑ Corresponding author. like load voltage, load current, power consumption and SoC of a
E-mail addresses: [email protected] (R. Sreedhar), drkkarunanithi@- battery. The BMS explored in Extended Kalman Filter (EKF)
veltech.edu.in (K. Karunanithi).

https://doi.org/10.1016/j.matpr.2020.12.136
2214-7853/Ó 2021 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the Emerging Trends in Materials Science, Technology and Engineering.

Please cite this article as: R. Sreedhar and K. Karunanithi, Design, simulation analysis of universal battery management system for EV applications, Mate-
rials Today: Proceedings, https://doi.org/10.1016/j.matpr.2020.12.136
R. Sreedhar and K. Karunanithi Materials Today: Proceedings xxx (xxxx) xxx


algorithm since it is one of the accurate methods to determine SoC Eb ¼ E0  Z  I  K  ðQ =Q  it Þ  it þ I0 þ A  eðBitÞ ð1Þ
of a battery. The objective of the paper is to develop and simulate a
common mathematical model of batteries and hence to progress a Lead-Acid:
management system for battery parameters. Mathematical mod-

Eb ¼ E0  Z  I  K  ðQ =Q  it Þ  it þ I0 þ et ð2Þ
elling of four type of batteries namely Lithium-Ion, Nickel-
Cadmium, Lead-Acid and Nickel-Metal-Hydride are constructed. Nickel-Cadmium:
The block diagram of the proposed Universal BMS, simulation dia-

gram and its working, graphical representation of the battery con- Eb ¼ E0  Z  I  K  ðQ =Q  it Þ  it þ I0 þ et ð3Þ
straints are shown in the upcoming sections of this paper. Nickel-Metal-Hydride:


2. Block diagram of proposed universal BMS Eb ¼ E0  Z  I  K  ðQ =Q  it Þ  it þ I0 þ et ð4Þ

Fig. 1 shows that the block diagram of the proposed BMS. Here Where
four battery parameters are taken into consideration for this BMS Eb = EMF of a loaded battery (V)
namely SoC, Power consumption, battery operating voltage and E0 = No-load battery EMF (V)
load current delivered by the battery. It is designed to select a par- K = polarization constant (V/(Ah)) or (X)
ticular type of battery from the given four batteries. At a time, a Q = Maximum capacity of the battery (Ah)
R
single battery can be analysed with charge–discharge cycles. It is it = idt Remaining charge of the battery (Ah)
also probable to start the charge–discharge cycle of the proposed A = exponential zone amplitude (V)
system with different SoC rating. All measurements are done with B = exponential zone time constant inverse (Ah)  1
the help of MATLAB/SIMULINK. Z = internal impedance of the battery (X)
Here controlled voltage source acts as battery source which is I = Load current delivered by battery (A)
connected to the load through current measurement block. The I’ = filtered current (A)
current measurement system ensures that the load current will There are number of methods are available to estimate the State
not exceed the specified C rating of a particular battery while the of Charge of a battery. Here SoC is calculated by measuring the
EV is in running condition. The SoC of the battery is measured by remaining charge in a battery. SoC can be manipulated by the fol-
calculating the available charge in the battery. The charge estima- lowing expression.
tion is used for two purpose namely to estimate the SoC of the bat-
SoC ¼ 100  ðð1  itÞ=Q Þ ð5Þ
tery as well as to measure the battery voltage. The power
consumption of a battery can be calculated by the multiplied result Where, it = Remaining charge of the battery
of voltage and current of a particular battery. The simulation is Q = Maximum battery capacity (AH)
designed such that, to select a battery type from the four different Here SoC is measured by using Extended Kalman Filter because
batteries. By automatically identifying the selected battery type, it is one of the methods to give most accurate SoC value.
the suitable modelling equation can be used for further process like
charge estimation, remaining battery voltage estimation, SoC mea-
surement etc., The scope will display the graphical representation 4. Simulation of the proposed universal BMS
of parameters and the display box in simulation will displays the
instantaneous values of battery parameters like voltage, current, The overall simulation diagram of the proposed system is
power consumption and SoC. shown in Fig. 2. Here a DC voltage source with a small value of ser-
ies resistance is taken as a battery source. Internal resistor is one of
3. Battery parameters estimation the major factors which decides the lifetime of the battery. The
internal resistance of a battery is directly proportional to the num-
Battery voltage will differ with respect to number of factors. It is ber of charge discharge cycles. i.e., When the battery becomes old
difficult to measure a battery voltage when it is in operating. The the internal resistance also increases. For simulation analysis a
following equations are used to estimate the voltage across the small resistive load is connected in series with equivalent circuit
selected four different batteries correspondingly. of the battery.
Lithium-Ion: Here a controlled voltage source is considered as a battery
source which converts Simulink input signals into an equivalent
voltage source. Input to this voltage source is controlled by the
charge and discharge blocks in simulation. So, the set up can be
simulated by both charge and discharge cycles. The charge block
is designed to select a battery type from the given battery types
1 to 4 respectively in such a way that each battery type can be
charged based on their respective charge equation given by math-
ematical modelling which is given in section III of this paper. A
common discharge equation is modelled for all type of batteries.
The final SoC after the discharge time can be calculated by using
Extended Kalman Filter (EKF).
The capacity and voltage level of battery pack is considered
commonly as 2500mAh, 12 V respectively regardless of the battery
type. For simulation purpose the aforementioned values are
selected. Because it is possible to compare the performance of
the battery if all batteries are having same capacity. Also, instead
of connecting a motor load, a R-L load is connected for this simu-
lation analysis. The value of R-L load is taken from the equivalent
Fig. 1. Block diagram of the proposed system. circuit parameter value of a normal DC motor (R = 3.9 O,
2
R. Sreedhar and K. Karunanithi Materials Today: Proceedings xxx (xxxx) xxx

Fig. 2. Overall simulation diagram of the proposed universal BMS system.

Fig. 3. Equivalent circuit to select the type of battery.

L = 12mH). All batteries are considered as initially having 100% the parameters like battery no load voltage (Em), Battery voltage
State of Charge. at the end (Eb), load power, load current and SoC.
The instantaneous values of four battery parameters namely The voltage and current fluctuations also can be monitored by
battery voltage, current, power consumption, SoC values can be the graphs and display units in simulation. The load voltage of
displayed in simulation. Also, the graphical representation can be the Lead acid battery highly deviates at the time of starting. Also,
viewed in scope. At instant, any one type of battery can be Ni-Cd and Ni-MH batteries possesses almost same characteristics.
simulated. These outputs indicate that the Lithium Ion battery shows best
Fig. 3 shows that the circuit used to select the type of battery. performance characteristics in most aspects.
Here four types of batteries are considered for this simulation anal-
ysis. According to the charge–discharge equation a small change is
needed based on the type of battery. 5. Conclusion
The following parameters of the selected four batteries are sim-
ulated for the same time duration to make the performance The working of the proposed BMS with various battery types
comparison. like Li-ion, Ni-Cd, Ni-MH and Lead-Acid batteries with same capac-
1. Voltage vs Time ity are simulated. The simulation work is carried out with same
2. Current vs Time value of RL load and SoC value at initial stage to compare the work-
3. Power vs Time ing of the proposed Battery Management System. Different instan-
4. SoC vs Time taneous values of battery parameters like voltage, current, power
The graphical representation of the performance characteristics consumption and SoC are analysed using MATLAB/Simulink and
of the four batteries are shown in Figs. 4, 5, 6, 7 respectively. This their performance characteristics has been viewed in graphical
proposed BMS will also help to compare the performance charac- representation. Through this comparison, it is identified that
teristics of different batteries. The comparison among the perfor- Lithium Ion battery possess the best performance in most of the
mance characteristics of the batteries are given in Table 1. Here, battery parameters when evaluated with aforementioned four bat-

3
R. Sreedhar and K. Karunanithi Materials Today: Proceedings xxx (xxxx) xxx

Fig. 4. Characteristics of Li-Ion battery.

Fig. 5. Characteristics of Lead-Acid battery.

4
R. Sreedhar and K. Karunanithi Materials Today: Proceedings xxx (xxxx) xxx

Fig. 6. Characteristics of Ni-Cd battery.

Fig. 7. Characteristics of nickel-metal-hydride battery.

5
R. Sreedhar and K. Karunanithi Materials Today: Proceedings xxx (xxxx) xxx

Table 1
Performance characteristics comparison.

Performance characteristics Li Ion Lead Acid Ni Cd NiMH

Battery Capacity (mAh) 2500 2500 2500 2500
Battery No load voltage (E0)) 12.866 12.866 12.866 12.866
Load Voltage after end of simulation (Eb) 12.77 11.91 12.67 12.67
Load Current (A) 3.38 3.09 3.31 3.31
Load Power (W) 43.16 36.80 41.93 41.93
SoC after 1500 sec 48.94 48.85 47.12 46.34
Voltage and current Fluctuations Low at starting High at starting Low at starting Medium at starting

tery types. This type of common BMS can be made to functionable battery management system.” In Engineers and computer scientists: 2012
IMECS: proceedings of the international Multiconference, 2012, 14-16 March,
in different applications.
2012, Hong Kong. Vol. 2. Hong Kong: Newswood Limited. 2012.
[3] Kumar Bijender, Neeta Khare, P.K. Chaturvedi. ‘‘Advanced battery management
6. Future work system using MATLAB/Simulink.” In 2015 IEEE International
Telecommunications Energy Conference (INTELEC), pp. 1-6. IEEE, 2015.
[4] Habiballah Rahimi-Eichi, Unnati Ojha, Federico Baronti, Mo-Yuen Chow,
The proposed system can be upgraded by using the succeeding Battery management system: An overview of its application in the smart
key points. grid and electric vehicles, IEEE Ind. Electron. Mag. 7 (2) (2013) 4–16.
[5] M. Glavin, W.G. Hurley. ‘‘Battery management system for solar energy
applications.” In Proceedings of the 41st International Universities Power
1. The proposed system is developed by selecting four commonly Engineering Conference, vol. 1, pp. 79-83. IEEE, 2006.
used batteries. It can be further developed to suitable for any [6] Xianzhi Gong, Rui Xiong, Chunting Chris Mi, Study of the characteristics of
battery packs in electric vehicles with parallel-connected lithium-ion battery
type of battery.
cells, IEEE Trans. Ind. Appl. 51 (2) (2015) 1872–1879.
2. The problems associated with practical hardware implementa- [7] Yamin Rami, Ahmed Rachid. ‘‘Modeling and simulation of a lead-acid battery
tion of the proposed system can be investigated and a possible packs in matlab/simulink: Parameters identification using extended kalman
filter algorithm.” In 2014 UKSim-AMSS 16th International Conference on
universal BMS for all type of batteries irrespective of applica-
Computer Modelling and Simulation, pp. 363-368. IEEE, 2014.
tions can be developed. [8] Chun-Sing Cheng, Ricky Wing-Hong Lau, Nand Kishor Rathi, Henry Shu-Hung
Chung, Extraction of intrinsic parameters of lead-acid batteries using energy
recycling technique, IEEE Trans. Power Electron. 34 (5) (2019) 4765–4779.
[9] Asmae El Mejdoubi, Hicham Chaoui, Hamid Gualous, Peter Van Den Bossche,
Declaration of Competing Interest Noshin Omar, Joeri Van Mierlo, Lithium-ion batteries health prognosis
considering aging conditions, IEEE Trans. Power Electron. 34 (7) (2019)
The authors declare that they have no known competing finan- 6834–6844.
[10] Bliss G. Carkhuff, Plamen A. Demirev, Rengaswamy Srinivasan, Impedance-
cial interests or personal relationships that could have appeared based battery management system for safety monitoring of lithium-ion
to influence the work reported in this paper. batteries, IEEE Trans. Ind. Electron. 65 (8) (2018) 6497–6504.
[11] Cheng Lin, Aihua Tang, Wenwei Wang, A review of SOH estimation methods in
Lithium-ion batteries for electric vehicle applications, Energy Procedia 75
References (2015) 1920–1925.
[12] M. Berecibar, I. Gandiaga, I. Villarreal, N. Omar, J. Van Mierlo, P. Van den
[1] Gregory L. Plett, Battery management systems, Volume I: Battery modeling. Bossche, Critical review of state of health estimation methods of Li-ion
Vol. 1. Artech House, 2015. batteries for real applications, Renew. Sustain. Energy Rev. 56 (2016) 572–587.
[2] Chen Chen, Ka Lok Man, T.O. Ting, Lei Chi-Un, Tomas Krilavičius, T.T. Jeong, J.K.
Seon, Sheng-Uei Guan, Prudence WH Wong. ‘‘Design and realization of a smart