electronics

Review
Digital Technology Implementation in Battery-Management
Systems for Sustainable Energy Storage: Review, Challenges,
and Recommendations
Gopal Krishna 1 , Rajesh Singh 1,2 , Anita Gehlot 1,2 , Shaik Vaseem Akram 1 , Neeraj Priyadarshi 3
and Bhekisipho Twala 4, *

1 Uttaranchal Institute of Technology, Uttaranchal University, Dehradun 248007, India

2 Department of Project Management, Universidad Internacional Iberoamericana, Campeche C.P. 24560, Mexico
3 Department of Electrical Engineering, JIS College of Engineering, Kolkata 741235, India
4 Digital Transformation Portfolio, Tshwane University of Technology, Staatsartillerie Rd, Pretoria West,
Pretoria 0183, South Africa
* Correspondence: [email protected]

Abstract: Energy storage systems (ESS) are among the fastest-growing electrical power system due
to the changing worldwide geography for electrical distribution and use. Traditionally, methods
that are implemented to monitor, detect and optimize battery modules have limitations such as
difficulty in balancing charging speed and battery capacity usage. A battery-management system
overcomes these traditional challenges and enhances the performance of managing battery modules.
The integration of advancements and new technologies enables the provision of real-time monitoring
with an inclination towards Industry 4.0. In the previous literature, it has been identified that limited
studies have presented their reviews by combining the literature on different digital technologies
Citation: Krishna, G.; Singh, R.; for battery-management systems. With motivation from the above aspects, the study discussed here
Gehlot, A.; Akram, S.V.; Priyadarshi,
aims to provide a review of the significance of digital technologies like wireless sensor networks
N.; Twala, B. Digital Technology
(WSN), the Internet of Things (IoT), artificial intelligence (AI), cloud computing, edge computing,
Implementation in Battery-
blockchain, and digital twin and machine learning (ML) in the enhancement of battery-management
Management Systems for Sustainable
systems. Finally, this article suggests significant recommendations such as edge computing with AI
Energy Storage: Review, Challenges,
and Recommendations. Electronics
model-based devices, customized IoT-based devices, hybrid AI models and ML-based computing,
2022, 11, 2695. https://doi.org/ digital twins for battery modeling, and blockchain for real-time data sharing.
10.3390/electronics11172695
Keywords: energy storage systems; battery-management system; artificial intelligence; digital twin;
Academic Editors: Stefano Feraco,
blockchain; edge computing
Angelo Bonfitto and Nicola Amati

Received: 29 July 2022

Accepted: 24 August 2022
Published: 27 August 2022 1. Introduction
Publisher’s Note: MDPI stays neutral The United Nations (UN) has emphasized implementing renewable energy for mini-
with regard to jurisdictional claims in mizing carbon emissions. As part of this, renewable energy is being widely adopted by
published maps and institutional affil- many countries. Prior to this, the implementation of ESS has gained wide attention [1].
iations. However, monitoring of these ESS has paved a way for implementing battery-management
systems to detect abnormalities and allow fault detection in ESS [2]. Figure 1 illustrates
the global market for battery-management systems for different applications, in which a
compound annual growth rate (CAGR) of 54.8% is anticipated due to wireless bifurcation
Copyright: © 2022 by the authors. based on connection [3]. Wireless battery-management systems are quickly gaining traction
Licensee MDPI, Basel, Switzerland.
with the need to reduce wires and the usage of the IoT.
This article is an open access article
Additionally, a battery-management system ensures that unusual circumstances
distributed under the terms and
in the architecture of a device will have pre-configured remedial measures. A battery-
conditions of the Creative Commons
management system further validates the proper method for controlling a gadget’s temper-
Attribution (CC BY) license (https://
ature because the temperature has an impact on the power-intake profile. In comparison to
creativecommons.org/licenses/by/
4.0/).
conventional battery technology, lithium-ion batteries charge faster because they have a

Electronics 2022, 11, 2695. https://doi.org/10.3390/electronics11172695 https://www.mdpi.com/journal/electronics

Electronics 2022, 11, 2695 2 of 24

higher energy density and provide a higher power density for longer battery life in a more
compact package [4]. When compared to nickel-based batteries, their self-discharge is less
than half as great, and they do not require prolonged priming (priming is a conditioning
cycle used as a service to improve battery performance during usage or after long periods
of storage) [5]. Li-ion batteries are also becoming more affordable, which makes them an
attractive option for electric vehicles and other applications [6].

Figure 1. Growth of Battery-management systems from 2020 to 2030 [3].

There are various traditional charging methods, such as constant current (CC), constant
voltage (CV), constant-current-constant-voltage (CCCV), and multi-stage constant current
(MCC) charging. CC charging is a charging method that uses a constant current to charge
the battery. The CV charging approach is environmentally friendly for fast charging; the
approach depends upon the battery’s technologies, but such charging harms the battery’s
capabilities. The CCCV charging method is a hybrid strategy that incorporates both
CC and CV [7]. The MCC charging technique consists of several phases with CC, and
the current progressively declines as the terminal voltage approaches a preset voltage
threshold. The battery is charged up to the point at which the conditions of the terminal
are met [8]. The dangers associated with conventional battery charging techniques include
overheating, overvoltage, deep discharge, overcurrent, pressure, and mechanical stress. A
supervisory system that makes sure batteries work properly in the intended application is
necessary to prevent battery failure and reduce potentially dangerous circumstances. A
battery-management system is the name of this monitoring device [9]. Nowadays, there
are many features available in BMS that help the battery operate more efficiently and
safely. Monitoring, battery protection, assessment of the state of health (SOH), state of
charge (SOC), mobile balancing, charging control, and thermal management are a few of
these functions.
A well-designed battery-management system is essential since there are issues about
the safety, dependability, and overall performance of lithium-ion battery systems, particu-
larly in stand-alone systems [10]. Currently, digital technologies such as WSN, IoT, cloud
computing, AI, ML, NN, deep learning, blockchain, big data, cyber security, etc., have
gained attention for real-time sensing, monitoring, fault detection, fault diagnosis, real-time
alert generation, and real-time analytics with prediction.
The cost of storing electricity is still high, and charging a battery fully takes a long
time. The cost of a battery also depends on the components that build up the battery.
Infrastructure for public charging is still lacking. A battery-management system has many
technologies applied to it, but there are still certain restrictions, such as cell balancing,
temperature control, charge control, environmental influence on the system, exact reading
of State of Health (SoH), State of Charge (SoC), and logbook functions, among others [11–21].
Studies have also conducted different systematic reviews of battery-management systems,
Electronics 2022, 11, 2695 3 of 24

such as the [22] study, which carried out an extensive literature review on state-of-health
estimating approaches, and [23] presented a comprehensive review of the most widely used
battery modeling and state estimation methodologies for battery-management systems.
Recently, a study [24] examined the evolutions and problems of cutting-edge battery
technologies and battery-management systems. Moreover, in data-driven electrothermal
models, data-driven technologies such as AI, cloud computing, and blockchain technologies
are examined. From this, it concluded that previous studies have focused on discussing the
review of individual technology implementation in battery-management systems.
With the motivation from the above aspects, this study discussed and reviewed the
progress and implementation of these technologies in battery-management systems, which
empowers an inclination towards industry 4.0. The novelty of this study is that in previous
studies it has been observed that the exploration of digital technology’s impact on battery-
management systems is discussed separately. Even though numerous approaches have
been offered, only a few kinds of literature have attempted a comprehensive assessment of
strategies for monitoring battery-management systems with multiple digital technologies
The authors of this work aim to present clearly and discuss the impact of digital technology
on battery-management systems by combining literature of digital technologies (WSN,
IoT, cloud computing, AI, ML, NN, deep learning, blockchain, big data, cyber security).
From the literature, we have concluded and discussed the vital recommendation that can
be applied as a part of the future research direction. The main contribution of the study is
as follows:
• The basic concept of battery-management systems with different technical terms and
architecture is discussed in detail.
• In order to analyze the impact of these technologies on battery-management systems,
we discussed various digital technologies such as WSN, IoT, Cloud Computing, AI, ML,
NN, deep learning, blockchain, big data, and cyber security for battery-management
systems using tabular and pictorial representation.
• Finally, from the analysis, the article discusses the limitations and presents vital
recommendations for future work.
The structure of the paper: Section 2 discusses an overview of battery-management
systems; Section 3 covers the technologies used in battery-management systems; Section 4
includes recommendations; Section 5 presents the conclusion.

2. Methodology for Review

In this section, we discuss the methods utilized to carry out and check the progress
of wireless technology implementation in battery-management systems. The methods are
provided in the following order: search strategy and selection criteria, data collection and
extraction, and data analysis. This review is largely concerned with the progress of the
various technologies involved in establishing a battery-management system.
The main research question is: “Which technologies are employed in battery man-
agement systems for sustainable energy resources?” Based on the discussed question,
we collected research articles from several databases such as the web of science and Sco-
pus. During the search of articles, the following keywords were primarily applied in the
database. “Wireless monitoring of battery management system”, real-time monitoring of
battery management system “; “IoT and battery management system”, WSN and battery
management system”, cybersecurity and battery management system”, “digital technolo-
gies and battery management system”, artificial intelligence and battery management
system”; “intelligent monitoring and battery management system”, “machine learning and
battery management system”, “deep learning and battery management system”
To decide whether an item should be included or removed from this review, the
following criteria were used.
• We did not select evaluation studies with identical results that used the same data sets,
methods, or algorithms.
Electronics 2022, 11, 2695 4 of 24

• Reviews were not accepted for research that discussed methods but did not conduct
experiments or validate results.
• Diploma theses and dissertations in bachelor’s and master’s programs were not evaluated.
• Scientific articles that were non-peer-reviewed were not reviewed.
The authors have analyzed the articles that were considered for review. Based on
the analysis, this review presents the statistics of different papers that were utilized to
study the different technologies implemented for automated feedback systems. Figure 2
illustrates a pie chart that shows the percentage of the technologies used in this literature
survey. The major parts of the technology reviewed were WSN at 11%, IoT is at 13%, Cloud
Computing at 8%, AI/ML, NN and DL at 37%, Big Data at 4%, Blockchain at 7%, and
Expert System at 4%. Based on this conclusion, this study aims to discuss the progress and
significance of these technologies’ implementation in battery-management systems. This
study considers certain parameters to address the different technologies’ applications with
algorithms, techniques, and advantages.

Figure 2. Bar Chart to show the percentage of technologies in the literature.

3. Overview of Battery-Management Systems

In this section, we discuss the overview of battery-management systems which is
addressed in detail, and a comparison of the environmental and technical efficiency impact
in tabular form is conducted. The battery-management system is a broad area with many
applications (Figure 3) and implementations that are both sophisticated and diverse. An
electrical power garage device’s several battery modules can have their total performance
monitored, managed, and optimized by a battery-management system. In the event of
abnormal circumstances, BMS can detach modules from the apparatus.
a. Structure of Elements and Arrangements
A battery-management system cannot be used as a stand-alone system in a machine
infrastructure. A smart electrical automation machine includes modules for managing
batteries, an interface for connecting the machine to the power grid, packs for storing
energy, and a system for supervising the battery and regulating energy usage [25]. Battery-
management system implementations come in a range of styles, including centralized,
distributed, and modular approaches. Multiple cables link the manipulator unit and battery
cells in a centralized structure. A modular BMS puts together the strengths and weaknesses
of the other two topologies and requires additional hardware and programming work.
Figures 4 and 5 show the battery-management system implementation topology. Lastly,
with a modular topology, a certain battery-moving device corresponds to several operating
devices, but the operating devices are linked [26]. A component-based battery-management
Electronics 2022, 11, 2695 5 of 24

system requires more programming coders and components (hardware), but it simplifies
troubleshooting and optimization for various network topologies.

Figure 3. Applications of battery-management systems.

Figure 4. Connections of a battery-management system and its integration [27].

b. Structure of Battery-management system

The software for managing a battery is created to make multitasking simple since
it is effective at identifying activities fast, as shown in Figure 6 [28]. Previously, it had
been impractical to continue both extraordinary commitments concurrently; one mission
had to be postponed in order to sustain the other mission. Battery-management systems
of the back-state can’t perform multi-tasking at the same time, but the current battery-
management system software architecture offers this capability. Now, new architectures of
battery-management systems represent that they can perform multiple tasks without any
barrier. The initial tasks are defined by the architecture of the battery-management system,
such as reading and calculation of voltage and current, over-current and voltage protection,
reading and calculation of the battery-management system, protective relay actuation, etc.
It must be performed promptly to ensure the safety of the battery-management system. A
Common Microcontroller Software Interface Standard (CMSIS) and a Hardware Access
Electronics 2022, 11, 2695 6 of 24

Layer (HAL) are connected to the microcontroller. For real-time functionalities, a real-time
operating system (RTOS) is introduced into the BMS software architecture [12].

Figure 5. Implementation topology for a battery-management system. (A) Centralized (B) Distributed.

Figure 6. Battery-management system software architecture.

c. Functionalities of Battery-management systems

There are lots of functionalities of a battery-management system. For the capacity
estimation of the battery, it calculates the current, temperature, and voltage percentage.
Temperature control devices can be operated via the IoT using controlling devices. These
measures also aid in extending the life of the battery [29].
d. Impacts of Battery-management systems
There are two types of impacts that can affect battery-management systems: environ-
mental and technical efficiency. This study discusses the electrochemical methods used
by EES structures, including batteries. Power terminals and batteries can be dangerous if
not used correctly. The environmental effects of small-scale power garage facilities were
examined in this paper. This investigation uncovered the causes of animal extinction as
well as soil and water pollution caused by cadmium. This paper also discusses other
compounds used in state-of-the-art electrochemical batteries and any shielding techniques
that could be used to make them secure and with a low environmental impact [30]. These
movements have the potential to cut emissions from software networks and electric flows,
successfully lowering air pollution and enhancing other policies and effects on people.
Electronics 2022, 11, 2695 7 of 24

Our gadgets and technology are powered by electricity, which transforms chemical
energy into electric energy. Electricity can flow to a digital device through a battery’s
electrical circuit, which is formed by the anode and cathode. Batteries must be properly
disposed of once this electric circuit is exhausted, however, tens of thousands of batteries are
thrown away every year [31]. Even while disposing of batteries can seem innocuous, doing
so might have disastrous effects on the environment. Each battery includes dangerous,
lethal, and corrosive elements including lead, lithium, cadmium, and mercury. Here are five
facts regarding batteries you should be aware of if you are worried about their impact on the
environment. Battery-management system concerns related to efficiency, the environment,
and other operational characteristics are presented and summarized in Table 1.

Table 1. Environmental and technical efficiency impact of battery-management systems.

Environmental Impact Impact of Technical Effectiveness

CO2 emissions reduction: Estimation of the real-time SoC:
In addition to adopting a battery-management system to store In addition to implementing a battery-management system to
off-height electricity to meet height demand, we think a fee of store electricity generated off-height to satisfy the demand
40% might cut CO2 emissions. for height.
Benefits of greenhouse gases (GHG): Optimal Charging:
If we employ more battery-management systems and smooth The target is a layout that is mostly based on layout
off-top electricity rather than surges, the benefits of batteries for characteristics and is exceptionally time-efficient, secure,
reducing greenhouse gas emissions may be doubled. and optimal.
Effects of metal depletion: Fast Characterization:
BMS could be an excellent option for charging and discharging
Accurate SOC and SOH characterizations are available from
batteries since it can manage charging and discharging cycles as
BMS. While SOH characterization is mostly focused on the
well as the operating frequency. On compounds with high
range of cycles of data, SOC models its conclusions using a
environmental and power impacts, this substance has a
single full cycle of data.
considerable impact.
Impacts of temperature regulation: Self-Evaluation:
A BMS may be used to control two separate temperatures: the BMS represents intricate battery functions, such as capacity,
electrochemical response temperature and the ambient power, hysteresis effects, and temperature effects using
temperature of the battery. mathematical formulae.

4. Technological Review of Battery-Management Systems

In this section, we discuss the distinct digital technologies that have been identified
through the analysis. Here, the individual technologies of battery-management systems
are addressed in detail. To show the representative battery operating states in electric
vehicle (EV) applications, battery modeling and the assessment of battery internal states
and characteristics initially play a significant role. After identifying these crucial factors,
a suitable battery charging strategy may be created to safeguard the battery, increase
energy conversion efficiency, and prolong battery life. It is challenging to ensure modeling,
estimating, and charging performance in actual applications, which might differ from test
settings or in a worst-case scenario. To tackle this difficult problem, it is necessary to study
the constraints or to establish a confidence interval for the methods that are described [32].

4.1. WSN in Battery-Management Systems

A battery pack with several separate cells contains many wire terminations that can
fail. To address the wiring issue, a wireless battery-management system relies on the
ZigBee communication protocol with voltage, temperature, and SOC sensors [33]. The
Battery-management system monitors runtime statistics, keeps a data log, and manages
load switching between photovoltaic power systems and utility. The current and voltage
sensors are connected to the FPGA through an Analogue-to-Digital Converter [34].
Table 2 gives a comparative evaluation of the evaluation research primarily based
totally on battery-management systems, along with sensors and a set of rules with the
Electronics 2022, 11, 2695 8 of 24

prevailing study. Little research offers the dialogue of wi-fi sensor community technology
as much as LoRa technology and wi-fi information acquisition. However, this text gives a
complete dialogue of lots of wireless sensor communities with information and commu-
nication technology (ICT), along with IoT and battery-management systems [35–38]. The
article also depicts the notion of IoT implementation in a battery-management system using
a wireless sensor network. Finally, this essay discusses the benefits and ideas for improving
battery-management systems using an advanced methodology and advises building the
architecture in WSN using 5G technology.

Table 2. A detailed survey of WSN in battery-management systems.

Ref. Objective Sensor Used Algorithm Used Advantage

The study describes a wireless battery
Battery control device that uses both the wi-fi
Enhancement of Voltage, current,
[33] electrochemistry architecture and the Zigbee
battery life temperature, and SOC
(lifepo4) conversation protocol to connect with
other devices.
The voltage and current sensors are
connected to an FPGA using an
Analog-to-Digital Converter (ADC).
Smart energy
Improvement of Current sensors, voltage An FPGA and strength control and
[34] management system
energy efficiency sensors, CT and PT sensors tracking center are two examples of
algorithm
equipment connected to a network
using the wireless communication
protocol Zigbee.
A hybrid-strength garage machine can
Increase the battery E-power management aid in preventing damage to the Wi-Fi
[35] Wireless sensor
lifetime algorithm. sensor’s typical battery during the
process of rapid discharge.
The study discussed presents a
Fulfillment of
conceptual design for a LoRa-based
[36] battery-based power Current and voltage sensor. -
Private Server Network-mode battery
demand
energy storage monitoring system.
The gadget that is the subject of the
study discussed helps to avoid
LSTM-based battery sudden battery failure and poor
[37] Current sensor -
voltage prediction functioning and is beneficial in
speeding up the repair and lowering
restoration expenses.

4.2. IoT in Battery-Management Systems

The state-of-charge parameters of a battery can be measured using different techniques
and this state of charge measures the amount of charge it can store or can show the current
charging status of the battery [38]. Overcharging the battery will not be a possibility if the
percentage is calculated correctly. However, because each has its restrictions, there may
be times when the battery is overcharged. The alternators will include a built-in voltage
regulator that can deliver steady voltage even when charging automobile batteries. Failure
might have several dangerous repercussions. Gases like hydrogen and oxygen, among
others, may be released as a result of overcharging. They are created by the electrolyte’s
aqueous solution evaporating [39]. The study discussed the progress of smart cells and
battery-management systems from various points of view using the possible integration
of sensing techniques, design, and innovation in battery-management systems [40]. The
study examines sensor noise estimation methodologies and error boundaries and finishes
with a look ahead at the research that will be required to enable quick charging, battery
repurposing for degradation prediction, grid energy storage, and defect-recognition [41]
by thoroughly analyzing the extant literature on the status of health estimating methods.
Electronics 2022, 11, 2695 9 of 24

There are two sorts of estimation methods: experimental and model-based estimation
approaches. In this work, thorough literature analysis and the methods for assessing the
health condition of the battery are presented in greater detail, and their respective merits
and weaknesses are evaluated [42].
The physical and digital embodiments of a battery interact closely in this cyber-
physical system, allowing for smarter control and longer battery life. The state-of-the-art
in-vehicle diagnostic tools, battery modeling, data-driven modeling methodologies, and
how these aspects might be merged into a framework for generating a battery digital twin
are all presented in these viewpoints [43]. Fiber optic sensors are being used more and
more in battery monitoring as a result of the growing demand for advanced battery control
structures with accurate reputation estimations. The purpose of this evaluation is to include
the advancements that have made it possible to use measurements of battery internal
parameters, along with the nearby pressure, strain, temperature, and refractive index for
renowned processes, as well as outside dimensions, along with the temperature gradient
and a gasoline sensor, to detect thermal runaway. Fiber optic sensors are characterized in
terms of battery structures of three different sizes including grid-scale battery structures,
battery packs for heavy-duty electric trucks, and electric cars [44].
The large current peaks during the data transmission method are one feature of the
LoRa technology. Thus, a hybrid energy storage device is implemented in preventing
the typical battery of a wireless sensor from degrading during rapid draining [35]. The
study discussed and detailed the abstract approach of employing a camera server network-
mode LoRa camera-powered energy-storage observation system [36]. The study discussed
offers a prediction approach for forecasting the subterranean management system’s battery
capacity evaluation. The technology guards against the improper operation and unexpected
battery failure [22]. With a 5G advanced battery-management system structure, the classic
BMS mostly uses comprehensive laboratory data to calibrate parameters, which makes it
challenging to satisfy the needs of extreme precision and real-time performance. The study
described the abstract design of the camera server network using a LoRa-based battery
energy-storage observation system. The trend for the future is a fact-based architecture of
personalized battery control systems, as seen in Figure 7.

Figure 7. Advanced battery-management system architecture with 5G.

Electronics 2022, 11, 2695 10 of 24

Table 3 depicts earlier research that used IoT in a wireless sensor network. The prior
research included in the table was largely concerned with error detection, fault tolerance,
and increasing energy density. The integration of IoT and battery-management system is
used to obtain the most efficient and sustainable solution.

Table 3. A detailed survey of IoT in battery-management systems.

Ref. Objective Sensor Used Algorithm Used Advantage

The percentage error is less using the
NN (Neural Network)
[38] Calculate the SoC Current sensor NN algorithm, as compared to
Algorithm
without a NN algorithm.
Calculate the Soc Temperature Establishment of a fault
[39] Current sensor
and SoH calculation algorithm diagnosis system
The work on the optical FBG sensor
yields some positive results and
SoC, SoH progresses of Estimation and control
[40] Current sensor demonstrated its ability to assess
smart cell algorithms
surface/inner pressure and
temperature in situ and operando.
Increasing energy density and
[41] Battery SOH estimation Current sensor Data optimization
associated vehicle range.
The method utilized to evaluate the
High-precision Adaptive filtering or
[42] Battery SOH estimation battery health level is based on
current sensor data-driven algorithm
real needs.
Many scientific works use a
combination of spectroscopic,
Estimation of fast Hall Effect sensor or k-nearest neighbors’
[43] physical, and electrochemical
charging algorithm Shunt resistor algorithm
methodologies to improve the
understanding of how batteries work.
The predicted sensing system costs for
Equivalent-circuit- standard fiber optic sensors, and one
Estimation of a sensing Temperature, low-cost
[44] model-based SOC of the restrictions in their practical
system for optical fiber. fiber optic sensors
estimation algorithms deployment into batteries is the
expensive interrogation cost.

4.3. Cloud Computing in Battery-Management Systems

Bluetooth 4.0 module usage, and subsequently Bluetooth network protocol usage,
results in larger battery energy savings, that is, a longer battery lifespan in all circumstances,
when contrasted with the results of the energy consumption calculation performed using
the XBee ZigBee antenna [45]. This is due to the module lacking different energy values
when transmitting and receiving data, and also lower module expenditure values when
active in contrast to ZigBee and Wi-Fi XBee antennae. End sensing, edge computing, cloud
computing, and a knowledge repository are all part of a layered cloud-to-things system,
such as a cloud-based battery management solution with status estimation capabilities.
Data visualization from the cell-battery vehicle transportation system at various scales can
be conducted. A hierarchical functional display is created using the Cyber Hierarchy and
Interactional Network (CHAIN) architecture [46].
In order to raise the processing power and data storage capacity of cloud computing,
the study offers a cloud-based battery-management system. All battery-related data is
monitored and wirelessly uploaded to the cloud via the Internet of Things to create a digital
replica of the battery system. The data is then analyzed by battery diagnostic algorithms,
which expose the battery state and aging window [47]. This is also the first study to
show that the battery’s capacity and power degrade concurrently. Figure 8 shows the
architecture of a cloud-connected battery-management system. The system’s functionality
and methods of diagnosis were tested with prototypes of a cloud battery-management
system in the field.
Electronics 2022, 11, 2695 11 of 24

Figure 8. The architecture of a cloud-connected battery-management system.

Techniques of ML will be ready in the future based on data received from cloud-
based battery-management systems for exact lifespan forecasting and system improve-
ments [48]. Table 4 provides the detailed function of cloud computing in wireless sensor
networks. The prior research included in the table contains the different types of sensors
with display systems, algorithms, and improved scheduling services for better battery
energy management.

Table 4. A survey on cloud computing in battery-management systems.

Ref. Objective Sensor Used Display System Algorithm Used Advantage

Parameter A cloud-based digital twin for
identification, battery systems improves the
Calculation of Numerous display meta-heuristics, computing power, data storage
[45] Current sensor
sleep-time types SOCs, cloud-suited capacity, and dependability of
battery diagnostic the battery-management
algorithms. system.
Air, humidity, Improved battery energy
Functions of state Numerous
[46] temperature, MQ-2 NA savings offered by the Bluetooth
estimation displays
gas, smoke flame network protocol.
For managing battery energy,
the intelligent scheduling
New intelligent
[47] Current sensor - IIS, PVE Algorithm service charging model is more
BMS
effective than the conventional
scheduling service.
A framework for a cloud-based
Monitoring the AEKF, PSO battery-management system is
[48] Current sensor -
battery cells algorithm proposed that makes use of an
end-edge-cloud architecture.

4.4. Big Data in Battery-Management Systems

Cyber-Physical System (CPS) technology and battery big data platforms are the foun-
dations of the study’s uniquely flexible and dependable battery management strategy. The
proposed GRNN algorithm and cross-validation technology-driven data cleaning tech-
Electronics 2022, 11, 2695 12 of 24

nique may effectively fix corrupt data in the cloud battery database under temperature
changes [49]. A machine learning-based data cleaning technique is proposed that is rele-
vant to the properties of huge data from electric car batteries. The work presented a deep
learning-enabled lithium battery model that can adapt to a big data environment.
The data cleaning method, which is based on a machine-learning algorithm, produces
favorable results when a terminal voltage is absent, for example, when the mean absolute
percentage error of filling is less than 4%, which has a greater impact on improving the
overall quality of the dataset [50].
Information is gathered using big data technologies, which include N.N., machine
learning, and deep-learning algorithms. However, after going through the data cleaning
procedure, one can obtain the most accurate data, which is crucial for the battery’s lifespan.
Table 5 gives a thorough analysis of big data in battery-management systems.

Table 5. A detailed survey on big data in battery-management systems.

Ref. Objective Sensor Used Algorithm Used Advantage

ELM, deep learning, Accurately restore the
SoC error
[49] Current Sensor conventional cloud battery database’s
estimation
data mining. corrupt data.
The method for cleansing
Simulation of SVR, deep learning,
data produces positive
[50] the battery Current Sensor machine learning,
outcomes using the ML
characteristic neural network.
algorithm.

4.5. AI—ML, NN, and Deep Learning in Battery-Management Systems

We will see an increased role of battery-management systems with next-generation
batteries. A model that evolves as it investigates a chemical area may be created using a
machine learning inverse design, allowing for the expansion of a model in areas of extreme
uncertainty and the identification of molecular space regions with desired attributes as a
role of composition. The challenges of modeling the links between material properties and
intricate physical parameters have been handled in recent years by ML approaches [51].
For cell-level capacity estimation, a deep-learning technique using deep convolutional
neural networks (DCNN) is used, which is based entirely on the current, voltage, and price
capability measurements throughout a half-charge cycle. With these aims in mind, this is
one of the first attempts to use deep learning to estimate the capacity of a Li-ion battery
online [52]. The major focus of this research is the creation of new deep learning (DL) with
a SOC estimation model for safe renewable energy management (DLSOC-REM) for HEVs.
Since battery damage from excessive charging and discharging is unavoidable, the BMS
should provide an accurate SOC calculation [53].
Today’s technology concentrates on the creation of clever algorithms for estimating
inaccuracy, SOC, SOE, SOH, centered structure, access characteristics, advantages, and
downsides. According to the study, clever algorithms have demonstrated improved overall
performance in terms of precision, flexibility, robustness, and battery efficiency when
using an estimate [54]. Because of their high electricity and energy density, lithium-ion
batteries are widely employed in the automotive sector (in electric motors and hybrid
electric motors). However, this creates more challenging protection and dependability
scenarios that necessitate the advancement of cutting-edge battery-management systems.
A BMS ensures a battery’s safe and reliable functioning and understands that it requires
solving a model. Modern BMSs, on the other hand, may not be able to deliver accurate
results at real-time prices and some points, in a vast operation range, thus they are not
designed to the specifications of the automotive sector [55].
The study looks at how battery-management systems have changed over time and
suggest a tiered design architecture with three progressive levels for improved battery
management. The algorithm layer aims to give full knowledge of the battery, while the
application level provides a secure and effective battery method through correct supervision.
Electronics 2022, 11, 2695 13 of 24

The foundation layer concentrates on the system’s theoretical underpinnings and physical
foundations [23]. By thoroughly analyzing the extant literature on the status-of-health
estimating methods, the study discussed seeks to act as a valuable resource for scholars
and practitioners. There are two types of these techniques: methods of estimation based on
experiments and models [56].
One study implemented a battery life forecast model that is geared towards operational
battery management optimization. The methodology has been developed for lithium-ion
(Li-ion) cells to take into account five operational factors: discharging and charging currents,
maximum and minimum cycling constraints, and operating temperature [57]. The proposed
SoC and SoH calculations are utilized to build an algorithm that can accurately estimate the
battery state. The SoC may be appropriately computed by applying the battery efficiency
to the open circuit voltage to minimize the initial fault of the Coulomb counting method
(CCM). The internal resistance of a battery increases while charging and discharging, while
the CC charging time decreases [58].
This work calculates the SoC of Li-battery systems for any applications like EV using a
variety of ML techniques such as support vector machines (SVM), artificial neural networks
(ANN), linear regression (LR), ensemble bagging, and Gaussian process regression (GPR)
(Figure 9). The model’s error analysis is used to optimize the battery’s performance
parameters. Finally, performance indexes are used to compare all six algorithms [59].
Energy storage systems (ESSs) need a battery-management system algorithm that can
control the battery’s condition since getting older causes a battery’s internal resistance to
increase and its capacity to diminish. To manage the battery status, this research presents a
battery-efficiency calculation formula. The proposed formula for calculating the battery
efficiency takes into account charging current, charging time, and battery capacity [60].

Figure 9. ML approaches in battery management.

The multipurpose control and planning (MCP) approach using three indices to define
the best BESS location and category: BESS capacity, OLTC and SVR tap operations, and
Electronics 2022, 11, 2695 14 of 24

PVP curtailment. In the simulated case study, BESSs were used for power smoothing
of the substation/PVPs and RPF prevention at the substation, simulating the needs of
Japanese power utilities [61]. The review begins with an introduction to machine learn-
ing’s conceptual framework and general application process, followed by a review of ML
progress in both enlightening battery material design and precise battery state estimation.
ML is thought to help accelerate the use and improvement of lithium-ion batteries on a big
scale [62].
The method for calculating the necessary parameters depends on the simulation of
the temperature from the battery measurements presented in the study discussed. A set of
rules first looks at the relationship between current steps and the terminal voltage that was
determined, using the assumption that a certain load is present in both the present and the
past. Second, by combining the Gauss-Newton approach and particle swarm optimization,
the first-predicted parameters from the primary methodology are appropriate for the
dimension data. Then, it is estimated how each simulation parameter depends on the
battery temperature and market reputation [63]. The five most extensively researched types
of device-learning techniques for estimating battery SOH are thoroughly examined. The
ML-assisted SOH estimation strategies are evaluated from three angles: the assessment
performance of several procedures using five performance indices, and training modes
based entirely on feature extraction and choice strategies [64].
In order to test lithium batteries, the educational data is divided using a special
evolutionary algorithm based entirely on the fuzzy C-approach clustering method. With
the help of the clustering findings, the antecedent parameters and the model’s topology
are found. The parameters are extracted using the recursive least-squares method, and
the antecedent and subsequent portions are then optimized simultaneously using the
backpropagation learning method. Studies have shown that the suggested estimator
is accurate and performs better than those produced using traditional fuzzy modeling
techniques [65]. Table 6 makes a distinction between different methodologies based on the
concept, kind, structure, and performance evaluation. Smart grids (SGs) and electric cars
are two examples of high-power applications that employ lithium-ion battery packs and
need a battery-management system.

Table 6. A detailed survey on AI-ML-NN and deep learning in battery-management systems.

Ref. Objective Sensor Used Algorithm Used Advantage

Ml algorithm, support
vector regression synced Active learning in the domain of
cross-validation simplex objective functions may lead to a
[51] SoC calculation Current sensor better knowledge of the appropriate
algorithm, and ANN
algorithm are all examples rewards to pursue when performing
ML.
of algorithms.
The proposed DL technique
The adaptive-observer
Accuracy in SoC algorithm, SVM, RVM, demonstrates significant efficiency in
[52] Current sensor capacity estimation, highlighting that
and SoH KNN regression, and
lazy-learning algorithm. a method is a suitable tool for online
Li-ion battery health management.
Bmo algorithm, SoC-rem
The dlsoc-rem technique can be used
algorithm, hybrid
[53] SoC estimation Current sensor to estimate SoCs in an accurate and
metaheuristic optimization
timely manner.
algorithms
The future of data-driven and
[66] Safety of battery Current, stress, fiber, Intelligent algorithms intelligence-based battery
Bragg grating,
management is examined.
Electronics 2022, 11, 2695 15 of 24

Table 6. Cont.

Ref. Objective Sensor Used Algorithm Used Advantage

Accurate SoC and SoH estimations
were proposed by applying battery
efficiency to the estimation process.
SoC and SoH Ocv, ccm, and proposed The estimated SoC and SoH were used
[58] Current sensor
estimation soc algorithm to improve not only the performance
of the BMS but also the battery safety
via a fault diagnosis algorithm with
accurate SoH estimation.
ANN, SVM, LR, Gaussian Analyzing the voltage and current in
[59] SoC estimation Current sensor process regression. the SoC estimation.

Adaptive filtering or This method is chosen to evaluate the

[60] SoH estimation High-precision current battery health level based on real
sensor data-driven algorithm demands.
SVM, ANN, linear, GPR,
SoC, voltage, and ensemble boosting, An analysis is conducted based on
[61] Current sensor
current estimation voltage and current.
ensemble bagging
Ml algorithms, clustering
algorithms, naïve Bayes, ML can be used for knowing the
[62] SoC, SoH estimation Current sensor
logistic regression, linear battery state.
regression

The deep-learning Calculations and modern material

[63] SoC, voltage, and Current sensor design demonstrate improved battery
current estimation algorithm, performance.
The new method shows the input
[64] Accuracy estimation High-precision current LR, KNN, SVM, ANN, and
measurement sensors EL ALGORITHM characteristics and the estimation
accuracy.
The least-squares
The learning mechanism works using
algorithm, subtractive
High-precision Hall the genetic fuzzy-clustering technique
SoC and SoH clustering, fuzzy clustering,
[65] estimation current sensor, and the direct search algorithm
current-sensor direct search algorithm,
leveraged to realize the
genetic algorithm,
antecedent parameters.
and ANN
BMS algorithms, optimal
Battery impedance, capacity
charging algorithms,
Charging and estimation, optimal charging
[49] Current sensors constant-current charging
discharging estimation strategies, and strategies to evaluate
algorithm, genetic
battery-management systems.
algorithm, BFG algorithms
MD, ANN, SVM, KNN, RF, This worked for the safety of the
[67] SoH estimation Current sensor ERT, DNN, SVR, KRR, PLS battery of the EV.
Swarm optimization
Performance estimation algorithm, kernel-based Compared to other models, the CNN
[68] Current sensor
of model learning algorithm, model performs better.
gradient descent algorithm,
A fast recursive algorithm,
Model size and computational cost are
Cost estimation using adaptive filtering
[69] Current sensor much lower than those of the original
models algorithms, least-squares
convolutional neural network model
algorithm
SVM, ANN, LR, GP Probability distribution has improved
[70] SoC, SoH estimation Current sensor and ANN the state-of-charge estimation.

A battery-management system requires a combination of software and hardware to

complete functions such as battery-state estimation, problem detection, monitoring, and
control [71]. The most recent research on the use of ML in battery development, involving
electrodes and electrolytes, is summarized. Meanwhile, battery state prediction is available.
Finally, numerous present issues are discussed, as well as a methodology for addressing
them in the future development of ML for rechargeable lithium-ion batteries [67]. To
increase the resilience and the projected 1D CNN network’s accuracy, the partial hyperpa-
rameters of the neural network are optimized by employing a weighted particle-swarm
Electronics 2022, 11, 2695 16 of 24

optimization method that is linearly decreasing. To account for the unpredictability of

charging behavior in practice, the 1D CNN model employs random sections of the charging-
voltage curve, differential charging-voltage curve, and charging-current curve as input data.
LDWPSO is also utilized to optimize the fundamental hyperparameters of the 1D CNN
model [68]. The article provides a novel framework for building compact CNN models on
a limited dataset with better-estimated performance that incorporates the ideas of transfer
learning and network pruning [69]. According to the findings, if a DNN has enough retired
layers, it can anticipate the SOC of unknown driving cycles during training. EVs and smart
grids are two examples of high-power applications that frequently employ lithium-ion
battery packs and a battery-management system which requires software and equipment
combined to complete duties such as battery state estimation, problem discovery, monitor-
ing, and control. The study discussed presents a thorough examination of the current level
of ML approaches to battery-management systems. It creates the difference between the
techniques based on concept, type, structure, and evaluation of performance [71].

4.6. Expert (Recommendation) Systems in Battery-Management Systems

A battery-powered device’s safety, effectiveness, and dependability are all guaranteed
by a device that controls the battery or battery-management systems. Numerous studies on
battery-management systems have been conducted over the years, and they have largely
improved the safety, effectiveness, and dependability of battery systems. However, there
are still issues that need to be resolved. In this article, we outline such issues and discuss
potential solutions. The difficulties of creating a battery-operated gadget that can be used
in upcoming destiny projects are discussed in this article. It also talks about some of the
responses that were given [4]. There are certain projects where you may find prototypes of
various players, usually from universities or government programs. Additionally, there
are a few duties for businesses that are participating in the market but are still in the
prototype and market testing phases. We have to decide whether to publish asynchronous
conversation modes based entirely on open specs and the well-known example of XML
because it is difficult to find globally general specifications and requirements for data
exchange, particularly with intelligent grid systems, public transportation systems, control
systems, and batteries inside the power industry [72]. The studies in Table 7 address the
sensors and advantages of implementing expert systems for battery-management systems.
The various algorithm-like hybridized intelligent algorithms enable users to recommend
a cost-effective and energy-saving strategy that can be executed in the customization of
battery-management systems.

Table 7. A survey on expert systems in battery-management systems.

Ref. Objective Sensor Used Algorithm Used Advantage

Battery impedance, capacity
Precise
BMS, optimal charging, estimation, optimal charging
characterization, and
[4] Current Sensor constant-current charging, BFG strategies, and strategies to
reliable battery
algorithms evaluate
estimation
battery-management systems.
Hybridized intelligent
Precise characterization A complete examination,
Temperature, Current algorithms, newly designed
[72] and reliable battery evaluation, and advice for
thermal sensor algorithms for eight-cell
estimation automotive engineers.
battery packs

4.7. Digital Twins in Battery-Management Systems

The materials and management techniques employed determine the lifespan of li-
battery-powered equipment. A digital twin of a battery is a digital variant of a battery
that interacts intimately with a cyber-physical system, allowing for greater control and a
longer lifespan [43]. Monitoring of the battery-management system is carried out to ensure
the greatest level of reliability and safety. The meta-model, which permits the creation of
Electronics 2022, 11, 2695 17 of 24

domain-specific models, reflects the architecture as seen in Figure 10. The three basic layers
of the idea are hardware, twin, and service level [73].

Figure 10. Layers of Architecture.

Based on the digital twin, we can conclude the many solutions for battery-management
systems, such as real-time state estimation, digital modeling, dynamic charging control,
dynamic equalization control, dynamic thermal management, etc., [74]. For developing
the digital twin of a battery-management system, all the relevant data should be processed
and stored on a cloud platform. The stage of each battery cell can be shown by the digital
twin [75]. The studies in Table 8 address the study of digital twins in battery-management
systems using IoT and cloud technology, and by inserting the SOC and SOH into the system
for the digital twin, we can fit battery models to the data [76].

Table 8. A survey on digital twins in battery-management systems.

Display
Ref. Objective Sensor Used Algorithm Used Advantage
System
Standard procedure SOC, SOAP,
Hall Effect and Intelligent control of battery systems
[43] on the database - CC-CV charging
other sensors using the ML approaches.
Management algorithm
digital twin Multi-discipline The proposed design provides a
[73] Integrated Sensor -
architecture for BMS algorithm roadmap for the life cycle of a BMS.
Application of Least squares Summarizes recent methods of
[74] RFID, sensors Soh display
digital twin in BMS algorithm research for future enhancement.
Open-loop,
Measurement of Voltage, current, BMS was developed based on cloud
[75] Web front end model-based,
SoC, SoH. and temperature computing and IoT
AEHF.
Voltages, least-squares,
Inserting the SoC, Stored data shows the state of the
[76] temperature, and Web front end Levenberg–
and SoH in the cloud battery with advancements.
current Marquardt

4.8. Blockchain in Battery-Management System

Nowadays, a limited range of battery life is the major problem for electric vehicles.
To address this, we can swap the batteries but there are few authorized battery-swapping
stations. In this situation, a strong battery-management system or battery-swapping sys-
tem (by station or driver) based on blockchain is required which can be continuously
monitorable [77]. Blockchain has found wide use in the energy sector because of its un-
derlying qualities of anonymity, decentralization, transparency, and dependability [78].
An upcoming battery-management system can be managed by critical activities and tasks
Electronics 2022, 11, 2695 18 of 24

involving the management of the battery, recovery, firmware security checks, patch gen-
eration, etc., [79]. Blockchain generation is used to defend an IoT-enabled battery control
gadget from undesirable cyberattacks and make certain verbal exchanges and statistics
security [80]. The studies in Table 9 address the sensors, algorithms, and advantages of
implementing blockchain for battery-management systems.

Table 9. A survey on Blockchain in battery-management systems.

Ref. Objective Sensor Used Algorithm Used Advantage

Consensus algorithm The user received either a battery or a
[77] Increase the reliability -
(Hashing). charge/swap station.
Charging scheduling The future generation of distributed
[78] Security enhancement - algorithm, consensus energy solutions can be designed
algorithm. using blockchain.
Reverse engineering for Embedded
Firmware checks and recovery are
[79] security check Current, voltage sensor battery-management
possible by blockchain.
and recovery system algorithms.
Leader election
Current Enhancing cybersecurity of the wbm
[80] Enhancing the Security algorithm, on-board
Sensor in blockchain-based IoT network
control algorithms

4.9. Cybersecurity in Battery-Management System

The topic of cyber-physical security of battery-energy-storage systems is complicated
because it not only involves information security principles but also calls for bridging the
knowledge gaps between the effects of cyberattacks on industrial control systems [81].
Due to the constant network connectivity of IoT devices, there is an increasing risk of
cyberattacks. There are lots of threats that can be possible such as unauthorized software
updates, unauthorized access, Man-in-the-Middle attacks, insecure network protocols,
unauthorized cloud access, SQL Injection, etc., [82]. For the detection of attacks, there are
lots of methods such as manipulated system command attack detection, battery attack
detection, training-set attack detection, etc. Protected IoT-cloud platforms will be made
available to BMSs to encourage better cybersecurity and spur the adoption of Li-ion battery
systems in cyber-physical settings [82]. Table 10 gives a comparative study based on the
cyber security based on the battery-management systems.

Table 10. A survey on cyber security in BMS.

Ref. Objective Sensor Used Algorithm Used Advantage

SoC estimation, EMS
To enhance the risk assessment of these
Cyber-attacks and algorithms, voltage-based
[81] Current sensor assets, threat models for BESS must be
prevention charge equalization
further developed.
algorithms
IoT-cloud platforms will be applied to
Health monitoring, IoT BMSs to increase cybersecurity and
Cyber-attacks and
[82] Current sensor network, SHA256 hashing accelerate the proliferation of Li-ion
prevention
algorithm battery systems in cyber-physical
environments.
Cyber-attacks and Battery SE such as SOC and SOH are
[82] Current sensor ML and ANN
prevention forecasted using ML and ANN.

5. Recommendations
In the above, we have detailed and discussed the significance of battery-management
systems and the integration of digital technologies in battery-management systems for
achieving digital-based monitoring with advanced features. Based upon the analysis,
we have discussed the challenges and suggested further recommendations for future
enhancement below.
Electronics 2022, 11, 2695 19 of 24

• Wide adoption of customized IoT sensor-based devices in the monitoring and ob-
taining of real-time data of battery-management systems [4]. Customization allows
the user to include features that are very significant for their battery-management
system. In addition to this, researchers need to adopt the materials in developing
IoT devices for making them resistant to the environmental conditions of the battery-
management system.
• The large amount of sensor data that is generated through IoT sensor-based devices
can be effectively utilized for the prediction of charging and discharging time, SoC,
SoH, aging, etc., [72]. Researchers need to focus on creating a hybrid model that
can detect different anomalies under different environmental conditions with a high
accuracy rate. To achieve this, AI-based computing units should also be integrated
into IoT-based devices.
• Edge computing in battery-management systems is implemented limitedly. Edge
computing needs to be integrated into IoT-based devices for processing the obtained
sensor data at the edge network itself [43]. In addition to this, AI models can be loaded
into the computing unit to perform prediction analytics on real-time data. This indeed
can empower the enhancement of the latency and minimize the load on the server for
performing the prediction.
• The digital twin is an emerging technology, and the integration of this technology will
empower the creation of a replica of a battery-management system under different
environmental conditions with customized features [73]. Few studies have already
introduced state estimation and cloud-inspired equalization for batteries. Moreover,
this study also enabled upgrading of the route of the battery with full life-cycle data.
• Blockchain technology in battery-management systems enables the securing of data
and also connects different entities in the distributed network for real-time monitoring
of the health of the battery-management system from any location [74]. In addition to
this, blockchain enables the removal of the barrier of accessing and sharing data of
battery-management systems among manufacturers, electricity consumers, and power
grid operators.
• The evolution of big data with ML and DL has overcome the challenges of complicated
modeling and insufficient data-feature extraction, making the extraction and life
prediction of lithium battery health assessment features practicable [75]. Big data
examines the effects of important elements on the use of batteries: current, voltage,
and temperature. It focuses on the impact of charge-current fluctuations, high charge
cut-off voltage, and temperature on the stability of lithium batteries based on an
investigation of batteries of various materials.

6. Conclusions and Future Scope

Battery-management systems have gained significant attention due to the wide adop-
tion of renewable energy generation for sustainability. The health monitoring of batteries is
crucial for reliably storing energy. Along with this, the evolution of digital technologies has
proven to be effective for monitoring the physical environment from any location. Based
on this motivation, this article discussed the significance of battery-management systems
and further discussed the implementation of these technologies in battery-management
systems. From the review of different articles, it can be concluded that battery health
estimation methodologies have been developed for monitoring the remaining capacity
and energy estimation, capacity prediction, life and health prediction, and alternative
essential indicators connected to battery balance and thermal management. Finally, this
article suggests recommendations such as edge computing with AI model-based devices,
customized IoT-based devices, hybrid AI models and ML-based computing, digital twins
for battery modeling, and blockchain for real-time data sharing.

Author Contributions: Conceptualization, R.S. and G.K.; methodology, A.G.; formal analysis, S.V.A.;
data curation, S.V.A.; writing—original draft preparation, G.K.; writing—review and editing, N.P.
Electronics 2022, 11, 2695 20 of 24

and B.T.; visualization, R.S. and A.G.; funding acquisition, N.P. and B.T. All authors have read and
agreed to the published version of the manuscript.
Funding: The APC was funded by Tshwane University of Technology, South Africa.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data sharing not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations

ADC Analog-to-Digital Converter

AI Artificial Intelligence
ANN Artificial Neural Network
BMO Barnacles Mating Optimizer
BMS Battery-management system
CAGR Compound Annual Growth Rate
CC Constant Current
CCCV Constant-Current-Constant-Voltage
CMSIS Common Microcontroller Software Interface Standard
CNN Convolutional Neural Network
CPS Cyber-Physical System
CT Current Transformer
CV Constant Voltage
DL Deep Learning
ESS Energy Storage Systems
EV Electric Vehicle
FPGA Field Programmable Gate Arrays
HAL Hardware Access Layer
ICT Information and Communication Technology
IoT Internet Of Things
KNN K-Nearest Neighbor
LDWPSO Linearly Decreasing Weight Particle Swarm Optimization
LI Lithium—Ion
L0RA Long Range Radio
LSTM Long Short-Term Memory
MCC Modern Constant Current
ML Machine Learning
NN Neural Network
OCV Optical Character Verification
PGD Projected Gradient Descent
PV Photovoltaic
REM Energy Management
RTOS Real-Time Operating System
RVM Reverse Vending Machine
SGs Smart Grid
SHA Secure Hash Algorithm
SoC State of Charge
SoD State of Discharge
SoE State of Emission
SoH State of Health
UN United Nations
Wi-Fi Wireless Fidelity
WSN Wireless Sensor Network
Electronics 2022, 11, 2695 21 of 24

References
1. Owusu, P.A.; Asumadu-Sarkodie, S. A review of renewable energy sources, sustainability issues and climate change mitigation.
Cogent Eng. 2016, 3, 1167990. [CrossRef]
2. Dai, H.; Jiang, B.; Hu, X.; Lin, X.; Wei, X.; Pecht, M. Advanced battery management strategies for a sustainable energy future:
Multilayer design concepts and research trends. Renew. Sustain. Energy Rev. 2020, 138, 110480. [CrossRef]
3. Battery Management System Market Research Report: By Battery Type, Connectivity, Topology, Vertical—Global In-
dustry Analysis and Forecast to 2030—Global Industry Analysis and Demand Forecast to 2030. Available online:
https://www.researchandmarkets.com/reports/5010717/battery-management-system-market-research-report?utm_source=
GNOM&utm_medium=PressRelease&utm_code=4cjcxw&utm_campaign=1549383+-+Global+Battery+Management+System+
Markets%2c+2020-2021+%26+Forecast+to+2030%3a+Automotive%2c+Consumer+Electronics%2c+Industrial%2c+Aerospace+
%26+Defense%2c+Telecommunications&utm_exec=chdo54prd (accessed on 28 July 2022).
4. Balasingam, B.; Ahmed, M.; Pattipati, K. Battery Management Systems—Challenges and Some Solutions. Energies 2020, 13, 2825.
[CrossRef]
5. Meng, J.; Luo, G.; Ricco, M.; Swierczynski, M.; Stroe, D.-I.; Teodorescu, R. Overview of Lithium-Ion Battery Modeling Methods
for State-of-Charge Estimation in Electrical Vehicles. Appl. Sci. 2018, 8, 659. [CrossRef]
6. Bilgin, B.; Magne, P.; Malysz, P.; Yang, Y.; Pantelic, V.; Preindl, M.; Korobkine, A.; Jiang, W.; Lawford, M.; Emadi, A. Making the
Case for Electrified Transportation. IEEE Trans. Transp. Electrif. 2015, 1, 4–17. [CrossRef]
7. Chu, Z.; Feng, X.; Lu, L.; Li, J.; Han, X.; Ouyang, M. Non-destructive fast charging algorithm of lithium-ion batteries based on the
control-oriented electrochemical model. Appl. Energy 2017, 204, 1240–1250. [CrossRef]
8. Lin, Q.; Wang, J.; Xiong, R.; Shen, W.; He, H. Towards a smarter battery management system: A critical review on optimal
charging methods of lithium ion batteries. Energy 2019, 183, 220–234. [CrossRef]
9. Zheng, T. Fault diagnosis of overcharge and overdischarge of lithium ion batteries. Chem. Eng. Trans. 2018, 71, 1453–1458.
[CrossRef]
10. Banguero, E.; Correcher, A.; Pérez-Navarro, Á.; Morant, F.; Aristizabal, A. A Review on Battery Charging and Discharging Control
Strategies: Application to Renewable Energy Systems. Energies 2018, 11, 1021. [CrossRef]
11. Mehar, S.; Zeadally, S.; Rémy, G.; Senouci, S.M. Sustainable transportation management system for a fleet of electric vehicles.
IEEE Trans. Intell. Transp. Syst. 2015, 16, 1401–1414. [CrossRef]
12. Gabbar, H.; Othman, A.; Abdussami, M. Review of Battery Management Systems (BMS) Development and Industrial Standards.
Technologies 2021, 9, 28. [CrossRef]
13. Tarascon, J.-M. Key challenges in future Li-battery research. Philos. Trans. R. Soc. London. Ser. A Math. Phys. Eng. Sci. 2010, 368,
3227–3241. [CrossRef] [PubMed]
14. Rigas, E.S.; Ramchurn, S.D.; Bassiliades, N. Managing Electric Vehicles in the Smart Grid Using Artificial Intelligence: A Survey.
IEEE Trans. Intell. Transp. Syst. 2014, 16, 1619–1635. [CrossRef]
15. Sancarlos, A.; Cameron, M.; Abel, A.; Cueto, E.; Duval, J.-L.; Chinesta, F. From ROM of Electrochemistry to AI-Based Battery
Digital and Hybrid Twin. Arch. Comput. Methods Eng. 2020, 28, 979–1015. [CrossRef]
16. Li, W.; Demir, I.; Cao, D.; Jöst, D.; Ringbeck, F.; Junker, M.; Sauer, D.U. Data-driven systematic parameter identification of an
electrochemical model for lithium-ion batteries with artificial intelligence. Energy Storage Mater. 2021, 44, 557–570. [CrossRef]
17. Korjani, S.; Facchini, A.; Mureddu, M.; Rubino, A.; Damiano, A. Battery management for energy communities—Economic
evaluation of an artificial intelligence-led system. J. Clean. Prod. 2021, 314, 128017. [CrossRef]
18. Raju, P.; Vijayan, S. Artificial Intelligence based Battery Power Management for Solar PV And Wind Hybrid Power System. Int. J.
Eng. Res. Gen. Sci. 2013, 1, 2.
19. Lombardo, T.; Duquesnoy, M.; El-Bouysidy, H.; Årén, F.; Gallo-Bueno, A.; Jørgensen, P.B.; Bhowmik, A.; Demortière, A.; Ayerbe,
E.; Alcaide, F.; et al. Artificial Intelligence Applied to Battery Research: Hype or Reality? Chem. Rev. 2021, 122, 10899–10969.
[CrossRef]
20. Su, W.; Eichi, H.; Zeng, W.; Chow, M.Y. A survey on the electrification of transportation in a smart grid environment. IEEE Trans.
Ind. Inform. 2012, 8, 1–10. [CrossRef]
21. Zhang, X.; Wang, Y.; Liu, C.; Chen, Z. A novel approach of battery pack state of health estimation using artificial intelligence
optimization algorithm. J. Power Source 2018, 376, 191–199. [CrossRef]
22. Xiong, R.; Li, L.; Tian, J. Towards a smarter battery management system: A critical review on battery state of health monitoring
methods. J. Power Source 2018, 405, 18–29. [CrossRef]
23. Wang, Y.; Tian, J.; Sun, Z.; Wang, L.; Xu, R.; Li, M.; Chen, Z. A comprehensive review of battery modeling and state estimation
approaches for advanced battery management systems. Renew. Sustain. Energy Rev. 2020, 131, 110015. [CrossRef]
24. Liu, W.; Placke, T.; Chau, K. Overview of batteries and battery management for electric vehicles. Energy Rep. 2022, 8, 4058–4084.
[CrossRef]
25. Abbas, M.; Cho, I.; Kim, J. Analysis of High-Power Charging Limitations of a Battery in a Hybrid Railway System. Electronics
2020, 9, 212. [CrossRef]
26. Arnieri, E.; Boccia, L.; Amoroso, F.; Amendola, G.; Cappuccino, G. Improved Efficiency Management Strategy for Battery-Based
Energy Storage Systems. Electronics 2019, 8, 1459. [CrossRef]
Electronics 2022, 11, 2695 22 of 24

27. Xing, Y.; Ma, E.W.M.; Tsui, K.L.; Pecht, M. Battery Management Systems in Electric and Hybrid Vehicles. Energies 2011, 4,
1840–1857. [CrossRef]
28. Lee, S.; Kim, J. Power Capability Analysis of Lithium Battery and Supercapacitor by Pulse Duration. Electronics 2019, 8, 1395.
[CrossRef]
29. Uno, M.; Ueno, T.; Yoshino, K. Cell Voltage Equalizer Using a Selective Voltage Multiplier with a Reduced Selection Switch Count
for Series-Connected Energy Storage Cells. Electronics 2019, 8, 1303. [CrossRef]
30. Kokkotis, P.I.; Psomopoulos, C.S.; Ioannidis, G.C.; Kaminaris, S.D. Environmental Aspects of Small Scale Energy Storage Systems.
In Proceedings of the Fifth International Conference on Environmental Management, Engineering, Planning & Economics,
Mykonos, Greece, 14–18 June 2015; pp. 399–406.
31. Battery Recycling Is Important for Environmental Health—Gallegos Sanitation/Republic Services. Available online: https:
//gsiwaste.com/battery-recycling-is-important-for-environmental-health/ (accessed on 28 July 2022).
32. Liu, K.; Li, K.; Peng, Q.; Zhang, C. A brief review on key technologies in the battery management system of electric vehicles.
Front. Mech. Eng. 2018, 14, 47–64. [CrossRef]
33. Rahman, A.; Rahman, M.; Rashid, M. Wireless Battery Management System of Electric Transport. IOP Conf. Ser. Mater. Sci. Eng.
2017, 260, 012029. [CrossRef]
34. Khattak, Y.H.; Mahmood, T.; Alam, K.; Sarwar, T.; Ullah, I.; Ullah, H. Smart Energy Management System for Utility Source and
Photovoltaic Power System Using FPGA and ZigBee. Am. J. Electr. Power Energy Syst. 2014, 3, 86. [CrossRef]
35. Petrariu, A.I.; Lavric, A.; Coca, E.; Popa, V. Hybrid Power Management System for LoRa Communication Using Renewable
Energy. IEEE Internet Things J. 2020, 8, 8423–8436. [CrossRef]
36. Yudho, S. Conceptual design of battery energy storage monitoring system using LoRa. In Proceedings of the 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, Semarang, Indonesia, 19–20 September 2020; pp. 374–377. [CrossRef]
37. Nurcahyanto, H.; Prihatno, A.T.; Jang, Y.M. Battery Management using LSTM for Manhole Underground System. In Proceedings
of the 3rd International Conference on Artificial Intelligence in Information and Communication, ICAIIC 2021, Jeju, Korea, 13–16
April 2021; pp. 500–503. [CrossRef]
38. IEEE Industrial Electronics Society, IEEE Singapore Section. Industrial Electronics Chapter, IEEE Singapore Section, and Institute
of Electrical and Electronics Engineers. In Proceedings of the 2017 12th IEEE Conference on Industrial Electronics and Applications
(ICIEA), Siem Reap, Cambodia, 18–20 June 2017.
39. Vehicular Technology Society and Institute of Electrical and Electronics Engineers. In Proceedings of the 2015 IEEE Vehicle Power
and Propulsion Conference (VPPC), Montreal, QC, Canada„ 19–22 October 2015.
40. Wei, Z.; Zhao, J.; He, H.; Ding, G.; Cui, H.; Liu, L. Future smart battery and management: Advanced sensing from external to
embedded multi-dimensional measurement. J. Power Source 2021, 489, 229462. [CrossRef]
41. Lin, X.; Kim, Y.; Mohan, S.; Siegel, J.B.; Stefanopoulou, A.G. Modeling and Estimation for Advanced Battery Management. Annu.
Rev. Control. Robot. Auton. Syst. 2019, 2, 393–426. [CrossRef]
42. Hu, X.; Xiong, R.; Egardt, B. Model-based dynamic power assessment of lithium-ion batteries considering different operating
conditions. IEEE Trans. Ind. Inform. 2014, 10, 1948–1959. [CrossRef]
43. Wu, B.; Widanage, W.D.; Yang, S.; Liu, X. Battery digital twins: Perspectives on the fusion of models, data and artificial intelligence
for smart battery management systems. Energy AI 2020, 1, 100016. [CrossRef]
44. Su, Y.-D.; Preger, Y.; Burroughs, H.; Sun, C.; Ohodnicki, P. Fiber Optic Sensing Technologies for Battery Management Systems and
Energy Storage Applications. Sensors 2021, 21, 1397. [CrossRef]
45. Digi XBee and XBee-PRO Zigbee RF Modules | Digi International. Available online: https://www.digi.com/products/
embedded-systems/digi-xbee/rf-modules/2-4-ghz-rf-modules/xbee-zigbee#specifications (accessed on 28 July 2022).
46. Sampaio, H.V.; de Jesus, A.L.C.; Boing, R.d.N.; Westphall, C.B. Autonomic IoT Battery Management with Fog Computing. In
Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics);
Springer: Geneva, Switzerland, 2019; Volume 11484, p. 89103. [CrossRef]
47. Khayyam, H.; Abawajy, J.; Javadi, B.; Goscinski, A.; Stojcevski, A.; Bab-Hadiashar, A. Intelligent battery energy management and
control for vehicle-to-grid via cloud computing network. Appl. Energy 2013, 111, 971–981. [CrossRef]
48. Yang, S.; Zhang, Z.; Cao, R.; Wang, M.; Cheng, H.; Zhang, L.; Jiang, Y.; Li, Y.; Chen, B.; Ling, H.; et al. Implementation for a cloud
battery management system based on the CHAIN framework. Energy AI 2021, 5, 100088. [CrossRef]
49. Li, S.; Zhao, P. Big data driven vehicle battery management method: A novel cyber-physical system perspective. J. Energy Storage
2020, 33, 102064. [CrossRef]
50. Li, S.; Li, J.; He, H.; Wang, H. Lithium-ion battery modeling based on Big Data. Energy Procedia 2019, 159, 168–173. [CrossRef]
51. Barrett, D.H.; Haruna, A. Artificial intelligence and machine learning for targeted energy storage solutions. Curr. Opin. Electrochem.
2020, 21, 160–166. [CrossRef]
52. Shen, S.; Sadoughi, M.; Chen, X.; Hong, M.; Hu, C. A deep learning method for online capacity estimation of lithium-ion batteries.
J. Energy Storage 2019, 25, 100817. [CrossRef]
53. Vellingiri, M.T.; Mehedi, I.M.; Palaniswamy, T. A Novel Deep Learning-Based State-of-Charge Estimation for Renewable Energy
Management System in Hybrid Electric Vehicles. Mathematics 2022, 10, 260. [CrossRef]
Electronics 2022, 11, 2695 23 of 24

54. Hossain Lipu, M.S.; Hannan, N.A.; Karim, T.F.; Hussain, A.; Saad, M.H.M.; Ayob, A.; Miah, M.S.; Mahlia, I. Intelligent algorithms
and control strategies for battery management system in electric vehicles: Progress, challenges and future outlook. J. Clean. Prod.
2021, 292, 44. [CrossRef]
55. Lencwe, M.J.; Chowdhury, S.P.D.; Olwal, T.O. Hybrid energy storage system topology approaches for use in transport vehicles: A
review. Energy Sci. Eng. 2022, 10, 1449–1477. [CrossRef]
56. Yao, L.; Xu, S.; Tang, A.; Zhou, F.; Hou, J.; Xiao, Y.; Fu, Z. A Review of Lithium-Ion Battery State of Health Estimation and
Prediction Methods. World Electr. Veh. J. 2021, 12, 113. [CrossRef]
57. Muenzel, V.; de Hoog, J.; Brazil, M.; Vishwanath, A.; Kalyanaraman, S. A Multi-Factor Battery Cycle Life Prediction Methodology
for Optimal Battery Management. In Proceedings of the 2015 ACM Sixth International Conference on Future Energy Systems,
Bangalore, India, 14–17 July 2015; pp. 57–66. [CrossRef]
58. Lee, J.; Kim, J.-M.; Yi, J.; Won, C.-Y. Battery Management System Algorithm for Energy Storage Systems Considering Battery
Efficiency. Electronics 2021, 10, 1859. [CrossRef]
59. Chandran, V.; Patil, C.; Karthick, A.; Ganeshaperumal, D.; Rahim, R.; Ghosh, A. State of Charge Estimation of Lithium-Ion Battery
for Electric Vehicles Using Machine Learning Algorithms. World Electr. Veh. J. 2021, 12, 38. [CrossRef]
60. Chiang, Y.H.; Sean, W.Y.; Ke, J.C. Online estimation of internal resistance and open-circuit voltage of lithium-ion batteries in
electric vehicles. J. Power Sources 2011, 196, 3921–3932. [CrossRef]
61. Akagi, S.; Yoshizawa, S.; Ito, M.; Fujimoto, Y.; Miyazaki, T.; Hayashi, Y.; Tawa, K.; Hisada, T.; Yano, T. Multipurpose control and
planning method for battery energy storage systems in distribution network with photovoltaic plant. Int. J. Electr. Power Energy
Syst. 2019, 116, 105485. [CrossRef]
62. Mao, J.; Miao, J.; Lu, Y.; Tong, Z. Machine learning of materials design and state prediction for lithium ion batteries. Chin. J. Chem.
Eng. 2021, 37, 1–11. [CrossRef]
63. Dvorak, D.; Bauml, T.; Holzinger, A.; Popp, H. A Comprehensive Algorithm for Estimating Lithium-Ion Battery Parameters from
Measurements. IEEE Trans. Sustain. Energy 2017, 9, 771–779. [CrossRef]
64. Sui, X.; He, S.; Vilsen, S.B.; Meng, J.; Teodorescu, R.; Stroe, D.-I. A review of non-probabilistic machine learning-based state of
health estimation techniques for Lithium-ion battery. Appl. Energy 2021, 300, 117346. [CrossRef]
65. Hu, X.; Li, S.E.; Yang, Y. Advanced Machine Learning Approach for Lithium-Ion Battery State Estimation in Electric Vehicles.
IEEE Trans. Transp. Electrif. 2015, 2, 140–149. [CrossRef]
66. Lv, C.; Zhou, X.; Zhong, L.; Yan, C.; Srinivasan, M.; Seh, Z.W.; Liu, C.; Pan, H.; Li, S.; Wen, Y.; et al. Machine Learning: An
Advanced Platform for Materials Development and State Prediction in Lithium-Ion Batteries. Adv. Mater. 2021, 34, 2101474.
[CrossRef]
67. Qian, C.; Xu, B.; Chang, L.; Sun, B.; Feng, Q.; Yang, D.; Ren, Y.; Wang, Z. Convolutional neural network based capacity estimation
using random segments of the charging curves for lithium-ion batteries. Energy 2021, 227, 120333. [CrossRef]
68. Li, Y.; Li, K.; Liu, X.; Wang, Y.; Zhang, L. Lithium-ion battery capacity estimation—A pruned convolutional neural network
approach assisted with transfer learning. Appl. Energy 2021, 285, 116410. [CrossRef]
69. How, D.N.T.; Hannan, M.A.; Lipu, M.S.H.; Sahari, K.S.M.; Ker, P.J.; Muttaqi, K.M. State-of-Charge Estimation of Li-Ion Battery in
Electric Vehicles: A Deep Neural Network Approach. IEEE Trans. Ind. Appl. 2020, 56, 5565–5574. [CrossRef]
70. Ardeshiri, R.R.; Balagopal, B.; Alsabbagh, A.; Ma, C.; Chow, M.-Y. Machine Learning Approaches in Battery Management Systems:
State of the Art: Remaining useful life and fault detection. In Proceedings of the 2020 2nd IEEE International Conference on
Industrial Electronics for Sustainable Energy Systems (IESES), Cagliari, Italy, 1–3 September 2020.
71. Uzair, M.; Abbas, G.; Hosain, S. Characteristics of Battery Management Systems of Electric Vehicles with Consideration of the
Active and Passive Cell Balancing Process. World Electr. Veh. J. 2021, 12, 120. [CrossRef]
72. Merkle, L.; Segura, A.S.; Grummel, J.T.; Lienkamp, M. Architecture of a Digital Twin for Enabling Digital Services for Battery
Systems. In Proceedings of the 2019 IEEE International Conference on Industrial Cyber Physical Systems (ICPS), Taipei, Taiwan,
6–9 May 2019; pp. 155–160. [CrossRef]
73. Wang, W.; Wang, J.; Tian, J.; Lu, J.; Xiong, R. Application of Digital Twin in Smart Battery Management Systems. Chin. J. Mech. Eng.
2021, 34, 57. [CrossRef]
74. Li, W.; Rentemeister, M.; Badeda, J.; Jöst, D.; Schulte, D.; Sauer, D.U. Digital twin for battery systems: Cloud battery management
system with online state-of-charge and state-of-health estimation. J. Energy Storage 2020, 30, 101557. [CrossRef]
75. Merkle, L.; Pöthig, M.; Schmid, F. Estimate e-Golf Battery State Using Diagnostic Data and a Digital Twin. Batteries 2021, 7, 15.
[CrossRef]
76. Florea, B.C. Electric Vehicles Battery Management Network Using Blockchain IoT. In Proceedings of the 2020 IEEE International
Conference on Automation, Quality and Testing, Robotics (AQTR), Cluj-Napoca, Romania, 21–23 May 2020; pp. 1–6. [CrossRef]
77. Bao, J.; He, D.; Luo, M.; Choo, K.-K.R. A Survey of Blockchain Applications in the Energy Sector. IEEE Syst. J. 2020, 15, 3370–3381.
[CrossRef]
78. IEEE Power Electronics Society, IEEE Power & Energy Society, IEEE Industry Applications Society, and Institute of Electrical and
Electronics Engineers. In Proceedings of the 2020 IEEE Transportation Electrification Conference & Expo (ITEC), Chicago, IL,
USA, 23–26 June 2020.
Electronics 2022, 11, 2695 24 of 24

79. Faika, T.; Kim, T.; Ochoa, J.; Khan, M.; Park, S.-W.; Leung, C.S. A Blockchain-Based Internet of Things (IoT) Network for
Security-Enhanced Wireless Battery Management Systems. In Proceedings of the 2019 IEEE Industry Applications Society Annual,
Baltimore, MD, USA, 29 September–3 October 2019.
80. Trevizan, R.D.; Obert, J.; De Angelis, V.; Nguyen, T.A.; Rao, V.S.; Chalamala, B.R. Cyberphysical Security of Grid Battery Energy
Storage Systems. IEEE Access 2022, 10, 59675–59722. [CrossRef]
81. Kumbhar, S.; Faika, T.; Makwana, D.; Kim, T.; Lee, Y. Cybersecurity for Battery Management Systems in Cyber-Physical
Environments. In Proceedings of the 2018 IEEE Transportation Electrification Conference and Expo (ITEC), Long Beach, CA,
USA, 13–15 June 2018; pp. 934–938. [CrossRef]
82. Kharlamova, N.; Hashemi, S.; Traholt, C. The Cyber Security of Battery Energy Storage Systems and Adoption of Data-driven
Methods. In Proceedings of the 2020 IEEE 3rd International Conference on Artificial Intelligence and Knowledge Engineering,
Laguna Hills, CA, USA, 9–13 December 2020; pp. 188–192. [CrossRef]