Renewable and Sustainable Energy Reviews 179 (2023) 113280

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

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

Digital twin of electric vehicle battery systems: Comprehensive review of

the use cases, requirements, and platforms
F. Naseri a, *, S. Gil a, C. Barbu a, E. Cetkin b, G. Yarimca b, A.C. Jensen c, P.G. Larsen a,
C. Gomes a, **
a
Department of Electrical and Computer Engineering, Aarhus University, Aarhus, Denmark
b
Department of Mechanical Engineering, Izmir Institute of Technology, Izmir, Turkey
c
Department of Energy and Climate, Danish Technological Institute, Aarhus, Denmark

A R T I C L E I N F O A B S T R A C T

Keywords: Transportation electrification has been fueled by recent advancements in the technology and manufacturing of
Artificial intelligence (AI) battery systems, but the industry yet is facing serious challenges that could be addressed using cutting-edge
Battery management system (BMS) digital technologies. One such novel technology is based on the digital twining of battery systems. Digital
Battery passport
twins (DTs) of batteries utilize advanced multi-layer models, artificial intelligence, advanced sensing units,
Battery recycling
Internet-of-Things technologies, and cloud computing techniques to provide a virtual live representation of the
Digital twin (DT)
Electric vehicle (EV) real battery system (the physical twin) to improve the performance, safety, and cost-effectiveness. Furthermore,
Fault diagnosis they orchestrate the operation of the entire battery value chain offering great advantages, such as improving the
Internet-of-things (IoT) economy of manufacturing, re-purposing, and recycling processes. In this context, various studies have been
Machine learning (ML) carried out discussing the DT applications and use cases from cloud-enabled battery management systems to the
Predictive maintenance digitalization of battery testing. This work provides a comprehensive review of different possible use cases, key
Remaining useful life (RUL) enabling technologies, and requirements for battery DTs. The review inclusively discusses the use cases,
Second-life
development/integration platforms, as well as hardware and software requirements for implementation of the
Software architecture
battery DTs, including electrical topics related to the modeling and algorithmic approaches, software architec­
tures, and digital platforms for DT development and integration. The existing challenges are identified and
circumstances that will create enough value to justify these challenges, such as the added costs, are discussed.

cause a thermal runaway leading to the fire or explosion of the battery

1. Introduction [4,5]. Batteries face challenges not only when they are in use, but also
across their whole value chain from raw material and supply chain to the
Transportation electrification is an essential pathway to limiting repurposing and recycling of batteries. Despite the decreasing trend in
global warming below 1.5 ◦ C as set out in the Paris Agreement [1]. their cost, EVs are still deemed expensive and unaffordable.
Electric vehicles (EVs) crucially rely on lithium-ion batteries to store the A battery management system (BMS) is usually used to ensure that
required energy for propulsion. Due to their significance, batteries have the EV battery is operated within safe limits [6]. However, BMSs often
been an active field in academia, industry, and among policymakers, use a low-cost microprocessor that hinders the use of best-in-class al­
and large investments have been secured to improve them. The ad­ gorithms for the optimum operation of the batteries. Novel digitalization
vancements in battery technology cannot be underestimated. However, techniques can address the challenges in the battery processes including
there are still unresolved challenges in the field that have slowed down manufacturing, assembling, operating, repurposing, and recycling [7].
the electric transportation paradigm, for example, long charging times One such technology is based on the digital twin (DT) concept. The DT of
and quick degradation, especially under fast charging conditions [2,3]. a battery is its live digital equivalent with prediction capabilities, which
The lithium-ion cells are sensitive to abusive operating conditions such is formed by employing multi-scale models, advanced data processing
as over-charging and elevated operating temperatures. Such conditions techniques based on artificial intelligence, machine learning (ML), and
may on rare occasions initiate unstable chain reactions which could internet-of-things (IoT) driven two-way data connectivity between the

* Corresponding author. Finlandsgade 22, Aarhus, 8200, Denmark.

** Corresponding author.
E-mail addresses: [email protected] (F. Naseri), [email protected] (C. Gomes).

https://doi.org/10.1016/j.rser.2023.113280
Received 7 September 2022; Received in revised form 31 January 2023; Accepted 27 March 2023
Available online 16 April 2023
1364-0321/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

Abbreviations ML Machine Learning

NN Neural Network
AAS Asset Administration Shell PDT Performance Digital Twin
AWS Amazon Web Services PF Particle Filter
BMS Battery Management System PT Physical Twin
CAD Computer-Aided Design RDF Resource Description Framework
CAN Controller Area Network RUL Remaining Useful Life
DT Digital Twin SCADA Supervisory Control and Data Acquisition
DTI Digital Twin Instance SoC State-of-Charge
DTP Digital Twin Prototype SoH State-of-Health
EMS Energy Management System SoP State-of-Power
EV Electric Vehicle SotA State-of-the-Art
IoT Internet-of-Things SoX State-of-X
KF Kalman Filter TMS Thermal Management System
LCC Life Cycle Cost VIT Voltage, Current, and Temperature
MAE Mean Absolute Error

physical twin (PT) and DT, to allow an accurate replication/prediction described some of the DT requirements and frameworks. Several re­
of the battery behavior [8]. The battery DT technology is advantageous view/survey works have also been published on this topic. Table 1 lists
in different terms as outlined in the following: these papers, the related publication year, as well as the focus area of
each paper.
• Application: It facilitates battery design optimization, improves The existing review/survey works have mostly focused on the
battery operation and maintenance, and makes batteries more effi­ possible DT applications and the DT conceptualization while some of
cient and cost-effective. them have referred to existing possible frameworks. These papers have
• Emissions/Environment: It contributes to mitigating the effects of addressed DTs only from the electrical perspective. However, battery DT
climate change by enabling more widespread use of clean energy and requires knowledge of multiple disciplines due to the interdependence
reducing emissions from transportation. It also supports the fulfill­ of the electrical, software, and digital systems. Except in Ref. [12], all
ment of environmental, social, and governance and United Nations review works have addressed battery DTs merely within the EV context
(UN) sustainable development goals by promoting sustainable en­ without reviewing DT opportunities across the whole battery value
ergy systems and reducing greenhouse gas emissions. chain such as the repurposing and 2nd life applications.
• Policy targets: It supports the fulfillment of policy targets related to Unlike other works, this review addresses the battery DT use cases
the energy storage, reduction in emissions, and sustainability, by across the whole value chain. In addition to discussing the electrical and
providing a comprehensive understanding of battery behavior and software requirements together, the review provides insight into the
performance over the entire lifetime. commercial and open-source DT design, development, and integration
• Regulation/Standards: It catalyzes the process of battery stan­ platforms, as well as their pros and cons. Furthermore, the results of an
dardization and regulation in terms of safety and performance. industry questionnaire are interpreted to identify the gaps and reveal
• Cost: It improves the life cycle cost (LCC) by improving the economy circumstances where the DT can create real value from a business
of manufacturing, reducing maintenance costs, helping batteries last standpoint to justify its costs. The audience of this review is very broad
longer, etc. including industry and academic experts working across the whole
battery value chain from EV users, fleet operators, and component in­
Despite being an active research topic in recent years, the battery DT tegrators to battery recyclers, as fully depicted in Fig. 1.
requirements and use cases are still unclear in an industrial setting. In The main objective of this review is to provide a useful indication of
addition, the simulation and software platforms that can be used for the the scope of the existing research literature including the ongoing trends
development and integration of the DTs have not been surveyed before. on different aspects of battery DTs including their use cases, re­
The works on battery DTs have been mostly focused on developing quirements, and platforms. The literature search is limited to the English
advanced BMS functionalities such as advanced modeling and state language, in Scopus, Science Direct, and IEEE Xplore for publications
estimation functions for batteries in addition to the works which have between 2003 to January 2023. Any peer-reviewed article containing
information about battery DT use cases, requirements, and DT devel­
opment/integration platforms is included in the review. Additionally,
Table 1
the DT platforms that are commercially available were found and
Review/survey papers published on the topic of battery DTs.
reviewed through an internet search. Some articles in which a review of
Ref. Year Topics covered their title and/or abstract disclosed little pertinent to the batteries were
published
excluded without further review. Reviewers with a relevant discipline
[9] 2022 • Brief review of DT use cases were assigned to abstract the information from different articles in the
• Review of the DT architecture
reviewed database. Information was collected on the type of DT use
[10] 2021 • Review of DTs in EV context including batteries, power
electronics, advanced driving assistance systems, etc cases considered and the achieved performance indicators, electrical,
• DT use cases focused on BMS algorithms mechanical, and software requirements, as well as the existing platforms
[11] 2021 • Review of use cases of battery DTs in EVs used for the design, development, and integration of the DTs and their
[12] 2021 • Review of DTs applications in the design, manufacturing, potential in terms of data storage, security, processing power, visuali­
operation, and post-operation phases
zation, etc. The analysis is fulfilled to provide an adequate level of detail
• Brief overview of the DT framework
[13] 2021 • Brief review of the use cases focused on EV operation about the battery DT potential, trends, and gaps while it also determines
[14] 2020 • Battery models and data processing for digital twining the value of undertaking a full systematic review of individual use cases,
• Battery state estimation technologies, and platforms.

2
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

Fig. 1. Potential stakeholders of the battery DT concept and audience of this review.

The presentation of the work is organized as follows: In Section 2, the platforms. The existing gaps, challenges, and opportunities, are dis­
definitions of the DTs are reviewed and the existing challenges in the EV cussed in Section 6, and finally, conclusions are provided in Section 7.
battery value chain are outlined. Section 3 provides a review of the state-
of-the-art (SotA) battery DTs and the relevant use cases to address the 2. Background on DTs and battery challenges
challenges of the batteries throughout the value chain. Then, in Section
4, the requirements of the battery DTs from electrical and software Based on the field of application, the DT is defined and understood in
perspectives are reviewed. Section 5 provides a review of the existing different ways. The DT has been referred to as a mega-model, avatar,
commercial and open-source battery DT development and integration mirrored system, digital shadow, or synchronized virtual prototype

Fig. 2. Difference between battery model, digital shadow, and DT [17].

3
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

[15]. Therefore, a generally accepted term and definition for DT is still interconnect stakeholders who play roles in different life cycle stages
lacking. Fig. 2 illustrates how a battery DT can be distinguished from a and to facilitate harmonization, legislation, and standardization of
battery digital model and a battery digital shadow [16]. Benchmark different battery technologies. A universal and transparent framework
definitions of the DT are also reviewed in Table 2. to benchmark battery technologies and validate/track progress toward
Depending on the application area and use case, this research pro­ sustainable and resource-efficient battery designs is still missing. In the
poses the following three kinds of DTs, as illustrated in Fig. 3(a). Each next Section, the DT’s potential to address some of these challenges is
DT type addresses specific needs in different life stages of the battery, i. reviewed.
e., the digital twin prototype (DTP) addresses the manufacturing of
batteries while the performance digital twin (PDT) deals with the 3. Review of the use cases and functionalities of the battery DT
operation of batteries in 1st life and 2nd life. The digital twin instance
(DTI) is the most complicated type of DT covering the whole lifecycle of The research on battery DT has sped up in recent years. New features
the batteries from manufacturing to material recycling. Stakeholders’ and functionalities are being introduced for the battery DT as the
interests in different DT types are shown in Fig. 3(b). technologies related to IoT, artificial intelligence, big data, and cloud
Generally, the DT has found several applications in the EV context computing are evolving. The DT can be used not only to address the
including autonomous driving, converters and inverters, digital design challenges in each stage of battery life but also to interconnect different
and manufacturing, health monitoring, advanced driving assistant sys­ life cycle phases of the battery orchestrating its performance and
tems, and battery systems. The applications cover various EV compo­ resulting in reduced LCC. The following subsections review the existing
nents and subsystems to improve performance and safety. Detailed and potential use cases of battery DTs.
descriptions of the DT use cases in the EV context can be found in
Ref. [11]. However, this review focuses on the use cases related to 3.1. SoX estimation and cell balancing
battery systems only. The use cases are driven by the challenges across
the whole battery value chain including battery raw material, The state-of-X (SoX) variables, i.e., state-of-charge (SoC), state-of-
manufacturing of cells, pack assembly, operation phases (EV and health (SoH), and state-of-power (SoP) are very important in the BMS
second-life), and recycling of batteries. Fig. 4 depicts the challenges in context since they are input to many algorithms that are responsible for
different stages of cycle life. At the material stage, the challenge is to find monitoring, controlling, and protecting the battery pack. The SoC is
and use effective and abundant battery materials to ensure sustainabil­ equivalent to the fuel gauge in non-electric cars. SoH indicates battery
ity. An important challenge in the manufacturing phases (cell and pack) health which is commonly characterized by the battery capacity or in­
is the production quality and quality assurance, which involves exten­ ternal resistance. The SoP determines the safe battery power boundaries
sive design effort, and long and expensive certification processes. In during normal EV operation or regenerative braking. Traditionally,
addition, the evolving battery materials mean that until the search for an these algorithms are implemented in the onboard BMS which runs on a
ultimate battery is not concluded, new designs and technologies at cell microprocessor with a capacity of a few hundred Mbytes [24]. Because
and pack levels should be brought forward, which is expensive and of the CPU and memory limits in the BMS, best-in-class SoX estimation
time-consuming. In EVs, the key challenge is battery durability and algorithms cannot be implemented due to the infeasibility of the
safety. Batteries require many algorithms to operate them, safely, and embedded platforms. This leaves an undesired state estimation error
expensive electronics are needed to achieve this. Likewise, batteries tend leading to potentially poorer battery performance and safety. On the
to degrade fast under harsh operating conditions such as elevated tem­ other hand, the DT sits on the cloud which offers considerably higher
peratures or high discharge rates. Design of a high-performance yet computational resources compared to the BMS allowing higher achiev­
cost-effective BMS to maximize safety and durability is challenging. able accuracy.
Battery repurposing and recycling phases are challenged mainly by the DT-based SoX estimation has been the subject of several publica­
complexity of processes and lack of regulations. A challenge that can be tions. In Ref. [25], SoC and SoH estimation is fulfilled on the cloud-based
attributed to the whole value chain is the lack of interoperability to DT using the adaptive H-infinity filter and particle swarm optimization,
respectively. As illustrated in Fig. 5, taken from Ref. [25], the VIT data
Table 2 (voltage, current, and temperature) are sent from the slave BMS to a
Benchmark definitions of the DT. Raspberry Pi by the controller area network (CAN) protocol. The data is
then transmitted to the DT using the MQTT protocol. The SoX algorithms
Author/institution Year DT concept/definition
are validated on a real uninterruptable power supply set-up resulting in
Michael Grieves [18] 2003 A “virtual digital expression equivalent to an SoC estimation mean absolute error (MAE) of 0.49%. Likewise,
physical products.”
NASA [19] 2010 An “integrated multi-physics, multi-scale,
DT-based SoH estimation was reported to achieve MAE of 0.74% and
probabilistic simulation of a vehicle or system 1.7% for capacity and resistance estimation, respectively. In Ref. [26], a
that uses the best available physical models, joint H-infinity filter and particle filter (PF) online algorithm has been
sensor updates, fleet history, etc., to mirror the proposed for cloud-based SoC estimation resulting in an MAE of 0.14%.
life of its flying twin.”
An effective approach for SoH estimation has been proposed in
Roland Rosen [20] 2015 A “new wave in modeling, simulation, and
optimization technology, which provides a big Ref. [27], where a battery DT based on the long short-term memory is
set of all digital artifacts.” developed and used to virtually discharge the battery to estimate its
Michael Schluse [21] 2016 A “virtual representation of a real-world subject” capacity. The MAE of SoH estimation is obtained at 2.86%. The
or a “real-world object.” DT-based SoC and SoH estimation has also been fulfilled in Ref. [28], in
Edward M. Kraft (U.S. 2016 A “multidiscipline simulation of a real-world
Air Force) [22] product, which uses data and sensor information
which different ML techniques based on random forest, light gradient
as input to model that mirror and predict the boosting, and deep NN were applied resulting in the MAE of 0.549% and
states and behavior over the lifespan of the 0.603% for SoC and SoH estimation, respectively. The DT is structured
physical system.” to fulfill regular estimation of the battery degradation while retraining
Rikard Soderberg [23] 2017 It “contains geometry representation of the
the SoC mechanisms to reflect the battery aging effect. In Ref. [29], the
assembly, kinematic relations, Finite Element
Analysis functionality, Monte Carlo simulation, battery DT has been used for SoX estimation and cell balancing control.
material properties and link to inspection The PF algorithm was used for SoX estimation, which yielded MAE of
database.” 0.3% and 0.5% for SoC and SoH estimations, respectively. Based on the
Billy Wu [14] 2020 A “digital replica of a physical entity with a close predictions of the actual capacity, the DT predicts the time that each cell
connection between the two.”
requires to be balanced and accordingly formulates a fast and accurate

4
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

Fig. 3. (a) Different types of battery DTs (b) Relevance of the DT type to stakeholders.

balancing strategy. The concept of the DT-assisted equalization strategy BMS concept proposed in Ref. [34]. The differences between the battery
is shown in Fig. 6 [29]. As seen, the BMS boards measure the key vari­ DT and the cloud BMS depend on the implementation philosophy and
ables of different batteries and this data will be sent to the DT on the the DT definition. In cases where the cloud BMS is defined as a PDT of
cloud, where balancing algorithms are placed. The necessary commands the battery, the terms can be used, interchangeably.
to control the balancing actuators will be generated by the DT and will
be sent back to the batteries. Other works have used DT for SoC and SoH
estimation based on extended Kalman filter-particle swarm optimization 3.2. Fault diagnosis and prognosis
[30], Kalman filter (KF) combined with least-squares support vector
machine and PF[31], ML [32,33], etc. Onboard BMS uses classic fault detection and protection methods
The battery algorithms can also be implemented within the cloud against the battery faults such as over-discharge, over-charge, short-
circuit, etc. The fault diagnosis is usually fulfilled using single-variate

5
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

Fig. 4. Battery challenges throughout the value chain.

Fig. 5. Cloud-based DT is used for SoX estimation. SoC is estimated continuously while SoH is updated from time to time and is provided as input to the SoC
estimation algorithm [25].

6
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

Fig. 6. DT-based equalization of the batteries [29].

techniques by comparing VIT variables to fixed threshold values dis­ number of cycles that have remained before reaching the end-of-life.
regarding the historical usage data. In the battery DT, the availability of Despite significant work that has been fulfilled in academia to perform
large computational resources and historical battery usage data offer the battery RUL estimation, none of the existing commercial BMSs have the
possibility to explore more advanced fault diagnosis techniques. The RUL estimation function up to now. However, RUL estimation on the DT
abundance of sensory data enables the use of advanced multi-variate and cloud offer several advantages as listed in Fig. 7.
condition monitoring techniques to improve accuracy in detecting and A large number of algorithms have been proposed to estimate the
locating faults. Fault prognosis is another interesting topic that can be RUL which rely on the model of the battery, data, or a combination of
explored in a DT context. Thanks to DT computation, advanced multi- both. A detailed review of RUL estimation algorithms can be found in
scale and physics-based models can be run online to detect or predict Ref. [38].
faults, an option that is conventionally beyond the BMS capability due to
the lack of resources. DT-enabled physics-based battery models can 3.4. Predictive maintenance
monitor the mechanisms and processes that eventually may trigger the
faults. Examples of such model-based methods are presented in Refs. EVs need regular service and maintenance. The conventional main­
[35,36], wherein the DT is created by combining the electrical, thermal, tenance schemes are inefficient and economically inviable since they are
and degradation models of the battery, reduced order models, or using normally applied either too soon, e.g. when the battery pack does not
the dynamic mode decomposition-based data-driven model. Intelligent need maintenance, or too late when a defect or expensive damage shows
monitoring of batteries using DT has been discussed in Ref. [37]. This up. With the battery DT, the RUL can be monitored in real-time, and
use case of the DT has also been studied in other fields, e.g. monitoring of thus, the maintenance can be scheduled only in case a repair or service is
steam turbines [8]. This type of DT falls under the category of PDTs. deemed necessary resulting in cost savings, improved battery lifetime
and durability, and avoiding unwanted shut-downs [39]. Likewise, the
3.3. RUL estimation manual battery check-up can be automated which saves time and money
for EV owners. The predictive maintenance use case for the battery DT
RUL gives information about the durability of the battery pack before using the Bayesian-based adaptive evolution method has been studied in
it reaches its end-of-life. Unlike the SoH which shows the battery status Ref. [40] wherein the lifetime prediction and reliability evaluation al­
at present time, the RUL is an indicator that predicts future battery gorithms have been developed to estimate the RUL. It was concluded
degradation trends. Normally, the RUL is estimated in terms of the that the battery maintenance costs can be reduced by up to 62% when

Fig. 7. Use cases of the RUL estimation on the cloud-located DT.

7
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

the predictive maintenance strategy was used. stored on the DT to facilitate the sharing services.

3.5. Battery repurposing, second-life, and recycling

3.7. Design and production optimization

The repurposing process from an EV battery to a second-use appli­

The battery DT (DTP type) can be used to improve the manufacturing
cation (e.g., stationary energy storage systems) is costly and time-
processes of batteries. Conventional methods for quality assurance, such
consuming due to the need for disassembly and manual lifetime
as those based on the house-of-quality or Pareto chart require significant
testing at the module or cell levels. With the DT concept, the SoH of the
resources by multi-disciplined teams to address a specific problem or
batteries can be continuously estimated, stored in the DT database, and
defect related to the product, which can take several weeks to months.
interpreted when needed, while this data can also be shared with
However, with a DT, it is easier to find the root cause of the defects in the
second-life stakeholders such as battery manufacturers for operational
manufacturing process and the quality assurance can be fulfilled within
planning, e.g., to forecast the availability of batteries exiting from first
a few days [23]. Likewise, with the battery DT, the information from the
life. Likewise, the economic value for different second-use applications
manufacturing level can be supplied to the EV operational stage, for
can be assessed, e.g., the energy and power capabilities of the batteries
instance, to calibrate some models and algorithms. Vice versa, the
can be accurately mapped to obtain the most viable application for the
operational data of the battery can be used as feedback to adjust the
second-life. Batteries with more power capability can be used for grid
design and optimization processes. Through providing this cross-stage
ancillary services (power conditioning, frequency support, etc.) while
exchange of data and information, the DT will create flexibility to
batteries with more energy potential can be used in uninterruptable
improve performance in different state-of-life of the battery. This
power supply systems. This use case has been considered in Ref. [41],
application falls under the DTI category.
wherein DT was used for battery residual value estimation as depicted in
The application of DT in battery manufacturing has been considered
Fig. 8.
in several publications. In Ref. [43], an efficient 3D discrete calendaring
element model has been proposed for the DT of the battery electrode,
3.6. Sharing (swapping) services which is the core of the battery manufacturing process. In Ref. [44], a 3D
resolved electrochemical model of electrodes was established to eval­
Battery sharing use case has been studied in Ref. [42], where a DT uate the effect of manufacturing parameters such as slurry formulation
was used to grasp the status of the shared batteries to facilitate their on the electrode microstructure and battery performance. Other works
replacement. This allows swapping stations to understand the status of include digital twining to consider 3D shapes of active material particles
the battery, how it was operated, and to what extent it was degraded [45], the dying model of electrodes [46], and carbon-binder spatial
when it was being used by a distinct EV. Likewise, EV owners can swap location [47]. To accelerate and ramp up the manufacturing processes,
batteries without being worried about the true health state of the bat­ in Refs. [48,49], DT has been used for fast and economic battery module
tery. Battery DT also creates opportunities for financial services and assembly and for flexible pouch cell stack formation to enable virtual
insurance companies since they know the reliability and degradation of testing and evaluation of different solutions before real manufacturing.
batteries and they can adjust fees and financial strategies, accordingly. In Ref. [50], a DT environment has been developed to automate the test
Stakeholders can develop more viable guarantee/warrantee plans when and evaluation of the batteries to speed up the manufacturing process. In
detailed conditions of the EV batteries are available online. The same Ref. [51], a DT has been developed for all-solid-state batteries to esti­
DT-based concept can be used in the case of EV rental and leasing mate experimentally inaccessible and difficult to obtain information
companies. An example of this use case is presented in Ref. [42], such as dead particles, specific contact area, and charge distribution in
wherein the SoC, SoH, and running distance of the EVs are estimated and 3D domain, which will help to map design and performance parameters

Fig. 8. Schematic of the cloud-based DT for battery repurposing and 2nd life decision-making [41].

8
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

for battery design optimization. The digitalization of battery testing will with the regulations [56]. Through effective life cycle management with
also reduce the time and cost associated with the experiments for battery the battery passport concept, second-life services can save money and
characterization. Design optimization of the battery thermal manage­ time by not testing batteries for a second time and recyclers can better
ment system (TMS) using the DT has also been considered in Ref. [52]. set the requirements for the recycling processes as detailed in Ref. [55].
Most of the reviewed works fall under the category of DTPs targeting Fig. 9 shows the schematic of a Battery Identity Global Passport pro­
manufacturing processes. posed in Ref. [55]. It is noteworthy that the Global Battery Alliance has
recently called for prompt actions to exchange battery data through the
3.8. Energy optimization battery passport concept [56].
The realization of the battery passport concept requires a framework
The performance of the energy management system (EMS) de­ that connects all operational phases. In this regard, an effective DT
termines the EV driving range, lifetime of batteries, EV acceleration, etc. framework has been proposed in Ref. [12] to interconnect the research
EMS is normally realized onboard the EVs and due to the limited pro­ and design phase, manufacturing phase, after-sale phase, and
cessing power, embedding advanced EMS strategies which are based on post-operation phase based on the cloud space and 5G communication.
sophisticated global and online optimization methods, stochastic algo­ Another DT framework for lifecycle management of the EV battery packs
rithms, ML, etc. Is difficult. On the other hand, the DT offers sufficient has been proposed in Ref. [57], wherein the design phase,
computational power which can be used to run best-in-class EMS algo­ manufacturing phase, and operations phase (including the second-life)
rithms improving the EV performance under different driving condi­ are equipped with their battery DTs (research and development DT,
tions, e.g., to decide the optimum power limits during the acceleration manufacturing DT, battery DT, and DTs of other assets) while all DTs
and regenerative braking. Examples of such advanced EMS strategies share information through centralized cloud storage in IT system. In
can be found in Refs. [53,54]. Detailed analysis of these algorithms is Refs. [58,59], the importance of blockchain technology to empower
beyond the scope of this review. sustainable manufacturing and lifecycle management in industry 4.0 is
highlighted. Likewise, the challenges and future of DTs in the manage­
3.9. Thermal management system ment of the product lifecycle are reviewed in Ref. [60].

The TMS is responsible to control the heating/cooling apparatus to 3.11. Battery charging and vehicle-to-grid (V2G) operation
maintain the battery temperature within a specific temperature range
and reduce the temperature gradients and temperature inhomogeneous Different protocols can be used for battery charging including Con­
across the pack. Conventionally, the TMS is implemented onboard. stant Current (CC), Constant Voltage (CV), CC-CV, pulse charging, etc.
However, the DT makes it possible to implement advanced predictive The charging protocol affects the charging performance in terms of ef­
and intelligent control strategies to improve the overall TMS perfor­ ficiency, battery degradation, charging time, and cost [61]. In this re­
mance and battery lifetime. One such example has lately been proposed gard, an effective use case for DT is the health-aware charging of
by Bosch™ Mobility Solutions. The so-called “Battery in the Cloud” batteries [62]. In the DT, the charging problem can be formulated into a
concept receives data from the battery and fleet and takes predictive multi-objective optimization problem with a cost function that can
actions to improve the battery’s performance. For example, the TMS consider degradation, charging cost, efficiency, and time, and decides
starts to adjust the battery temperature a few minutes before the EV the optimized protocol and charging parameters such as frequency and
reaches a charge station that is already booked. This not only reduces the width of charging pulses during pulse charging. The cost function can be
charging time (by preparing the battery to accept higher charge cur­ optimized, for example, to maximize the lifetime of the battery pack (by
rents) but also results in less battery degradation under fast charging reducing the temperature gradients, preventing lithium plating, etc.)
conditions. while reducing the charging time to the extent possible. In addition,
constraints such as the maximum permissible charging power can be
3.10. Battery passport introduced to the charging optimization problem. An example of the use
case is the convex multiperiod optimization strategy proposed in
A battery DT can be used to manage a so-called battery passport to Ref. [63] to coordinate the battery charging while accounting for
monitor, collect, and integrate battery data and metadata starting at the maximum charging power and voltage rise. Similarly, the application of
cradle (manufacturing) toward the end-of-life gate (recycling) [55]. The artificial intelligence based on deep reinforcement learning for opti­
battery passport is defined as a digital entity that conveys the social, mized fast charging of batteries has been reported in Ref. [64]. In this
governance, and environmental requirements to guarantee compliance work, fast charging is realized through a multiphysics-constrained

Fig. 9. Schematic of the Battery Identity Global Passport proposed in [55].

9
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

strategy with the consciousness of thermal safety and battery aging, and Table 3 summarizes the reported use cases of the battery DT. Fig. 10
an extension of the battery life by about 15% at an equivalent charging shows that the battery digital twining is moving quickly and the number
speed is reported. Thanks to the large computational resource available of applications and use cases reported has steadily increased in recent
on the cloud, the DT makes it possible to use advanced algorithms for years.
solving large optimization problems like this.
Batteries also offer the possibility of V2G operation to support the 4. Key enablers and requirements
electric grid when needed, e.g. to contribute to grid stabilization, peak-
shaving, providing backup power, etc. [65,66]. Similar to battery The technologies, components, and requirements of the battery DTs
charging, a cost function can be formulated to optimize the V2G oper­ depend on the DT type, application area, and the considered use cases.
ation using the DT, e.g. minimize the battery degradation when oper­ The elements are listed in Fig. 11 and key technologies are discussed in
ating in V2G mode, increase the revenue, etc. Likewise, the usage profile the following subsections.
and data of the EVs, batteries, charge stations, and grid can be stored on
the DT and this data can be used for the optimization of charging 4.1. Measurement of the key battery variables
stations.
Several sensors are usually integrated into the battery PT to measure
the key variables including the VIT data. Voltage is normally measured
Table 3 for every cell while current is measured at pack level using current
Review of the DT use cases. shunts or hall effect-based sensors. Temperature is normally measured
Use case Ref. Description for every other cell. Advanced battery packs also include multi-sensing
units to measure additional variables such as gas, pressure, and strain
SoX estimation [25] Cloud BMS with H-infinity filter-based SoC
estimation algorithm and SoH estimation based
to detect hazardous situations. The sensory measurements are the input
on particle swarm optimization to the DT models and algorithms. These data should thus be communi­
[26] DT-based BMS with combined H-infinity and PF- cated to the cloud-located DT with proper sampling rates. Sample rates
based SoC estimation depend on the use cases. However, the common practice is 10HZ for
[27] Battery DT was used to virtually apply a complete
voltage and current and 1 Hz for temperatures. The IoT gateway can
discharge cycle to measure SoH (discharge
capacity) listen to the vehicle CAN bus to read the data and pass them to the
[29] DT-assisted SoC, SoH, and SoP estimation using battery DT. Alternatively, the IoT gateway can directly listen to the BMS
PF algorithm communication interfaces conditional on compatibility with the IoT
[30] DT-based BMS for extended Kalman filter-based gateway. The DT uses this data to update the models, perform data
SoC estimation and particle swarm optimization-
based SoH estimation
analysis, and operate algorithms for monitoring, optimization, SoX/RUL
[31] DT-based BMS with KF-least-squares support estimation, etc.
vector machine SoC estimation and AR-PF SOH
estimation 4.2. Battery models
[32] Cloud-enabled DT for SoC and SoH estimation of
batteries
[28] DT-based SoC and SoH estimation using data- Models serve several services and tasks related to battery operation
driven methods based on random forest, light [67]. Generally, the DT considers different modeling levels from mate­
gradient boosting, and deep NN rial and cell components to the pack. Lithium-ion transport in the active
[33] Cloud-enabled DT for SoC and SoH estimation material and between the electrodes takes place on the nanometer and
using ML
Battery cell [29] DT-assisted balancing control of the cells
micrometer scales whereas, at the cell level, behavior is described on the
equalization millimeter scale such as in heat transport [68]. Therefore, multi-scale
Battery Monitoring [35] Module-level modeling of batteries based on modeling is required to involve different time scales. In this regard,
cloud-enabled DT for monitoring of the batteries 3D models are the most complicated type incorporating cell
[37] Intelligent monitoring based on battery DT
multi-physics to model the electrode porosity and inhomogeneous cell
[36] Dynamic mode decomposition-based data-driven
model for battery DT behavior along the 3D coordinates. The 3D models are very accurate and
Battery sharing services [42] Battery DT was developed to enable services that robust, but they contain nonlinear coupled partial differential equations
facilitate battery sharing, e.g., running distance that are too heavy to be solved in real-time. Thus, the key challenge has
calculation been about finding effective model order reduction techniques that can
Battery design and [43] Battery DT model based on discrete element
manufacturing method for manufacturing optimization
reduce the complexity of the models while maintaining fidelity. Various
[48] DT development for flexible cell stack formation models have accordingly been developed. The electrochemical model
of pouch battery cells based on the so-called Newman or P2D assumes homogeneous electrode
[49] DT was proposed to develop robotic workcells for particles of the same size and predicts the behavior in axial coordinates.
fast and economic battery module assembly
Several model order reduction methods for discretizing and approxi­
[51] DT-driven battery model developed for design
optimization and to reveal experimentally mating the P2D model based on the finite-difference method, Pade
inaccessible information approximation, etc. Have been proposed resulting in several
[50] Battery DT was used to speed up development electrochemical-thermal model combinations with different scales, e.g.,
and manufacturing phases by eliminating actual P2D+0D, P2D+1D (also known as P3D), and P2D+2D/3D. In the
system tests
[52] GPR-based battery virtual DT developed for
simplest case, spatial lumping has been proposed to form the so-called
design optimization of the TMS single particle model wherein each electrode will be modeled only
Repurposing and [41] Cloud-enabled battery DT to facilitate second-life using one spherical particle and the electrolyte dynamics will be
second-life decision-making process via DT-based aging neglected. The model assumes a uniform temperature distribution
prediction
within the cell. However, simplified models fall short in certain oper­
Predictive Maintenance [39] Semi-analytical DT model of battery pack
developed for real-time temperature monitoring ating regimes such as C-rates typically higher than 3C. As shown in
and predictive maintenance Ref. [69], the reduced version of the P2D model with thermal dynamics,
[40] DT was used to fulfill life prediction and also named P2D-T, can be extended to capture the degradation mecha­
reliability evaluation for predictive maintenance nisms such as the loss of lithium inventory. The model is then discretized
implementation
and applied for the estimation of battery SoC using the singular

10
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

Fig. 10. (a) Publications per year on the battery DT concept (b) Share of battery DT use cases.

Fig. 11. Key elements of the battery DTs.

evolutive interpolated KF algorithm. In Ref. [70], distributed fiber op­ parameters, heat generation rate, SoC, and maximum capacity. A more
tical sensors have been embedded in the battery and the data has been detailed review of thermal models can be found in Ref. [71]. Data-driven
used to build a one-state thermal model for co-estimation of thermal models such as those based on neural networks (NNs) or hybrid models

11
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

(model-based combined with data-driven) can improve the DT gener­ union of different models related to other systems, subsystems, and
alization capability and overcome some of the limitations of reduced processes. For example, the cell models can be combined to build
models. The availability of large quantities of data on the DT, module and pack models while the pack model could in addition contain
user-friendly software frameworks [72,73], and specialized hardware the models of the BMS and TMS. This is conceptually shown in Fig. 13. In
[74] have made it simpler to apply NNs to the task of predicting the this regard, one should highlight the effective model development/
behavior of batteries. A good example of such data-driven modeling is improvement toolbox proposed in Ref. [88] wherein all DT modeling
the hybrid lumped-thermal-NN model proposed in Ref. [75], in which a requirements such as cross-integration of multiple cell models, execu­
mechanism-driven distributed lump thermal model is combined with an tion of models in “what-if” scenarios, and model update calculation are
ML-based axial thermal gradient compensation segment. See Refs. well considered.
[76–78] for an overview of NN architectures in use for the simulation of
dynamical systems.
The model structures are designed such that some parameters affect 4.3. DT architecture
the model behavior if certain inputs are given, i.e., it could be the case
that some parameters are never found by calibration. In these cases, The common feature of DT frameworks is the use of services that can
noise filtering techniques [79] and the design of experiments [80] have be integrated into the platform [89,90]. Hao et al. argued in Ref. [91]
been demonstrated to be useful. Another challenge is the time-varying that the most important services for DT are state estimation (also known
model behaviour which results in variations of the model parameters as inverse modeling or sensor fusion), predictive maintenance, fault
in different conditions such as different temperatures, SoC, C-rates, and diagnosis, decision-making support, and self-adaptation. The architec­
degradation levels. Various real-time model parameterization algo­ ture of the DT can therefore be represented by layers, where the lower
rithms have been proposed to address this challenge. Most popular layer enables connectivity and data exchange between the DT and the PT
methods are based on recursive filtering techniques such as the (effectively working as an IoT framework), and the upper layer provides
least-squares family [81] and more recently, the application of ML for more sophisticated services. To denote these layers [92], introduced the
battery model parametrization has also been explored [82]. term 3D DT, to denote DTs that only have the lower layer.
The equivalent circuit models capture the battery behavior using the The term 5D DT, introduced in Ref. [93], extends the 3D DT by the
electrical circuit components. Despite relatively higher computational data and service dimensions. The 5D DT then provides a central storage
efficiency, the equivalent circuit models cannot represent the physio- location connected to the physical, virtual, and service space, and a
chemical attributes of the cells. However, they are found to be effec­ service system for implementing data-driven services. The need for 5D
tive for specific tasks such as SoX estimation and fault diagnosis. As an DT appeared after applying the 3D DT in the industry, where the new
example, in Ref. [83], an equivalent circuit model based on Thevenin’s services allow new functionality with a higher value.
structure has been used to develop a model-based internal short-circuit Fig. 14 illustrates a basic architecture for a DT implemented in
detection technique, which features robustness against degradation and Ref. [94]. It uses a RabbitMQ server as a broker for communication
noise effects. Stochastic battery models have also been proposed to between different services and InfluxDB as a database to store, access,
predict aging behavior due to the randomness of different operating and retrieve data. The middle three blocks in the DT represent the
conditions and aging mechanisms [84]. A detailed review of battery services.
models is beyond the scope but some good review papers merely dedi­ All components communicate via the RabbitMQ message exchange,
cated to the modeling topic have already been published [85–87]. and the data is stored in a time series database (InfluxDB). In Ref. [95], a
Fig. 12 shows the advantages and disadvantages of each model type. DT architecture has been proposed which consists of three different
The cell model is the building block but in practice, the DT can be the layers as shown in Fig. 15(a). The hardware and connectivity layer is
responsible to collect all sensory data and fulfilling edge-processing to

Figiure 12. Potential of different battery models for use in the DT.

12
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

Fig. 13. Physical battery architecture and DT battery models for cell#1 to cell#N. All BMS/TMS algorithms are replicated on battery DT.

Fig. 14. Exemplary communication of incubator DT in relation to a self-adaptation service (from Ref. [94]).

remove the outliers. Battery simulations, models, and algorithms oper­ The DT functional architecture proposed in Ref. [96] consists of five
ate in the DT layer which also includes the databases. High-level services different layers as shown in Fig. 16. The architecture is similar to
and aggregation operate in the service layer. The database in the twin Ref. [95] except that two additional layers are introduced for the
layer considers four types of data, i.e., the master data (such as cell database and connectivity. In Ref. [97], a new framework for
metadata), transaction data (such as VIT data), state data (processed DT-assisted enterprise resource planning of battery manufacturing sys­
data at the twin layer), and link data as illustrated in Fig. 15(b). tems has been proposed. The framework enables real-time access to

13
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

Fig. 15. (a) DT architecture proposed in Ref. [95] (b) Overview of data types in the DT architecture.

Fig. 16. The DT functional architecture proposed in [96].

multiple data types such as battery real-time operational data, produc­ sizes, message routing, handle packet dropout, and minimize commu­
tion line data, and customer feedback data, which will help to optimize nication delays while being resource efficient. Some communication
the production chain in manufacturing and to accelerate the technologies like RabbitMQ even enable load balancing and therefore
decision-making processes. increased resilience. Failures in communication typically can lead to
A seven-layer DT architecture has also been proposed in Ref. [88] as disastrous consequences in a real-time battery monitoring scenario, such
shown in Fig. 17. The layered structure shown in Fig. 17 is similar to the as overcharge, over-discharge, and thermal runaway.
one presented by Ref. [96] except that an additional layer for the se­ Fortunately, there are many mature communication techniques in
curity of the DT is foreseen. place: RabbitMQ and Apache Kafka, to name a few. On the research side,
As summarized, most DT architectures follow the layered architec­ a wide review of IoT related to energy systems is carried out in Ref. [98],
tural style combined with service-oriented architecture. The upcoming wherein the role, impact, challenges, and constraints in different sce­
subsections detail the different steps and most common services narios and subdomains are described. They also listed some IoT pro­
required to form a DT. tocols and technologies that can be used in different applications,
including 6lowPAN, Power Line Carrier, ZigBee, among others. Refer­
ence [99] also highlights the capabilities of 5G-IoT and its comparison to
4.4. IoT and connectivity other technologies, such as LoRa, Wi-Fi, ZigBee, and Bluetooth. NB-IoT
and the 3 GPP low power wide area technologies offer a different
The communication framework is an integral part of the DT. It needs promising alternative for IoT communication in smart grid and
to support point-to-point communication, wildly different message

14
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

Fig. 17. The DT architecture proposed in [88].

Vehicle-to-Grid applications [99]. Reference [100] presents an orches­ battery PT and its environment; spawn new what-if simulations from
tration of different IoT protocols in the electrical industry and a mapping current and historical data; replay past states; and display predictive
of the protocols HTTP, MQTT, OPC UA, and LoRaWAN to use cases in maintenance results.
different scenarios. It can be challenging to strike the right balance in detail, as data
visualization needs to promote real-time, perceptual, and scalability
4.5. Data storage [106]. For more details, the reader can refer to Ref. [107].

The most appropriate technology to store data from battery PT is the 4.7. State estimation
time series databases since these are optimized for querying time series
data [101,102]. Other types of data storage methods that are suitable for Key battery state variables include SoC, SoH, and SoP. Other states
DTs can be based on the semantics that is normally used along with data such as state-of-safety, state-of-temperature, state-of-energy, and state-
models. A common technology for this is the resource description of-function can be of interest in some use cases. Battery state variables
framework (RDF), which allows the storage of data using ontological or are highly correlated and vary with time. State estimation combines the
relational models as the scheme of the databases. This way, data can be available sensory data and battery models to indirectly elicit these state
stored and maintained on top of semantic models [103]. Description variables. State estimation methods are very diverse and can be gener­
documents and metadata of DTs also need to be considered in large-scale ally categorized into model-based, data-driven, and hybrid methods.
DT setups and managing them in Git repositories with web servers for The model-based methods can be based on stochastic techniques such as
the DT Web, similar to the one proposed in Ref. [104]. Metadata and KF and its variations (extended Kalman filter, Unscented KF, etc.), PF, H-
description files of the DT Web can also be mapped to RDF structures, infinity filter, etc., or deterministic techniques such as state observers.
enabling a bridge between static and dynamic data storage and querying Many data-driven methods have been proposed for battery state
using SPARQL, and even reasoning, such as presented in Ref. [105]. estimation-NNs, long short-term memory networks, and SVM to name a
few. Model-based methods fall short in certain operating conditions due
to low fidelity, e.g., during the fast charging of the battery. On the other
4.6. Visualization hand, the data-driven methods have better generalization capability, but
they rely on an extensive amount of experimental data to be trained.
Visualization tools have matured over recent years. Currently, it is Obtaining the data requires time-consuming and expensive laboratory
possible to produce 3D interactive animations using tools such as Unity tests. Hybrid methods combine the advantages of the other two ap­
(https://unity.com), Qt (https://www.qt.io), iTwin (https://www.it proaches. An Example of the hybrid method is presented [108] which
winjs.org), Gazebo, and dashboard interfaces with Grafana, Dash, to combines PFs and NNs to predict the battery SoH. A detailed review of
name a few. the battery state estimation can be found in Ref. [109].
The challenge remaining is the ability to create such interfaces
quickly from CAD and semantic models of the battery. For example, it
should be possible to, use such interfaces; selectively visualize the 3D

15
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

4.8. Monitoring and anomaly detection and accidents leading to injury or death [124]. To this end, the battery
DT can be perceived as a prime target for potential attackers, similarly to
Monitoring is crucial for the evaluation of the PT’s behavior and for supervisory control and data acquisition (SCADA) systems [112,125,
checking whether the original assumptions about the environment in 126]. Several ways of preventing attacks at industrial and specifically
which the PT operates still hold [110]. In the SotA [111,112], one can CPSs have been proposed, such as among others, the use of formal
distinguish between online and offline monitoring techniques. Online methods to create secure architectures [127], integration of different
monitoring is done as each new data sample arrives, whereas offline security controls [128], or the use of state estimators [129,130], where
monitoring is suitable for use cases where an added delay is tolerated. In attack resilient state estimators have been proposed [131]. Securing DTs
offline monitoring, a time series database stores all data from the PT and consequently requires not only considering access control, network se­
is consulted at leisure by the monitor. Monitors can also attempt to curity, and transmitted data integrity but also the integrity of the model
identify the cause of problematic behavior by replaying the historical itself.
data through models of the system, while artificially injecting faults in As examples of a DT as a security tool, one has [132,133] both
the system until they find a potential explanation for the problematic proposing using the DTs assets in the design of the security aspects and
behavior observer [113]. Monitors are not limited to checking for un­ attack modeling and mitigation. In Ref. [132], the authors propose a
desirable situations. They can also check for expected and desirable framework to generate DTs from specifications for SCADA systems. In
behavior of the DT services themselves, and not just the PT. For example, addition to being a security tool, the work of [133] proposes to use the
concerning the safety of the battery PT, it can be very valuable to be able DT in training and simulation, testing exercises for security engineers.
to monitor when the discrepancy between the monitored and predicted Further information on this topic can be found in Refs. [134–136].
behaviors starts and thereby enable anomaly detection. The reader can Relevant standards that apply to the DT domain in terms of security
refer to the SotA in run-time monitoring [111,112] and advanced bat­ and privacy include IEC 27400–2022 entitled “Cybersecurity — IoT
tery monitoring strategies [114]. security and privacy — Guidelines” and IEC 62443 which defines
cybersecurity for industrial automation networks.
4.9. What-if (Co-)simulation
5. Review of the existing DT platforms
A DT foundation as a decision support system is the ability to run
simulations using hypothetical scenarios reproduced from historical The implementation process of the battery DTs is twofold: develop­
data or containing predictions of the future environment of the PT. For ment and integration. For each phase, several commercial and open-
instance, after an anomaly is detected, many simulations can be run to source tools and platforms are available, which are reviewed in the
investigate what the probable cause is. Naturally this forces such sim­ following subsections.
ulations to be faster than real-time, so as to find a solution and intervene
before the anomaly becomes a fault. In cases where the system is rep­ 5.1. Development platforms
resented by coupled sub-models that have been produced by different
tools, then co-simulation [115] can be used to realize the coupling. The development of DTs includes the design and establishment of the
What-if simulations can also be used to optimize the battery configu­ appropriate battery models which could replicate the real battery
rations as the battery operates in new conditions or environments. behavior to the desired purpose. Several tools can be used for battery DT
However, one of the challenges is to quantify and manage the uncer­ development, which are reviewed in the following.
tainty inherent in these simulations and subsequence choices based on
them. 5.1.1. ANSYS Twin Builder
ANSYS Twin Builder offers multiphysics system solvers which
4.10. Self-adaptation involve applications in 1D and 3D space with easy and fast system setup.
It makes simulation models closer to the real batteries by combining 3D
Self-adaptation can be denoted to the process of optimizing the PT physics solvers and the reduced order models and thus, the electrical,
configuration as a response to changes in its environment, as proposed thermal, and mechanical models can be coupled and tested in synergy.
earlier in autonomic computing, thereby increasing its resilience. DTs ANSYS heating/cooling library offers a wide range of models for TMS
are perfect candidates to deploy self-adaptation loops. However, components such as heat exchangers, turbomachinery, valves, etc. It is
designing these loops requires domain experts who, incidentally, do not also possible to model two-phase flows which are needed to model the
usually possess the software engineering background to deploy these TMS’s refrigeration cycles. ANSYS Fluent and ANSYS Twin Builder have
loops. In this regard, complex event processing frameworks [116] which been applied by Electronic Cooling Solutions Inc. For thermal prediction
provide a useable interface to encode behavioural rules can be of help. and optimization [137].
Domain specific language engineering provides the tools to quickly A typical battery DT developed in Ansys is shown in Fig. 18 [138].
create such code-free development environments that can be used by The battery DT simulation steps are shown in Fig. 19.
domain experts [113]. Fig. 20 shows how the PT and DT can be connected using ANSYS
After a new configuration for the PT has been found, it is necessary to Twin Builder. The real-time data measured by sensors are transmitted to
make sure not only that it is safe, but also that the act of changing the PT the battery DT using IoT. The DT perceives the data as input, performs
from its current configuration to the new one, is safe. Here formal all the simulations, and sends feedback to the battery PT.
methods [117,118] and reachability analysis [119–121] can play a role.
As for the main steps in any self-adaptation loop, the reader can refer 5.1.2. COMSOL Multiphysics®
to the MAPE-K Loop, proposed in Ref. [122] and detailed in Ref. [123], COMSOL Multiphysics® is a commercial Finite Element software
as a reference architecture for designing self-adaptive loops. The work in with strong tools for battery DT development. The battery’s electrical,
Ref. [94] adapts the MAPE-K loop to the context of DT. thermal, and mechanical aspects can be numerically investigated in
synergy. Similar to ANSYS, the COMSOL DT incorporates lightweight
4.11. Privacy and security models such as reduced order models and lumped models to improve
computational efficiency. The concept of COMSOL’s battery DT is shown
This aspect of battery DTs is especially important for industrial use in Fig. 21. To link the battery with its DT, COMSOL offers an application
because compromised security could lead to dangerous situations such programming interface communication tool powered by Java. The web
as unstable operation of the battery, physical and economic damages, service transfers the measured battery data to the DT. Upon stimulation

16
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

Fig. 18. (a) A battery pack example with 28 cells, (b) battery pack configuration with connections of 4 modules, (c) Visualization of results.

Fig. 19. System simulation steps of battery DT in the Ansys Digital Twin Builder [138].

of the scenarios, the control parameters and reports are sent back to the ready-made multi-physics libraries with applications and solutions that
battery. Additionally, COMSOL Multiphysics comes with a dedicated enable accurate analysis supported by powerful platform features.
and effective Battery Design Module, which can be used to develop a Simcenter can be easily integrated into computer-aided design,
range of models for battery DTs at different scales from cell level to pack computer-aided engineering, and control software packages [47].
level. The DT models can be easily structured and calibrated to cover the A brief comparison of the battery DT development platforms is
right modeling and simulation needs, e.g. modeling of different aging provided in Table 4.
phenomena, thermal runaway, etc.
5.2. Integration platforms for battery DTs
5.1.3. Siemens Simcenter Amesim and digital twin
Simcenter Amesim software is an integrated, scalable mechanical
Since DT technology has been a trending topic in recent years for
system simulation platform that enables design engineers to virtually
both industry and academia, the necessity for ready-to-use DT platforms
evaluate and optimize system performance. This will increase the evo­
has increased. The backbone of these platforms lies in the IoT infra­
lution of systems engineering, from producing more realistic simulations
structure for the connectivity layers with additional modeling, simula­
at earlier design stages to final performance verification and control
tion, and visualization layers. The available options vary from
calibration. The software seamlessly combines system simulation and
commercial to open-source solutions and can be also combined with
testing, helping to predict performance on critical features ahead of time
other toolsets for DTs, such as [140,141]. There is also the ISO standard
and throughout its entire lifecycle. Simcenter Amesim allows you to
ISO 23247:2021 [142], which proposes a set of consensual terms for DTs
practically build models and perform realistic analysis by combining
and common criteria for DT frameworks. According to the standard, a

17
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

Fig. 20. Connectivity of battery and DT in ANSYS Digital Twin Builder [138].

Fig. 21. Battery DT concept in COMSOL Multiphysics® [139].

DT framework requires accuracy to describe at an appropriate level of usage, compatibility to exchange information with other systems, us­
fidelity, communication that enables synchronization, data acquisition ability to effectively achieve user goals, security of the connection,
with local or remote sensors, built-in data analysis, data integrity to maintainability to adapt the software product, and portability of soft­
correctly describe the state of the PT, compatibility to a range of ap­ ware and hardware artifacts.
plications, granularity for different levels of detail, identification of the The asset administration shell (AAS) is a German industrial frame­
data to a unique PT, management for resource optimization, coverage of work initiative for DTs [144], which has several open-source imple­
different product life-cycle phases, security to communicate only with mentations, including Eclipse BaSyx, SAP I4.0 AAS, NOVAAS, and AASX
authorized agents, built-in simulation, synchronization (data, state Package Explorer. AAS intends to provide a consensual metamodel for
estimation, etc.) between DT and PT, dynamic viewpoint according to DTs hand to hand with industrial standards.
objectives, and support hierarchical modeling to fit in a defined iTwin and Unity offer a toolkit for 3D-enabled DTs. The former is
hierarchy. open-source while the latter is commercial that has free licenses for
Regarding the available frameworks [143], reviewed and analyzed students.
Amazon web services (AWS) IoT Greengrass, Microsoft Azure DTs, and Going to a different domain, TerriaJS and digital twin cities centre
Eclipse Ditto in conjunction with Eclipse Hono and Eclipse Vorto in platform, both open-source, provide a foundational geospatial data DT
seven dimensions, covering 13 requirements in total. The frameworks framework, where the data associated with the PT can be mapped to
are assessed regarding overall functionality, the performance of resource identify patterns in the geospatial context.

18
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

Table 4
Comparison of well-known software packages for DT development ( = High-perfor­
mance, = Mid-performance, = Low-performance).

When analyzing the requirements for DT frameworks with the the overall performance depends on several aspects because the DT
reviewed platforms, it is possible to identify the strengths and weak­ application is composed of different components, such as communica­
nesses of each one in the following dimensions. tion, modeling, data analysis, data storage, and so on.
DT development process: The DT development process is not trivial Data storage: The current DT frameworks are built upon IoT
with the currently available DT frameworks. This is mainly because DT frameworks, which means that they already implement a data storage
engineering is a complex task that requires several components. Inte­ layer as part of the system, which is an advantage. Additionally, they can
grating the models into the connectivity layers and visualizations re­ provide different data storage methods, such as SQL and No-SQL data­
quires time and handwork to set it up. Although, modular bases, but also plain text and structured files, such as CSV and JSON,
implementations, such as commercial platforms, facilitate this by respectively. Metadata is also relevant to be stored because the seman­
providing user-friendly components that can reduce the complexity tics of data and systems can be strategically used for extracting valuable
considerably. inferences and learning from data, systems, and their composition. Here,
Connectivity: Since most of the DT frameworks are built upon IoT structured data with common schemes, standardized meta-structures,
infrastructure, the connectivity is usually good and flexible. The and RDF structures can be used within aware systems and entities.
frameworks integrate IP-based communication protocols, such as Visualization: Visualization applies to both 2D and 3D visualiza­
MQTT, AMQP, HTTP, and WebSockets, which enable communication tions, which can be required in different contexts, depending on the
within the system and with third parties. However, some frameworks purposes and scope of the DT use case. If the requirement is regarding 3D
are not built upon and oriented to connectivity and their communication aspects or the PT requires a 3D representation, DT frameworks like
capabilities make them less prone to external integrations and less iTwin and Unity can supply this aspect. In case geospatial 2D graphs are
flexible regarding interoperability and synchronization. required, TerriaJS can be a better option. It can also be the case that 2D
Security: Common security aspects rely on the security of the con­ visualizations are needed for monitoring. For these cases, any of the
nectivity layer, such as authorization, authentication, and encryption. frameworks can be integrated with dashboards to obtain built-in 2D
However, other security aspects regarding access control, security self- visualizations.
aware DTs, and similar are not covered by the existing frameworks. Modeling and simulation: This aspect is not completely covered by
Processing power, performance, and scalability: Depending on the current available DT frameworks. Modeling categories that are
the scope of the framework, the processing power, performance, and covered by the frameworks include data models, 3D models, or ML
scalability of DT platforms can be different. Frameworks, such as iTwin, models. Simulation, on the other hand, relies on high-fidelity physical
which is oriented to 3D scenarios, require more processing power than models, which are not offered as built-in modules so far. Development
IoT-based DT frameworks, such as Ditto. The same happens with the platforms described in subsection 5.1 can be useful here.
scalability; the more processing power a framework requires, the fewer An empirical analysis of the frameworks is shown in Table 5 [143].
instances of DT - PT pairs it can run at the same time. On the other hand,

Table 5
Qualitative comparison of different DT integration platforms ( = High-performance, = Mid-performance,
= Low-performance).

19
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

6. Discussions on challenges, gaps, and opportunities the SotA related to each use case is reviewed. The number of published
research papers on the battery DT topic has increased by about 200%
The challenges and gaps of the battery DTs are discussed as follows: since 2020, which demonstrates the increasing interest in the concept.
Among these research works, the majority have focused on using battery
• Lack of Standards and legislation: There is no standard or consensus DT for SoX estimation, design optimization, manufacturing, and moni­
about the definition, architecture, functional requirements, manda­ toring of batteries. More recently, the application of battery DTs for
tory or nice-to-have features of the battery DTs. The transparency of predictive maintenance and 2nd life applications has also gained inter­
battery data throughout its value chain will expose the technical and est. The potential use cases that can be developed within the battery DT
trade secrets of manufacturers, recyclers, etc. Therefore, proper framework such as energy optimization, optimized battery charging,
legislation concerning the privacy and transparency level of the data lifecycle management, and battery passport are also introduced. The
is required. reports show promising results for battery DT application, e.g. 60%
• Cybersecurity of battery DTs: There is a certain risk that the reduction in maintenance costs, a 15% improvement in the lifetime
communication channels between the battery DT and PT can be using optimized charging protocols, and a DT-based SoC estimation
compromised and manipulated by adversaries. Manipulation of the MAE of 0.14%.
sensory data or DT feedback misleads the battery control and pro­ The multi-disciplinary elements and requirements of the battery DTs
tection algorithms which can lead to risks of battery fires and acci­ are categorized into three groups, namely software, hardware, and IoT.
dents. Therefore, research should focus to identify the DT security Each group is then reviewed and the corresponding technical aspects are
gaps and potential hazards that may arise and their effect on the PT discussed in detail. Possible sensor networks to measure the key battery
operation and accordingly devise appropriate countermeasures to information such as VIT sensors, gas sensors, and GPS data, and IoT/
maintain the cybersecurity and integrity of the DT. connectivity options such as 5G to transmit the data to the cloud-located
• Complexity and cost of implementation: Battery DT development DT and vice-versa are reviewed. In terms of software requirements,
requires significant design effort to develop multi-disciplinary sys­ several key topics are covered including the co-simulation of multi-scale
tem designs and frameworks and relies on expensive infrastructures models, cyber-security, review of databases suitable for time-series data,
such as gateways, cloud servers, super-processors, and potentially and so on. The battery DT architectures are also reviewed. The three
additional sensor units. Despite that the additional cost and main layers in the DT architecture are the connectivity layer to receive
complexity can be justified in the medium-term due to the benefits and pre-process various types of data, the twin layer to operate the
for a wide spectrum of stakeholders, this still poses a challenge to the battery multi-scale models and algorithms, and the service layer which
initial investment and development of the pilot set-ups or full-scale provides higher-level services such as visualization, user interfaces,
development at higher technology readiness levels. application programming interface, etc. It is worthwhile in the DT ar­
• Technical barriers: The VIT data of the cells and heterogeneous data chitecture to include the metadata and metamodels related to different
from other sensor units (e.g. gas or pressure sensors) should be types of data (such as master data, transactions data, etc.) and models
collected and communicated to the cloud at a relatively high rate. (cell models, module models, BMS, TMS, etc.), which will improve the
The EV battery pack may include thousands of cells which means findability and interoperability of the data and models on the battery
millions of real-time data will limit the communication bandwidth. DT.
This necessitates using advanced communication technologies to The study also contributes to the review of the existing DT devel­
facilitate real-time data connectivity. The use of new generation opment platforms including ANSYS Digital Twin Builder and COMSOL
communication interfaces such as 5G technology is thus needed for Multiphysics. The commercial integration platforms such as Microsoft
the implementation of the battery DTs. In addition, processing a huge Azure and AWS and open-source integration platforms such as AAS and
amount of data on the cloud requires strong processors and deep its different implementations such as NOVAAS and AASX are reviewed
fusion via advanced signal processing algorithms to achieve precise and compared in terms of their security and capabilities in modeling,
mapping and a fast flow of data, and sharp responsiveness of the DT. processing, visualization, connectivity, and storage of the DT data. The
comparison provides valuable information to decide the best develop­
Taking these challenges into account, it is very important to consider ment/integration platform depending on the requirements of the
more targeted DT applications and use cases to avoid excessive resources implemented use case(s) and the specific modeling/simulation needs.
on tasks that do not provide too much value. To investigate the indus­ Finally, the existing challenges in front of the battery DT technology
trial potential of the DT, a technical questionnaire was recently devel­ are reviewed. The strategic challenge is the lack of standards and
oped as part of the European Project HELIOS and a few partner legislation, which complicates the transparency of data across the bat­
companies in the BMS/EV sector were asked to complete it. The feed­ tery value chain. The economic challenge is related to the large invest­
back revealed that a major bottleneck is related to the DT cost which ment costs due to the need for additional sensor networks, cloud
hardly can be justified for some use cases such as SoC estimation since services, etc. Concerning the technical barrier, the complicated
the achievable results on BMS are already good. Concerning the SoC modeling and algorithmic processes can be pointed out. In addition,
estimation, another bottleneck is the real-time performance which is communication and online processing of a massive amount of real-time
difficult to realize due to the high sample rates required, communication data related to a large number of cells is a challenging task. The review
delays, and computational burden related to cell-level SoC estimations. also summarizes the results of an industry survey about the effectiveness
The use of DT for SoH estimation could make more sense as the required of the existing battery DT use cases. Companies believe that the addi­
update intervals are much longer. The advanced life extension algo­ tional costs associated with the battery DT must be carefully justified
rithms operated by DT could improve the lifetime and thus reduce the considering the use cases. In this regard, SoC estimation on the DT is
life cycle cost of batteries. It would also be worthwhile to consider DT considered to be less effective because the achievable results on the BMS
use cases that provide value for a fleet and not individual EVs. Con­ are already good. Use cases such as health estimation, battery passport,
cerning this, use cases related to fleet management, repurposing of and predictive maintenance that will benefit a larger number of stake­
batteries, and battery passports can be very promising. holders such as EV users, fleet operators, leasing and car-sharing com­
panies, etc. Are deemed to be more promising.
7. Concluding remarks While the use cases are comprehensively outlined, a detailed review
of each use case was not included due to space limitations. However, the
This work provides a comprehensive review of the battery DTs. work can be used to judge if it would be worthwhile to do a full sys­
Different existing and potential use cases of the DTs are discussed and tematic review of each use case. In terms of open-source DT platforms,

20
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

only the most known ones are considered in the review. Therefore, [18] Grieves M. Digital twin: manufacturing excellence through virtual factory
replication. White paper, 2014;1:1–7.
future work can evaluate a wider range of DT platforms and assess the
[19] Shafto M, et al. NASA Report. Modeling, simulation, information technology &
possibility of applying them in the battery context. processing roadmap, 32; 2012. p. 1–38.
[20] Rosen R, Von Wichert G, Lo G, Bettenhausen KD. About the importance of
autonomy and digital twins for the future of manufacturing. IFAC-PapersOnLine
2015;48(3):567–72.
Declaration of competing interest [21] Schluse M, Rossmann J. From simulation to experimentable digital twins:
simulation-based development and operation of complex technical systems. In:
The authors declare that they have no known competing financial 2016 IEEE international Symposium on systems engineering (ISSE). IEEE; 2016.
p. 1–6.
interests or personal relationships that could have appeared to influence
[22] Kraft EM. The air force digital thread/digital twin-life cycle integration and use of
the work reported in this paper. computational and experimental knowledge. In: 54th AIAA aerospace sciences
meeting; 2016. p. 897.
Data availability [23] Söderberg R, Wärmefjord K, Carlson JS, Lindkvist L. Toward a digital twin for
real-time geometry assurance in individualized production. CIRP Annals 2017;66
(1):137–40.
No data was used for the research described in the article. [24] Panwar NG, Singh S, Garg A, Gupta AK, Gao L. Recent advancements in battery
management system for Li-ion batteries of electric vehicles: future role of digital
dwin, cyber-physical systems, battery swapping technology, and nondestructive
Acknowledgment testing. Energy Technol 2021;9(8):2000984.
[25] Li W, Rentemeister M, Badeda J, Jöst D, Schulte D, Sauer DU. Digital twin for
battery systems: cloud battery management system with online state-of-charge
We are grateful to the Poul Due Jensen Foundation, which has sup­ and state-of-health estimation. J Energy Storage 2020;30:101557.
ported the establishment of a new Centre for Digital Twin Technology at [26] Tang H, Wu Y, Cai Y, Wang F, Lin Z, Pei Y. Design of power lithium battery
Aarhus University. We would also like to thank the sponsors of the management system based on digital twin. J Energy Storage 2022;47:103679.
[27] Qu X, Song Y, Liu D, Cui X, Peng Y. Lithium-ion battery performance degradation
Digital Transformation Laboratory in the Ringkøbing-Skjern. Part of this
evaluation in dynamic operating conditions based on a digital twin model.
work is fulfilled within the framework of the HELIOS project which has Microelectron Reliab 2020;114:113857.
received funding from the European Union’s Horizon 2020 research and [28] Alamin KSS, Chen Y, Macii E, Poncino M, Vinco S. A machine learning-based
innovation program under grant agreement No. 963646. The permis­ digital twin for electric vehicle battery modeling. In: 2022 IEEE international
conference on omni-layer intelligent systems (COINS). IEEE; 2022. p. 1–6.
sions to reuse Figs. 5, 6, 9, 16 and 17 in this review paper are arranged [29] Wang Y, Xu R, Zhou C, Kang X, Chen Z. Digital twin and cloud-side-end
with the corresponding copyright holders. We acknowledge the per­ collaboration for intelligent battery management system. J Manuf Syst 2022;62:
missions granted to reuse these figures. 124–34.
[30] Zhou M, Bai L, Lei J, Wang Y, Li H. A digital twin model for battery management
systems. In: Concepts, algorithms, and platforms," in the international Conference
References on image, Vision and intelligent systems (ICIVIS 2021). Springer; 2022.
p. 1165–76.
[31] Peng Y, Zhang X, Song Y, Liu D. A low cost flexible digital twin platform for
[1] P. Agreement. Paris agreement. In: Report of the conference of the parties to the
spacecraft lithium-ion battery pack degradation assessment. In: 2019 IEEE
United Nations framework convention on climate change (21st session, 2015:
International Instrumentation and measurement technology conference (I2MTC).
Paris), vol. 4. HeinOnline; 2015. p. 2017. Retrieved december.
IEEE; 2019. p. 1–6.
[2] Naseri F, Schaltz E, Stroe D-I, Gismero A, Farjah E. An enhanced equivalent
[32] Merkle L, Pöthig M, Schmid F. Estimate e-golf battery state using diagnostic data
circuit model with real-time parameter identification for battery state-of-charge
and a digital twin. Batteries 2021;7(1):15.
estimation. IEEE Trans Ind Electron 2022;69(4):3743–51.
[33] Tang X, Sun Y, Zhao Y, Pei W, Li N, Kong L. Digital twin based BESS state
[3] Naseri F, Farjah E, Ghanbari T. An efficient regenerative braking system based on
estimation and operating opimization. In: 2021 IEEE 5th Conference on energy
battery/supercapacitor for electric, hybrid, and plug-in hybrid electric vehicles
Internet and energy system integration (EI2). IEEE; 2021. p. 3402–5.
with BLDC motor,". IEEE Trans Veh Technol 2017;66(5):3724–38.
[34] Tran M-K, Panchal S, Khang TD, Panchal K, Fraser R, Fowler M. Concept review
[4] Held M, et al. Thermal runaway and fire of electric vehicle lithium-ion battery
of a cloud-based smart battery management system for lithium-ion batteries:
and contamination of infrastructure facility. Renew Sustain Energy Rev 2022;
Feasibility, logistics, and functionality. Batteries 2022;8(2):19.
165:112474.
[35] Miguel E, Iraola U. Module-Level Modelling Approach for a Cloudbased Digital
[5] The Danish Emergency Management Agency. Brandsikkerhed i garageanlæg,
Twin Platform for Li-Ion Batteries. 2020. p. 1–6.
oplag af litium-ion batterier og batterier til solcelleanlæg i bygninger, 02/02/
[36] Sancarlos A, Cameron M, Abel A, Cueto E, Duval J-L, Chinesta F. From ROM of
2022. 2022 [Online]. Available, https://bpst.dk/da/Byggeri/Lister/Nyheder/
electrochemistry to AI-based battery digital and hybrid twin. Arch Comput
Nyheder/2022/06/Nyhed-om-risikoen-for-brand-i-el-og-hybridbiler.
Methods Eng 2021;28(3):979–1015.
[6] Naseri F, Karimi S, Farjah E, Schaltz E. Supercapacitor management system: A
[37] Cheng G, Wei W, Liu Z. Research on intelligent operation and maintenance
comprehensive review of modeling, estimation, balancing, and protection
system of battery based on digital twin. In: 2021 2nd international conference on
techniques. Renew.Sustain Energy Rev 2022;155:111913.
computer engineering and intelligent control (ICCEIC). IEEE; 2021. p. 154–7.
[7] Di Silvestre ML, Favuzza S, Sanseverino ER, Zizzo G. How Decarbonization,
[38] Wang S, Jin S, Bai D, Fan Y, Shi H, Fernandez C. A critical review of improved
Digitalization and Decentralization are changing key power infrastructures.
deep learning methods for the remaining useful life prediction of lithium-ion
Renew Sustain Energy Rev 2018;93:483–98.
batteries. Energy Rep 2021;7:5562–74.
[8] Yu J, Liu P, Li Z. Hybrid modelling and digital twin development of a steam
[39] Soleymani A, Maltz W. Real time prediction of Li-Ion battery pack temperatures
turbine control stage for online performance monitoring. Renew Sustain Energy
in EV vehicles,". In: International electronic packaging technical conference and
Rev 2020;133:110077.
exhibition. vol. 84041. American Society of Mechanical Engineers; 2020.
[9] Gururaj H, Hegde V. Digital twinning of the battery systems—a review. Adv
V001T01A004.
Renew Energy Electr Veh 2022:139–47.
[40] Yang D, et al. A Digital Twin-Driven Life Prediction Method of Lithium-Ion
[10] Bhatti G, Mohan H, Singh RR. Towards the future of smart electric vehicles:
Batteries Based on Adaptive Model Evolution. Materials 2022;15(9):3331.
Digital twin technology. Renew Sustain Energy Rev 2021;141:110801.
[41] Baumann M, Rohr S, Lienkamp M. Cloud-connected battery management for
[11] Wang W, Wang J, Tian J, Lu J, Xiong R. Application of digital twin in smart
decision making on second-life of electric vehicle batteries. In: 2018 thirteenth
battery management systems. Chin J Mech Eng 2021;34(1):1–19.
international Conference on ecological Vehicles and renewable energies (EVER).
[12] Garg A, Panigrahi BK. Multi-dimensional digital twin of energy storage system for
IEEE; 2018. p. 1–6.
electric vehicles A brief review. Energy Storage 2021;3(6):e242.
[42] Tanizawa T, Suzumiya T, Ikeda K. Cloud-connected battery management system
[13] Singh S, Weeber M, Birke KP. Implementation of battery digital twin: approach,
supporting e-mobility. Fujitsu Sci Tech J 2015;51(4):27–35.
functionalities and benefits. Batteries 2021;7(4):78.
[43] Ngandjong AC, et al. Investigating electrode calendering and its impact on
[14] Wu B, Widanage WD, Yang S, Liu X. Battery digital twins: Perspectives on the
electrochemical performance by means of a new discrete element method model:
fusion of models, data and artificial intelligence for smart battery management
towards a digital twin of Li-Ion battery manufacturing. J Power Sources 2021;
systems. Energy AI 2020;1:100016.
485:229320.
[15] Barricelli BR, Casiraghi E, Fogli D. A survey on digital twin: Definitions,
[44] Liu C, Lombardo T, Xu J, Ngandjong AC, Franco AA. An Experimentally-Validated
characteristics, applications, and design implications. IEEE Access 2019;7:
3D Electrochemical Model Revealing Electrode Manufacturing Parameters Effects
167653–71.
on Battery Performance. Energy Storage Material 2022;54:156–63.
[16] Botín-Sanabria DM, Mihaita A-S, Peimbert-García RE, Ramírez-Moreno MA,
[45] Xu J, et al. Lithium Ion Battery Electrode Manufacturing Model Accounting for 3D
Ramírez-Mendoza RA, Lozoya-Santos JdJ. Digital twin technology challenges and
Realistic Shapes of Active Material Particles. Journal of Power Sources 2023;554:
applications: a comprehensive review. Rem Sens 2022;14(6):1335.
232294.
[17] Bergs T, Gierlings S, Auerbach T, Klink A, Schraknepper D, Augspurger T. The
concept of digital twin and digital shadow in manufacturing. Procedia CIRP,
2021;101:81–4.

21
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

[46] Lombardo T, Ngandjong AC, Belhcen A, Franco AA. Carbon-binder migration: a [75] Wei Z, et al. Machine learning-based hybrid thermal modeling and diagnostic for
three-dimensional drying model for lithium-ion battery electrodes. Energy lithium-ion battery enabled by embedded sensing. Appl Therm Eng 2022;216:
Storage Mater 2021;43:337–47. 119059.
[47] Chouchane M, Rucci A, Lombardo T, Ngandjong AC, Franco AA. Lithium ion [76] Kosmatopoulos EB, Polycarpou MM, Christodoulou MA, Ioannou PA. High-order
battery electrodes predicted from manufacturing simulations: assessing the neural network structures for identification of dynamical systems. IEEE Trans
impact of the carbon-binder spatial location on the electrochemical performance. Neural Network 1995;6(2):422–31.
J Power Sources 2019;444:227285. [77] Kalogirou SA. Artificial neural networks in renewable energy systems
[48] Husseini K, Schmidgruber N, Weinmann H, Maibaum K, Ruhland J, Fleischer J. applications: a review. Renew Sustain Energy Rev 2001;4(5):373–401.
Development of a Digital Twin for Improved Ramp-Up Processes in the Context of [78] Legaard CM, Schranz T, Schweiger G, Drgona J, Falay B, Gomes C, Iosifidis A,
Li-Ion-Battery-Cell-Stack-Formation. Procedia CIRP, 2022;106:27–32. Abkar M, Larsen PG. Constructing neural network-based models for simulating
[49] Sharma A, Kumar Tiwari M. Digital twin design and analytics for scaling up dynamical systems. ACM Comput Surv 2021:3567591. https://doi.org/10.1145/
electric vehicle battery production using robots. International Journal of 3567591.
Production Research 2022:1–35. [79] Rhodes I. A tutorial introduction to estimation and filtering. IEEE Trans Automat
[50] Deda S, Eder A, Mhetre V, Kuchling A, Greul R, Koenig O. Designing a battery Control 1971;16(6):688–706.
emulator/tester from scratch to prototyping to automated testing within a HIL [80] Telford JK. A brief introduction to design of experiments. Johns Hopkins APL
digital twin environment. In: PCIM Europe digital days 2020; International Tech Dig 2007;27(3):224–32.
exhibition and conference for power electronics, intelligent motion, renewable [81] Adaikkappan M, Sathiyamoorthy N. Modeling, state of charge estimation, and
energy and energy management. VDE; 2020. p. 1–8. charging of lithium-ion battery in electric vehicle: a review. Int J Energy Res
[51] Park J, et al. Digital twin-driven all-solid-state battery: Unraveling the Physical 2022;46(3):2141–65.
and Electrochemical Behaviors. Adv Energy Mater 2020;10(35):2001563. [82] Tran MK, et al. Python-based scikit-learn machine learning models for thermal
[52] Xu Z, Xu J, Guo Z, Wang H, Sun Z, Mei X. Design and optimization of a novel and electrical performance prediction of high-capacity lithium-ion battery. Int J
microchannel battery thermal management system based on digital twin. Energy Res 2022;46(2):786–94.
Energies 2022;15(4):1421. [83] Hu J, He H, Wei Z, Li Y. Disturbance-immune and aging-robust internal short
[53] Li W, et al. Deep reinforcement learning-based energy management of hybrid circuit diagnostic for lithium-ion battery. IEEE Trans Ind Electron 2021;69(2):
battery systems in electric vehicles. J Energy Storage 2021;36:102355. 1988–99.
[54] Li W, et al. Cloud-based health-conscious energy management of hybrid battery [84] Tao L, Ma J, Cheng Y, Noktehdan A, Chong J, Lu C. A review of stochastic battery
systems in electric vehicles with deep reinforcement learning. Appl Energy 2021; models and health management. Renew Sustain Energy Rev 2017;80:716–32.
293:116977. [85] Wang Y, et al. A comprehensive review of battery modeling and state estimation
[55] Bai Y, Muralidharan N, Sun Y-K, Passerini S, Whittingham MS, Belharouak I. approaches for advanced battery management systems. Renew Sustain Energy
Energy and environmental aspects in recycling lithium-ion batteries: Concept of Rev 2020;131:110015.
Battery Identity Global Passport. Mater Today 2020;41:304–15. [86] Thelen A, et al. A comprehensive review of digital twin—part 1: modeling and
[56] Clark S, et al. Toward a unified description of battery data. Adv Energy Mater twinning enabling technologies. Struct Multidisciplin Optimizat 2022;65(12):
2022;12(17):2102702. 354.
[57] Anandavel S, Li W, Garg A, Gao L. Application of digital twins to the product [87] Thelen A, et al. A comprehensive review of digital twin—part 2: roles of
lifecycle management of battery packs of electric vehicles,. IET Collab Intell uncertainty quantification and optimization, a battery digital twin, and
Manuf 2021;3(4):356–66. perspectives. Struct Multidisciplin Optimizat 2023;66(1):1.
[58] Leng J, et al. Blockchain-empowered sustainable manufacturing and product [88] Singh S, Weeber M, Birke K-P. Advancing digital twin implementation: a toolbox
lifecycle management in industry 4.0: A survey. Renew Sustain Energy Rev 2020; for modelling and simulation. Procedia CIRP, 2021;99:567–72.
132:110112. [89] Leng J, Wang D, Shen W, Li X, Liu Q, Chen X. Digital twins-based smart
[59] Teng SY, Touš M, Leong WD, How BS, Lam HL, Máša V. Recent advances on manufacturing system design in Industry 4.0: a review. J Manuf Syst 2021;60:
industrial data-driven energy savings: Digital twins and infrastructures. Renew. 119–37.
Sustain Energy Rev 2021;135:110208. [90] Tao F, Zhang M. Digital twin shop-floor: a new shop-floor paradigm towards
[60] Yu W, Patros P, Young B, Klinac E, Walmsley TG. Energy digital twin technology smart manufacturing. IEEE Access 2017;5:20418–27.
for industrial energy management: Classification, challenges and future. Renew [91] Feng H, Gomes C, Thule C, Lausdahl K, Iosifidis A, Larsen PG. Introduction to
Sustain Energy Rev 2022;161:112407. digital twin engineering. In: 2021 annual Modeling and simulation conference
[61] Meng J, Yue M, Diallo D. Nonlinear extension of battery constrained predictive (ANNSIM). IEEE; 2021. p. 1–12.
charging control with transmission of Jacobian matrix. Int J Electr. Power Energy [92] Grieves M. Digital twin: manufacturing excellence through virtual factory
Syst 2023;146:108762. replication. White paper 2014;1:1–7. 2014.
[62] Gao Y, et al. Health-aware multiobjective optimal charging strategy with coupled [93] Tao F, Zhang M, Liu Y, Nee AY. Digital twin driven prognostics and health
electrochemical-thermal-aging model for lithium-ion battery. IEEE Trans Ind Inf management for complex equipment. Cirp Annals 2018;67(1):169–72.
2019;16(5):3417–29. [94] Feng HG, Claudio, Gil Santiago, Mikkelsen Peter H, Tola Daniella, Larsen Peter
[63] Morstyn T, Crozier C, Deakin M, McCulloch MD. Conic optimization for electric Gorm. Integration of the MAPE-K loop in digital twins. In: Presented at the 2022
vehicle station smart charging with battery voltage constraints. IEEE Tran Trans annual modeling and simulation conference (ANNSIM), San Diego, California,
Elect 2020;6(2):478–87. USA; 2022.
[64] Wei Z, Quan Z, Wu J, Li Y, Pou J, Zhong H. Deep deterministic policy gradient- [95] Merkle L, Segura AS, Grummel JT, Lienkamp M. Architecture of a digital twin for
DRL enabled multiphysics-constrained fast charging of lithium-ion battery. IEEE enabling digital services for battery systems. In: 2019 IEEE international
Trans Ind Electron 2021;69(3):2588–98. conference on industrial cyber physical systems (ICPS). IEEE; 2019. p. 155–60.
[65] Narasipuram RP, Mopidevi S. A technological overview & design considerations [96] Yang S, et al. Implementation for a cloud battery management system based on
for developing electric vehicle charging stations. J Energy Storage 2021;43: the CHAIN framework. Energy AI 2021;5:100088.
103225. [97] Parra RV, et al. Digital twin-driven framework for EV batteries in automobile
[66] Mopidevi S, Narasipuram RP, Aemalla SR, Rajan H. E-mobility: impacts and manufacturing. In: Transdisciplinary Engineering for resilience: Responding to
analysis of future transportation electrification market in economic, renewable system disruptions: Proceedings of the 28th ISTE international Conference on
energy and infrastructure perspective. Int J Powertrains 2022;11(2–3):264–84. transdisciplinary engineering, July 5–July 9, 2021, vol. 16. IOS Press; 2021.
[67] Yoro KO, Daramola MO, Sekoai PT, Wilson UN, Eterigho-Ikelegbe O. Update on p. 181.
current approaches, challenges, and prospects of modeling and simulation in [98] Bedi G, Venayagamoorthy GK, Singh R, Brooks RR, Wang K-C. "Review of Internet
renewable and sustainable energy systems. Renew Sustain Energy Rev 2021;150: of Things (IoT) in electric power and energy systems. IEEE Internet Things J
111506. 2018;5(2):847–70.
[68] Nikdel M. Various battery models for various simulation studies and applications. [99] Li S, Da Xu L, Zhao S. "5G internet of things: a survey. J Ind Inf Int 2018;10:1–9.
Renew Sustain Energy Rev 2014;32:477–85. [100] Gil S, Zapata-Madrigal GD, García-Sierra R, Cruz Salazar LA. Converging IoT
[69] Li Y, Wei Z, Xiong B, Vilathgamuwa DM. Adaptive ensemble-based protocols for the data integration of automation systems in the electrical industry.
electrochemical–thermal degradation state estimation of lithium-ion batteries. J Elect Sys Inf Tech 2022;9(1):1–21.
IEEE Trans Ind Electron 2021;69(7):6984–96. [101] Padgavankar M, Gupta S. Big data storage and challenges. Int J comput sci info
[70] Wei Z, Hu J, He H, Yu Y, Marco J. Embedded distributed temperature sensing technol 2014;5(2):2218–23.
enabled multistate joint observation of smart lithium-ion battery. IEEE Trans Ind [102] Mazumdar S, Seybold D, Kritikos K, Verginadis Y. A survey on data storage and
Electron 2022;70(1):555–65. placement methodologies for cloud-big data ecosystem. J Big Data 2019;6(1):
[71] Lin J, Liu X, Li S, Zhang C, Yang S. A review on recent progress, challenges and 1–37.
perspective of battery thermal management system. Int J Heat Mass Tran 2021; [103] Damjanovic-Behrendt V, Behrendt W. An open source approach to the design and
167:120834. implementation of Digital Twins for Smart Manufacturing. Int J Comput
[72] Paszke A, et al. Pytorch: an imperative style, high-performance deep learning Integrated Manuf 2019;32(4–5):366–84.
library. Adv Neural Inf Process Syst 2019;32. [104] Autiosalo J, Siegel J, Tammi K. Twinbase: open-source server software for the
[73] Abadi M, et al. Tensorflow: large-scale machine learning on heterogeneous digital twin web. IEEE Access 2021;9:140779–98.
distributed systems. 2016. arXiv preprint arXiv:1603.04467. [105] Bader SR, Maleshkova M. The semantic asset administration shell. In:
[74] Mittal S, Tolk A, Pyles A, Van Balen N, Bergollo K. Digital twin modeling, co- International Conference on Semantic Systems. Springer; 2019. p. 159–74.
simulation and cyber use-case inclusion methodology for IoT systems. In: 2019 [106] Agrawal R, Kadadi A, Dai X, Andres F. Challenges and opportunities with big data
Winter Simulation Conference (WSC). IEEE; 2019. p. 2653–64. visualization. In: Proceedings of the 7th International Conference on Management

22
F. Naseri et al. Renewable and Sustainable Energy Reviews 179 (2023) 113280

of computational and collective intElligence in Digital EcoSystems; 2015. [126] Kazemi Z, Safavi A, Pouresmaeeli S, Naseri F. A practical framework for
p. 169–73. implementing multivariate monitoring techniques into distributed control
[107] Chan WW-Y. A survey on multivariate data visualization, Vol. 8. Department of system,. Control Eng Pract 2019;82:118–29.
Computer Science and Engineering. Hong Kong University of Science and [127] Kulik T, Boudjadar J, Tran-Jørgensen PW. Security verification of industrial
Technology; 2006. p. 1–29. no. 6. control systems using partial model checking,". In: Proceedings of the 8th
[108] Zhang C, Zhu Y, Dong G, Wei J. Data-driven lithium-ion battery states estimation International Conference on Formal Methods in Software Engineering; 2020.
using neural networks and particle filtering. Int J Energy Res 2019;43(14): p. 98–108.
8230–41. [128] Krutz RL. Securing SCADA systems. John Wiley & Sons; 2005.
[109] Hu X, Feng F, Liu K, Zhang L, Xie J, Liu B. State estimation for advanced battery [129] Kazemi Z, Safavi AA, Naseri F, Urbas L, Setoodeh P. A secure hybrid dynamic-
management: key challenges and future trends. Renew Sustain Energy Rev 2019; state estimation approach for power systems under false data injection attacks.
114:109334. IEEE Tran Ind Inf 2020;16(12):7275–86.
[110] Bartocci E, et al. Specification-based monitoring of cyber-physical systems: a [130] Kazemi Z, Safavi AA, Arefi MM, Naseri F. Finite-time secure dynamic state
survey on theory, tools and applications. In: Lectures on Runtime Verification. estimation for cyber-physical systems under unknown inputs and sensor attacks.
springer; 2018. p. 135–75. IEEE transactions on systems, man, and cybernetics: systems 2022;52(8):4950–9.
[111] Mo Y, et al. Cyber–physical security of a smart grid infrastructure. Proc IEEE [131] Pajic M, et al. Robustness of attack-resilient state estimators. In: 2014 ACM/IEEE
2011;Vol. 100(1):195–209. international Conference on cyber-physical systems (ICCPS). IEEE; 2014.
[112] Paridari K, O’Mahony N, A E, Mady D, Chabukswar R, Boubekeur M, Sandberg H. p. 163–74.
A framework for attack-resilient industrial control systems: Attack detection and [132] Eckhart M, Ekelhart A. A specification-based state replication approach for digital
controller reconfiguration. Proc IEEE 2017;Vol. 106(1):113–28. twins. In: Proceedings of the 2018 workshop on cyber-physical systems security
[113] Ziade H, Ayoubi RA, Velazco R. A survey on fault injection techniques. Int Arab J and privacy; 2018. p. 36–47.
Inf Technol 2004;1(2):171–86. [133] Bécue A, et al. CyberFactory# 1—securing the industry 4.0 with cyber-ranges and
[114] Dai H, Jiang B, Hu X, Lin X, Wei X, Pecht M. Advanced battery management digital twins. In: 2018 14th IEEE international Workshop on factory
strategies for a sustainable energy future: multilayer design concepts and research communication systems (WFCS). IEEE; 2018. p. 1–4.
trends. Renew Sustain Energy Rev 2021;138:110480. [134] Moreira N, Molina E, Lázaro J, Jacob E, Astarloa A. Cyber-security in substation
[115] Gomes C, Thule C, Broman D, Larsen PG, Vangheluwe H. Co-simulation: a survey. automation systems,. Renew Sustain Energy Rev 2016;54:1552–62.
ACM Comput Surv (CSUR) 2018;51(3):1–33. [135] Jasiūnas J, Lund PD, Mikkola J. Energy system resilience–A review. Renew
[116] Keskisärkkä R. Semantic complex event processing for decision support. In: Sustain Energy Rev 2021;150:111476.
International Semantic Web Conference. Springer; 2014. p. 529–36. [136] Kim T, et al. An overview of cyber-physical security of battery management
[117] Cámara J, De Lemos R, Ghezzi C, Lopes A. Assurances for self-adaptive systems: systems and adoption of blockchain technology. IEEE J Emerg Selected Topics
principles, models, and techniques. Springer; 2013. Power Electron 2022;10(1):1270–81.
[118] Weyns D. Software engineering of self-adaptive systems. In: Handbook of [137] Ansys. "Ansys twin builder | create and deploy digital twin models." ANSYS. http
software engineering. Springer; 2019. p. 399–443. s://www.ansys.com/products/digital-twin/ansys-twin-builder (accessed July
[119] Kurzhanski AB, Varaiya P. Ellipsoidal techniques for reachability analysis. In: 2022.
International workshop on hybrid systems: Computation and control. Springer; [138] Ansys. "How to build a digital twin from a systems simulation." Ansys. https
2000. p. 202–14. ://www.ansys.com/blog/how-to-build-a-digital-twin-from-a-systems-simulation
[120] Asarin E, Bournez O, Dang T, Maler O. Approximate reachability analysis of (accessed August 10, 2022.
piecewise-linear dynamical systems. In: International workshop on hybrid [139] COMSOL. “Digital twins and model-based battery design.”. https://www.comsol.
systems: Computation and control. Springer; 2000. p. 20–31. com/blogs/digital-twins-and-model-based-battery-design/. [Accessed 15 August
[121] Girard A. Reachability of uncertain linear systems using zonotopes. In: 2022]. COMSOL.
International Workshop on Hybrid Systems: Computation and Control. Springer; [140] Qi Q, et al. Enabling technologies and tools for digital twin. J Manuf Syst 2021;58:
2005. p. 291–305. 3–21.
[122] Kephart JO, Chess DM. The vision of autonomic computing. Computer 2003;36 [141] Liu M, Fang S, Dong H, Xu C. Review of digital twin about concepts, technologies,
(1):41–50. and industrial applications. J Manuf Syst 2021;58:346–61.
[123] Arcaini P, Riccobene E, Scandurra P. Modeling and analyzing MAPE-K feedback [142] I. D. 23247–1. Automation systems and integration–digital twin framework for
loops for self-adaptation. In: 2015 IEEE/ACM 10th international Symposium on manufacturing–Part 1: overview and general principles. In: International
software Engineering for Adaptive and self-managing systems. IEEE; 2015. organization for standardization Geneva, Switzerland; 2020.
p. 13–23. [143] Lehner D, et al. Digital twin platforms: requirements, capabilities, and future
[124] Kumbhar S, Faika T, Makwana D, Kim T, Lee Y. Cybersecurity for battery prospects. IEEE Softw 2021;39(2):53–61.
management systems in cyber-physical environments. In: 2018 IEEE [144] Ye X, Hong SH. Toward industry 4.0 components: insights into and
transportation electrification Conference and expo (ITEC). IEEE; 2018. p. 934–8. implementation of asset administration shells. IEEE Ind Elect Magn 2019;13(1):
[125] Mo Y, Chabukswar R, Sinopoli B. Detecting integrity attacks on SCADA systems. 13–25.
IEEE Trans Control Syst Technol 2013;4:1396–407.

23