International Journal of Information Security (2024) 23:1171–1195

https://doi.org/10.1007/s10207-023-00784-x

REGULAR CONTRIBUTION

A review on digital twins for power generation and distribution

Jessica B. Heluany1 · Vasileios Gkioulos1

Published online: 1 December 2023

© The Author(s) 2023

Abstract
This paper presents a systematic literature review on the application of digital twins in the energy sector. Initially, we
generated an overview through a survey of prior reviews, independent of market vertical, then followed by a more detailed
review concentrating on the power production and distribution domains, as per the NIST (National Institute of Standards
and Technology) smart grid standard. We implemented a rigorous method, which included seven stages, beginning with the
collection of 2238 articles. We observed that the energy sector range was too broad and filtered by generation and distribution
during the practical screening, resulting in 275 for further screening. This amount was then condensed to 81 papers that
matched the quality screening criteria for synthesis and examination. In summary, digital twin architectures and frameworks
include five components: the physical entity, bidirectional communication, the virtual entity (with modeling and simulation),
data management, and services. Our study contributed by determining that distribution management is the most pertinent
application of digital twins in the distribution domain and fault diagnosis in the generation domain. Furthermore, we found
that digital twins involve multiple stakeholders whose role is rarely discussed in studies, and we identified a similar absence
of emphasis for security. Research on security often presents the digital twin as an additional layer of protection, yet rarely
investigates the security of the digital twin by itself. The potential limitations of our study to answer some of the technical
research questions may be because of the criteria for the selection of papers. However, as the emphasis of this study is on the
energy sector, it enabled domain-specific findings for generation and distribution.

Keywords Digital twin · Smart grid · Generation · Distribution

1 Introduction tructure functional in adverse situations that may include

a wide range of actors, equipment, systems, markets, ser-
Critical infrastructures are the main sectors that provide vices, and stakeholders. Moreover, according to the NIST
means for us to keep our way of life. Their malfunction framework for smart grids, investments on this study field
can debilitate a country’s safety and/or security in different are estimated to generate economic benefits that can achieve
forms, reason for which CISA (Cybersecurity Infrastructure billions of dollars [2]. Thus, possible technologies that may
and Security Agency) states that from all critical sectors, be suitable to address the energy sector challenges, such as
the energy sector is of unique criticality. Their argument for digital twins, are worth exploring. Between the advantages
such uniqueness is that “it provides an ‘enabling function’ of a digital twin implementation, we can mention real-time
across all critical infrastructure sectors” [1]. Given this con- simulation to improve decision making, performance, diag-
text, it is clear the importance of improving the resilience nosis, among others. This makes them suitable for addressing
and reliability of the power grid to keep this critical infras- common problems of the energy sector, help in the energy
transition, and upmost, provide more resilience to the grid.
B Jessica B. Heluany This illustrates the value of exploring current literature to
[email protected] identify the advancements and applications of digital twin
Vasileios Gkioulos technologies within the energy sector.
[email protected] In light of this, our study seeks to cover the following
1
topics:
Department of Information Security and Communication
Technology (IIK), Faculty of Information Technology and
Electrical Engineering (IE), NTNU – Norwegian University
of Science and Technology, Gjøvik, Norway

123
1172 J. B. Heluany, V. Gkioulos

– To identify and classify research papers approaching dig-

ital twin concepts and applications in the energy sector;
– To identify design architectures, operational paradigms,
current tools, and modeling techniques for the develop-
ment of digital twins;
– To identify how the digital twin security has been
addressed;
– To investigate existing applications and identify use cases
that can contribute to foster digital twin application in the
energy sector;

By researching these topics, this work will generate fresh

insights into some challenges of the power systems, since Fig. 1 NIST smart grid conceptual model
the identified use cases reflect relevant problems faced by
this sector, and the topics that cover digital twin technical
aspects are helpful for future implementations. Interest- vice providers. The participants within each domain are
ingly, as stated by [3], several governments are planning to named Actors. Therefore, a given power plant, for example,
incorporate DERs (Distributed Energy Resources) as part is an actor of the generation domain. Figure 1 summarizes
of the energy transition strategy, specially from renewable the domains and the communication and electricity flows
generation, which can be weather dependent resulting in between them.
intermittent production. This fact raises even more a chal- Another aspect that this standards highlights is the inter-
lenge that is widely discussed by the scientific community: operability, defined as “the capability of two or more net-
the growing management complexity of the grid. In a lit- works, systems, devices, applications, or components to work
erature review of power systems challenges, besides the together, and to exchange and readily use information—
distribution management, [4] also pointed climate changes securely, effectively, and with little or no inconvenience to
and environmental conditions (that are related to the dis- the user.”
cussion of DERs incorporation by [3]), disturbances or Once a smart grid is constituted of an increasing array of
unexpected events from faults sources, and cyber-attacks. interoperable systems, the ability to share information plays
Considering the possible cascading effects of such risks, [5] a vital role in the continuous digitalization, and the need
argued that smart grids naturally leads to decision making by for the energy transition, requires a consistent understanding
stakeholders that can be assisted by data-driven algorithms, of the grid language to improve the communication across
resulting in yet another challenge of “data-driven decision stakeholders, including solutions of hardware, software, and
support models.” Therefore, we can see that digital twins services.
can be an enabling technology to address the power sector From the digitalization perspective, digital twin appeared
challenges, considering that they would be a virtual replica of in many technology trends reports in the last years. A brief
an entity of interest, allowing real-time simulations that can way of describing digital twins is that they constitute a virtual
guide stakeholders decisions and provide solutions to some replica of a real entity. By doing so, this technology would
of these challenges. enable a series of applications that might be helpful for further
developments of smart grids. Digital twin has been listed by
1.1 Background Gartner as a technology trend for three consecutive years,
from 2017 to 2019 [7–9], and is still part of the trends within
The reason for naming grids as smart relies on the fact that Hyperautomation in 2020 Gartner report [10]. Digital twins
more and more digital components and services are being have also been a popular topic with increasing publications
applied in the electricity network. The US Department of in academic journals [11].
Energy’s Advanced Grid Research [6], states that what makes In a larger context, European initiatives such as the Euro-
a grid smart is the “digital technology that allows for two-way pean Technology & Innovation Platforms (ETIPs) for Smart
communication between the utilities and its customers, and Networks for Energy Transition (SNET), argue that digital
the sensing along the transmission lines.” However, the term technologies are enablers to achieve a better system and put
smart grids is very broad, approaching all domains of the digital twin and energy transition at the core of a low-carbon
energy sector. The NIST conceptual model of smart grids [2] economy [12].
defines seven main domains: system transmission operators
(TSOs); system distribution operators (DSOs); generation
(including DERs); customer; markets; operations and ser-

123
A review on digital twins for power generation and distribution 1173

1.2 Motivation The literature search was conducted on two applicable bib-
liographic databases: ScienceDirect and IEEEXplore, using
The topics covered in the previous sections highlight the the title query (“digital twin” AND (“review” OR “survey”
importance of the energy sector digitalization and its con- OR “state of the art”)). Only scientific/technical articles were
nection with digital twins as one of the leading technology examined, leading to 139 studies that were reduced to 41
trends of digitalization. However, as explained in the Back- after applying format and content criteria during practical
ground, the energy sector is composed of seven domains, screening, where Specific use cases and indirect studies were
each with several possible actors. Regarding digital twins, excluded. These studies were filtered by rigorous systematic
the possible applications are equally broad, making it hard review methodologies remaining 21 for further analysis.
to see what advantages of digital twins are unique for smart
grids. Our main motivation is to better understand how digital 2.1 Definition considerations
twins can contribute to specific smart grid domains, map-
ping what applications are more relevant according to the The term digital twin has been around for approximately 20
selected domains. Given the author’s knowledge background, years, since Grieves first proposed it in 2002 [13]. However,
distribution and generation were selected for further investi- no common definition has been accepted by both academic
gation. We aim to contribute in this research topic by mapping and industrial communities. To demonstrate and discuss the
the most common digital twins applications for these two lack of consensus and standardization, some authors com-
domains, and also evaluating the most common actors for piled multiple definitions into tables to evaluate similarities
each domain because it provides information on what are the and differences [14–19], but this effort did not change the
real entities and how this can impact the way the digital twin most accepted definition, which was formalized by NASA
is implemented for each use case. in 2012, and is cited in 15 from the 29 papers analyzed [11,
14–27]. According to NASA’s definition, a digital twin is
“an integrated multi-physics, multi-scale, probabilistic sim-
1.3 Structure of the paper
ulation of a vehicle or system that uses the best available
physical models, sensor updates, fleet history, etc., to mirror
This paper is organized in seven sections, starting from the
the life of its flying twin. The digital twin is ultra-realistic
introduction in Sect. 1, digital twin conceptualization in Sect.
and may consider one or more important and interdependent
2, and related work in Sect. 3 to provide an overview, followed
vehicle systems” [28].
by an explanation of the method applied for this systematic
It is out of the scope of this study to find the com-
literature review in Sect. 4. The findings are discussed in
mon ground between definitions, but some scholars have
Sect. 5, organized according to the research questions. Sec-
attempted to do so. As established by the meta-analysis in
tion 7 outlines some possibilities for future work, and Sect. 8
[29], the authors inferred that the common elements among
is dedicated to discussion, concluding our analysis.
the numerous definitions are i) virtual representation, ii) bidi-
rectional connection between the real and virtual entities, iii)
simulation, and iv) connection across all life cycle stages.
2 Main aspects of digital twins Upon analyzing these elements in correlation to NASA’s
definition, and contrasting them with published use cases,
To gain an in-depth comprehension of digital twins’ con- it becomes apparent that not all digital twin applications
ceptualization and applications, this research started with an employ all these elements, thus indicating a misuse of the
analysis of previous reviews regardless of the market verti- term. As an example, some studies focus solely on modeling
cal. The studies were analyzed aiming to map the application and/or simulating a system of interest without considering
domains, architectures and frameworks, tools, and security. the bidirectional communication, while explicitly referring
Definition evaluation was not initially included in the to developing a digital twin instance.
objectives, but was quickly identified as a key factor in ensur- In an effort to address this issue, a classification based on
ing that understanding and expectations are in sync for those the level of data integration was proposed in [30], with the
involved in the study of digital twins. Following this dis- data flow exchange between the digital and physical objects
cussion, further study was conducted into the application serving as the principal criterion. This work gave rise to
domains and use cases of digital twins. the terms Digital Model, Digital Shadow, and Digital Twin,
A review of the primary tools and techniques employed which have been cited in multiple other researches. Nev-
for modeling and simulation, as well as common architec- ertheless, there are other components that are not present
tures and frameworks, was conducted to assess the feasibility in this approach (such as the fidelity of the virtual repre-
of creating digital twins. Furthermore, it was determined if sentation), hindering the differentiation between traditional
security was being addressed or not. solutions (i.e., simulation) and digital twins, and, in turn,

123
1174 J. B. Heluany, V. Gkioulos

Fig. 2 System design hierarchy

and its knowledge domains [31]

prolonging the market time for more complex real-world 2.2 Application domains
applications.
It is our opinion that the misconception about the defini- With the advancement of research on digital twins, it has
tion of digital twins has its roots in the obscure meanings become possible to ascertain patterns and the most recurrent
of the words “ultra-realistic,” “mirror,” and “system,” and its fields of study. Some authors list the papers analyzed in their
relationship to complex systems organization. When refer- review according to market verticals [15, 20, 23], while others
ring to complex engineering systems, if no model is clearly categorize per lifecycle phase [18, 32]. Additionally, some
defined at a particular abstraction layer, there is an inher- reviews are already based on a specific market vertical and
ently ambiguity associated with the term “system” and its they list use cases without referring to the lifecycle phase
constituent parts that would be “mirrored.” [14, 17, 22, 33, 34].
In Fig. 2, extracted from [31], the author defines a sys- The outcomes of such studies vary depending on the sector
tem design hierarchy as composed of: parts, subcomponents, and the phase of the lifecycle that the digital twin addressed.
components, subsystems, and systems. Along with this struc- The investigation conducted in [35] (2022), which analyzed
ture, several knowledge domains must be taken into consider- 42 papers, revealed that manufacturing and energy were the
ation, evidencing how complex a digital twin can be assuming verticals with the most publications, followed by aerospace
a high-fidelity virtual replica that considers all these lev- and automotive. From a lifecycle perspective, [18] mapped
els and knowledge domains. In this way, thinking about a 240 papers published from 2010 to 2019, against each life-
“mirror” does not necessarily relate to “ultra-realistic” in cycle phase concluding that the amount of publications is
all levels, which, in turn, conveys the idea of modeling a increasing especially for the production and service phases.
given piece of the hierarchy for all knowledge domains that It is worth noting this reported increase for the services
could go down to atomic levels. Thus, it is understandable phase, given that [33] analyzed applications in this area with
why scholars simplify the use cases, misusing the term, a cutoff date of December 2018, explicitly highlighting a
and approaching real systems of interest at the component need for digital twins for services, since in several indus-
level and only for specific knowledge domains. At the same tries the profit margin from services frequently surpasses the
time, it reinforces the need for a more clear definition that margin from product sale. Nonetheless, over the years, [18]
leads this technology to a higher maturity level, instead of shows that there has been a meaningful increase in produc-
calling digital twin many traditional solutions that are in tion/manufacturing, and service phase, approaching the gap
place for years, such as modeling, simulation, and monitor- indicated by [33].
ing. Taking the work [18] as a basis and compiling results from
other papers, it is possible to summarize common use cases
that might be applicable for more than one lifecycle phase:

123
A review on digital twins for power generation and distribution 1175

Fig. 3 Example of digital twin architecture extracted from [38]

– Verification; optimization; validation; production control – Building blocks: potentially reusable component that can
and planning; what-if analysis; predictive maintenance; be combined with other building blocks to deliver archi-
new business and business models; reduce of capital tectures and solutions.
investment; improvement of flexibility to adapt to spe-
cific consumer needs; improve brand loyalty; real-time
state monitoring; asset management; traceability; data
Given that TOGAF conceptualization is broader than our
management; man–machine interaction; costs reduc-
architectural scope, we also based this discussion on authors
tion; improvement of vertical and horizontal integration;
who define architecture and framework from a computer
health monitoring and analysis; support after sales recon-
science view. If taken from programming perspective, a
figuration; fault detection and diagnosis; reduction of
framework is related to implementation of an architecture,
operational downtime; improvement of change manage-
providing multiple classes that abstract particular concepts,
ment of documents and assets; increase of customer
defines how these abstraction layers work together, provides
interaction and support.
reusable components for application-specific features, and
organizes patterns at higher levels [37].
2.3 Architectures and frameworks It can be argued that with no consensus on the defini-
tion of digital twins, the way the architecture and framework
We have noticed that some papers apply the terms architec- terms are used varies across the studies, although these could
ture and framework interchangeably, yet they have distinct be two distinct cases of term misuse. However, there is a
meanings. According to TOGAF standard definition [36], basis of agreement that a typical architecture has the pres-
a framework “provides the methods and tools for assisting ence of a physical and a virtual entity, in addition to the
in the acceptance, production, use, and maintenance of an relationship that connects them. An example of an archi-
Enterprise Architecture.” In our case, we are not dealing tecture that contains these dimensions is shown in Fig. 3
with an enterprise architecture, but with one of the possible extracted from [38]. Regarding the framework and assum-
abstraction levels. In this sense, we can assume the logi- ing a combination of the TOGAF definition and as described
cal abstraction level for a digital twin system, referring to by [37] in the previous paragraph, we assume that a digital
the design, i.e., the structure, data flows, functions, rules, twin framework approaches the architecture implementa-
and methods that guide the implementation. The TOGAF tion and would be composed by building blocks that are
architecture framework defines three categories with types potentially reusable. This premise is aligned with the frame-
of architectural work: work taken as reference in Fig. 4, where we observe multiple
building blocks for the physical entity platform, virtual
entity platform, data management platform, and services
– Deliverable: a work defined contractually by stakehold- platform. Each of the sub constituents of these building
ers, blocks may be reusable in the same project, or even dif-
– Artifact: a work that describes an aspect of the architec- ferent projects depending on how they adhere to each
ture and may or may not be considered a deliverable, context.

123
1176 J. B. Heluany, V. Gkioulos

Fig. 4 Example of digital twin framework extracted from [38]

Nonetheless, these terms are not observed in a unified 2.4 Tools

manner across the studies. Actually, attempting to detail
digital twin layers, a wide range of terms have been used, This section maps tools based on categories that were found
including “properties,” “characteristics,” “enabling technolo- during the development of this study. Recalling the archi-
gies,” among others. Table 1 from our previous work [39] tectural functional blocks, similar blocks are applied for the
is an extraction of different authors summarizing how they tools, ranging from the integration between the physical and
refer to the layers of further characterization. Upon evalu- virtual entities, data-related services, modeling, and simula-
ating the description provided by the authors, it is possible tion.
to identify overlapping terms even if they are categorized It is important to keep in mind that some solutions, such
differently. For example, [19] labels the update frequency as cloud services from Microsoft and Amazon, can fit more
as a dimension of the digital twin, while in [15, 20] they than one category. This is because some businesses provide
label as characteristics and utilize the term “twinning rate” a variety of solutions. The identification of all tools and their
instead. Similarly, some authors seek to detail digital twins by associated categories, though, is outside the purpose of this
enumerating enabling technologies. In [23], they classified study. Additionally, the tool set changes depending on the
the technologies according to domains (application, middle- market vertical, and small fixes could be included in a longer
ware, network, and object), but in [17], the enablers were list. Based on the research of [26] and the additional attribu-
listed without association do domains or building blocks. tion of tools listed in other studies to the suggested categories,
The aim of this table is not to suggest a new terminology or Table 2 [39] offers an overview, where the solution suppliers
categorize digital twins features in groups, but to highlight are identified in brackets.
how the lack of consensus on digital twin concept impacts
the way it is described and detailed across the studies. Along 2.5 Security
this table, higher group labels are separated with semicolons,
and terms used for these groups are separated with colons, Security was mentioned as a possible solution provided by
giving an idea of the proposed hierarchy in the studies. digital twins [16, 20, 41], but there is no discussion whether
This may be a consequence of the researchers’ varied back- this would be related to the security of the digital twin itself.
grounds, but it makes harder for the scientific community to Although some authors have stated that digital twins require
have an intuitive understanding of a digital architecture and an additional security strategy due to the introduction of new
framework, and it demonstrates the need for future research vulnerabilities [18, 22], most reviews have only mentioned
on standardizing all terms relevant for digital twin detail- it as a challenge [15, 23, 29, 32, 42]. Section 5.5 provides
ing. further discussion on this topic for the energy sector papers.

123
A review on digital twins for power generation and distribution 1177

Table 1 Compilation of terms used to detail digital twins in several studies highlighting the lack of uniformity related to the technology conceptu-
alization and implementation
References DT details label Terms used to describe DT features

[15, 20] Characteristics Integrated system, clone, counterpart, ties, links, description, construct, information,
simulation, test, prediction; virtual mirror, replica, physical entity, physical twin; virtual
entity, virtual twin, physical environment, virtual environment, state, realization,
metrology, twinning, twinning rate, physical-to-virtual connection/twinning,
virtual-to-physical connection/twinning, physical processes, virtual processes)
[20] Parameters Form, functionality, health, location, process, time, state, performance, environment
[16] Components radio-frequency identification, wireless sensor networks, radio-frequency identification
sensor networks, unit level, system level, system of systems level, middleware
(service-oriented architecture), communication protocol, communication protocol
interface (AutomationML), wireless communication, programming interface through
application programming interface, data-driven methods, geometry model, physical
model, behavior model, collaborative information model, decision making model,
scalability, model interoperability, fidelity, dynamicity, modularity, application interface
layer
[23] Domains, Enabling technologies Application Domain: model architecture and visualization, software and Application
Programming Interfaces, data collection and pre-processing; Middleware Domain: storage
technology, data processing; Network Domain: communication technology, wireless
communication; Object Domain: hardware platform, sensor technology
[11] Category, Dimensions Context: reference object, tangible product life cycle phase, benefits, application domain;
Data: data storage, data scope, data quality, data sources, data interpretation; Computing
capabilities: trigger types, model look-ahead perspective, computing timing capabilities,
update frequency of inputs, update frequency of outputs; Model: digital twin creation
approach, modeled characteristics, digital model types, model authenticity, model
maintenance, modularity; Integration: digital twin interaction, hierarchy, connection
mode, user focus, inter-organizational integration, collaboration; Control: level of
cognition, level of autonomy, learning capabilities; human–machine interaction: types of
interaction devices, human interaction capabilities
[17] Enablers Artificial intelligence, Internet of things, industrial internet of things, virtual reality,
augmented reality, hardware, communication technologies, knowledge building, design
process, development technologies
[40] Building blocks, Properties Physical Entity Platform: physical object (is observed), physical node (observes), human;
Virtual Entity Platform: semantic model with geometric model, physical model,
behavioral model, rule model, process model; Data Management Platform: data models,
data management methods; Service Platform: service models (physical/virtual), service
management layers
[19] Dimension, Level Update frequency: immediate real-time, event driven, every day, every week; Connectivity
modes: automatic, bidirectional, unidirectional; Integration breadth: world (full object
interaction), field environment, near field production system, product, machine; Product
lifecycle: begin of life, mid of life, end of life; Human-interaction: smart devices, virtual
reality, augmented reality, smart hybrid; Digital model richness: geometry, kinematics,
control behavior, multi-physical behavior; Simulation capabilities: look-ahead
perspective, Ad-Hoc, dynamic, static; Cyber Physical System intelligence: autonomous,
partial autonomous, automated, human triggered
[18] Key technologies Data-related technologies, high-fidelity modeling technologies, model based simulation
technologies
[26] Supporting tools types Integration and simulation, digital twin modeling, bridging and twin control, big data
processing, big data storage, artificial intelligence–machine learning and application
programming interfaces

123
1178 J. B. Heluany, V. Gkioulos

3 Related work of digital twins in the energy IEC 61850 and IEEE C37.118 because of their flexibility.
market Moreover, when it comes to the energy sector, some bench-
marks are useful for developing study cases on a common
While the previous section served as the conceptual basis research ground. By applying benchmark IEEE 118-Bus Sys-
on digital twins, this section objective is to analyze previous tem, [48] explored the adaptability of conventional models
works of energy-related digital twins reviews and highlight for developing a power system digital twin (PSDT), what
the differences between them and our work. includes enabling technologies that were discussed by [49]
As a starting discussion, it is worth noticing that many such as IoT platforms, and advanced data analytics including
studies refer to power systems without detailing the domain. artificial intelligence and machine learning.
As an example of general application in smart grids, [4] Such studies reinforce that the perceived benefits are com-
mention use cases in distribution utilities, distributed energy mon sense among scholars. However, given that most of them
management systems, operation centers, fault diagnosis, bat- are not domain-specific, there is a gap of knowledge on what
teries systems, and renewable energy generators. Besides can be the final actors of each domain and most common
these use cases are specific to the energy market, there is use cases, and that is the significance of our work. We aim
no quantitative analysis to provide the most relevant ones. at contributing with the energy sector community by doing
Following a similar broad view of smart grids, in [44] the a quantitative analysis of use cases in generation and dis-
authors map digital twin applications in power industry into tribution domains, specifying the final actors of these two
four groups: grid, plant, equipment, and other levels. Some domains, such as detailing services provided by digital twins
of the mentioned use cases include power grid structure and which for these actors.
design, control centers emulation, network security, trans-
formers, turbines, converters, relays, and fault diagnosis,
which are alike to the examples given in [4]. Research in 4 SLR method
[45] also investigated potential uses of the technology, the
most common being anomaly detection, smart grid man- This literature review differs from the one performed for the
agement, dynamic monitoring, and demand forecast. These related work section due to its rigorous and systematic steps.
studies have revealed the significance of the digital twin con- It is based on the methodologies described by [50] and [51]
cept across all aspects of smart grids, suggesting a number consisting of seven steps: (1) Identify the purpose, (2) Draft
of useful applications. However, they did not provide quan- Searching Protocol, (3) Apply practical screen, (4) Apply
titative analysis to identify the most relevant use cases per quality screen, (5) Extract Data, (6) Synthesize studies, and
domain, which is one of the reasons we narrowed our review (7) Write the review.
to two domains, being able to contribute specifically with Besides the databases, other tools utilized during this
generation and distribution. research were EndNote 20 for references management (prac-
For studies focused on specific domains, it has not been tical and quality screening), NVIVO and Excel to extract data
demonstrated the most relevant use case either. In their dis- and synthesize information.
cussion of Local Energy Communities (LEC) orchestration
[43], the authors goal is to develop digital twins that support 4.1 Purpose of the literature review
the balancing reserves of the electricity grid (distribution
domain). They contributed with modeling and simulation This literature review sought to identify the current state of
methods providing relevant references for physics-based knowledge of digital twins in the energy sector. In order to
models, and data-driven models. Their framework (Fig. 5) find information for technical aspects of the purpose, the
highlights prediction, optimization, and control strategies as following research questions were elaborated and grouped
services provided by the digital twin. according to specific categories. Group 1 (NIST domains
Some scholars have conducted research on enabling stan- and use cases) aims to provide information for quantitative
dards and technologies for information models suitable for analysis of use cases. Groups 2 and 3 (reference architec-
digital twins. Taking into account the wide range of existing tures, frameworks, and tools) allow a comparison between
solutions, [46] put forward the idea of utilizing the common the conceptual aspects studied in Sect. 2 to evaluate if the
information model (CIM), a vendor neutral open standard architectures and frameworks applied in the energy sector
for power systems. They argue that CIM is already used differ from general market. Group 4 (network and binding)
for developing interoperable applications (hardware or soft- focuses on the communication between the real and virtual
ware), and, given that interoperability is also a requirement entities to assess if the studies detail the bidirectional com-
for digital twins, it may provide a framework that enables munication, while group 5 is focused on security, which was
information sharing. Also approaching standards, in [47] the expected to be present in energy-related studies, given that
authors propose the use of communication protocols such as it is a critical infrastructure sector. Lastly, Group 6 has the

123
A review on digital twins for power generation and distribution 1179

Fig. 5 Digital twin framework for local energy communities digital twins extracted from [43]

Table 2 Compilation of some tools and techniques according to categories

Category Tool/technique

Bridging/integration Azure (Microsoft), AWS (Amazon), MindSphere (Siemens), Predix (GE), ThingWorx
(PTC), IBM Maximo Asset Health Insights (IBM), RFID, MTConnect, OPC UA,
MQTT, ZigBee, XML, IndraMotion MTX (Rexroth), Beacon (Fii-Foxconn),
TwinCAT (Beckhoff), SAP (SAP), Codesys (Codesys Group), edge/foggy computing
Data processing Data fusion algorithms, BigQuery (Google),
Spark/Storm/S4/Hive/Mahout/Flink/Pig/Impala (Apache), edge/foggy computing,
VoltDB (VoltDB), Azure (Microsoft), AWS (Amazon)
Data storage MongoDB (MongoDB), MySQL(Oracle/Others), Hadoop/Hbase/Kafka (Apache),
Oracle, Azure (Microsoft), AWS (Amazon), BigQuery (Google)
Data analytics AI algorithms (e.g., feature selection, feature extraction, pattern recognition, stochastic
optimization, evolutionary, etc.), ML algorithms (neural networks, fuzzy logic, etc.),
TensorFlow (Google), Azure (Microsoft), AWS (Amazon), BigQuery (Google)
Modeling Meta-information and semantics, ontologies, AutomationML, finite element, finite
element alternating method, AnyBody Modeling System (AnyBody Technology),
service-oriented architecture (SOA), representational state transfer (REST), Matlab
(MathWorks), Matpower (Matpower), InterPSS (InterPSS), OOPS, PowerFactory
(DIgSILENT), Modelica (Modelica), Markov chain, ANSYS Twin Builder (ANSYS),
NX (Siemens), SolidWorks (Dessault Systèmes), AutoCAD (Autodesk), 3D Max
(Autodesk), FreeCAD (Freecadweb), Azure (Microsoft), AWS (Amazon)
Simulation FEM simulation, Montecarlo simulation, CFD simulation, DDSIM (Damage and
Durability Simulator), S2S DFS, Simulink (MathWorks), CAE-based simulation,
(CATIA) Dassault Systemes, CIROS Studio (VEROSIM), Simcenter 3D (Siemens),
ANSYS Twin Builder (ANSYS), PSS R NETOMAC (Siemens), MWorks (Tongyuan),
SUMO (Eclipse), Open Simulation Platform (DNV-GL), Azure (Microsoft), AWS
(Amazon)

123
1180 J. B. Heluany, V. Gkioulos

objective to identify current challenges and directions for Xplore, ACM Digital Library, ResearchGate, ScienceDirect,
future work. Scopus, ProQuest, and Semantic Scholar.
Group 1: NIST domains and use cases The databases search was conducted with the following
query: ((“Document Title” OR “Abstract”: “digital twin*”)
– RQ1: What is the domain and actor addressed? AND (“Document Title” OR “Abstract”: “energy” OR “elec-
– RQ2: What is the reported use case? tric*” OR “power” OR “smart grid*”)). This set of keywords
– RQ3: Is the study practical or theoretical? intends to perform the initial filter of papers relating digital
– RQ4: What is the method applied? twin technology with the energy sector, considering differ-
– RQ5: Who are the stakeholders involved? ent vocabulary that were retrieved from a previous broad
analysis. The collection retrieved 3675 papers including
Group 2: Reference Architecture duplicates. After eliminating duplicates, 2238 were consid-
ered for the practical screening.
– RQ1: What is the architecture adopted?
– RQ2: What are the operational requirements of each layer 4.3 Practical screening
of the framework?
The practical screening phase consists of identifying articles
Group 3: Framework and its tools and techniques that may contain relevant research to answer the questions
stated in the literature review purpose. The inclusion and
– RQ1: What are the modeling techniques and tools used? exclusion criteria adopted in this work are given in Table 3.
– RQ2: What are the simulation techniques and tools used? The papers assessment was conducted by the main author
– RQ3: What are the data-related (processing, storage and (70%) and a laboratory assistant* at IIK, NTNU (30%). In
analysis) tools used? order to make the selection uniform, after each one had con-
cluded the selection, both assessed 10% of the selected by
Group 4: Network and binding each other and 5% of excluded. After aligning differences
and applying the inclusion/exclusion criteria explained here,
– RQ1: How is the data flow between the physical and the initial 2238 papers were reduced to 275.
virtual entities (manual, unidirectional or bidirectional?
– RQ2: How is the communication implemented in terms 4.3.1 Inclusion criteria
of architecture, components and services?
1. Language: papers written in English.
Group 5: Security 2. Content: the content analyzed through title/abstract must
be related to the research questions.
– RQ1: How is the security of the digital twin itself 3. Format: scientific/scholarly articles published in confer-
addressed? ences, workshops and journals.
– RQ2: How is the security of the communication between 4. Date: no restrictions were made regarding the year of pub-
the digital twin entity and the physical entity addressed? lication because digital twin is an emerging technology.
– RQ3: How is the security of the communication within 5. Market vertical: papers must be related to digital twins
digital twins addressed? in the energy sector. It is worth commenting though, that
– RQ4: Does the study address any other layer of security? when it comes to consumers, papers that approach digital
Which one? twins focusing on energy consumption efficiency in other
fields, were not considered, otherwise the results would
Group 6: Challenges and Future Work be very broad and out of the scope of this study;
6. Smart grid domain: papers should be about generation
and distribution domains from NIST conceptual model of
– RQ1: What are the limitations of this study?
smart grids
– RQ2: What are the challenges?
– RQ3: What is recommended as future work?
4.3.2 Exclusion criteria
4.2 Searching protocol
1. Format: Letters, editorials, reports, posters, magazines,
The protocol was designed to allow the searching evalua- trade journals, presentations, front and back matter pages
tion and replication by defining the databases and keywords such as title pages, abstract pages, index pages, table of
expression. The sources used to identify articles were: IEEE contents, and stand-alone images.

123
A review on digital twins for power generation and distribution 1181

Table 3 Inclusion and exclusion

Inclusion criteria Type
criteria
1. Written in English Language
2. Title eligible Content
3. Scientific articles published in conferences, workshops and journals Format
4. Publication year Date
5. Energy sector M. Vertical
6. Generation or distribution Domain

Exclusion criteria Type

1. Letters, editorials, reports, posters, magazines, trade journals, and presentations Format
2. Secondary studies Content
3. Articles that do not approach energy sector M. Vertical
4. Articles of the energy sector, but outside the generation and distribution domains Domain
5. Articles not found for full text download Availability

2. Content: secondary studies in the form of reviews, sur-

veys, systematic literature reviews, and generic digital
twin frameworks.
3. Market vertical: articles outside the energy sector that do
not meet inclusion criteria five explanations.
4. Smart grid domain: papers about transmission, opera-
tions, service providers, markets, and customer domains
from NIST conceptual model of smart grids are out of the
scope.

Fig. 6 Summary of papers filtering applied in this SLR

4.4 Quality screening

4.5 Data extraction
The 275 studies selected during the practical screening were
rated according to a quality assessment defined in a standard
This step consists of systematically extracting data from each
form used for each paper. Again, the papers assessment was
paper selected during the quality screening. The extraction
conducted by the main author (70%) and a laboratory assis-
form was prepared to store data that contributes to answering
tant* at NTNU, IIK (30%). The papers that did not meet the
the research questions in a consistent and uniform way, and
criteria were excluded, remaining 81 for the following steps.
was performed by the coding tool available at NVIVO.

4.4.1 Exclusion criteria 4.6 Synthesis of studies

1. Research Questions: papers that do not answer at least two Given the qualitative nature of most papers on digital twins,
questions from groups 2, or group 3, or group 5, and neces- the procedure adopted to analyze them and obtain a concept-
sarily one from group 4. The intention of this combination centric focus was based on synthesis by interpretation and
is to cover at least some aspects related to architec- explanation.
ture, framework, security, and, necessarily approach the
network, once the bidirectional communication was iden-
tified as a key possible differential between a digital twin 4.7 Writing the review
and traditional solutions.
2. Component level: papers in which the use case is at com- The findings of this literature review are described follow-
ponent level are too specific for the aim of this study. ing standard writing principles aiming to make a theoretical
3. Physical Model: papers that focus on the physical mod- contribution by being critical and using evidence to support
eling or structural properties instead of network/data. the claims.

123
1182 J. B. Heluany, V. Gkioulos

5 SLR analysis – Generation actors: defined power plant types includ-

ing wind, solar, nuclear, hydro, and thermal, added to
The following subsections address the topics as organized in a generic category labeled as power plant;
the research questions. – Distribution actors: substation, microgrid, distributed
energy resources (DERs), distribution centers;

5.1 NIST domains and use cases For the use cases categorization, the following assump-
tions were made:
As explained in the Background section 1.1, the terminology
of this paper is based on NIST Smart Grid Interoperability – Distribution management: papers related to multiple
Framework [2]. energy sources to be integrated into the grid consider-
This work started addressing previous reviews regardless ing the uncertainty factor.
of the market vertical, to map the main scientific findings on – Fault diagnosis: papers that approach monitoring, health
digital twins, and then narrowing down to an overview of the state, operational reliability, anomalies detection, percep-
energy sector reviews that approached all NIST smart grid tion, among others.
domains. This step showed a gap in domain-based studies. – Generation forecast: it is applicable for any energy
Attempting to cover this gap, we focused this paper on two source, but the trend was more noticed for renewable
domains: generation and distribution. sources, being wind and solar, the most common targets
After completing the practical screening and start classi- of forecast due to its intermittent power generation nature.
fying the papers, we noticed that only the NIST domains – Performance optimization: besides the term performance
classification would not be enough because many studies itself, other approaches of state estimation related to per-
approach more than one domain at the same time; thus, formance were also included in this group.
some assumptions were made to enable the categorization – Predictive maintenance: in addition to explicit prediction-
and provide an overview on what combination of domains related studies, papers approaching variables estimation,
was more present before the quality screening. We catego- fault prediction, and continuous monitoring were also
rized the papers into four groups: generation, distribution, considered in this category.
multiple domains, and all domains. Nonetheless, for cases – Reliability forecast: this category approaches both dis-
where more than one domain is approached, for example, tribution management and generation forecast, however
transmission and distribution, the amount of papers were not the focus is on system resilience.
enough to enable analyzing a trend for this combination. For – Security enforcement: papers that approach multiple
this reason, the most relevant domain was taken into account security use cases, such as online network monitor-
for the categorization, with the following considerations: ing, security testing, zero trust architectures, and secure
embedded solutions.
– Staff management: related to human aspects, behavioral,
– Smart grid papers that are generic were considered emergence awareness, compliance with regulations, and
addressing all NIST domains and labeled in the category decision making.
“All”;
– Microgrid papers or synonyms as “Local Energy Com- Although one of the practical screening exclusion crite-
munity” usually embrace generation, distribution, and ria was to be from the generation/distribution domain, as
sometimes transmission. In this work, we labeled them explained previously, some papers are not specific and a
as “Distribution” because this is the main domain single attribution is not possible. When papers approaching
approached, similar to power flow papers; multiple or all domains were attending the criteria (answer
– Papers on electrical power systems or equipment that can at least two questions form groups 2, or 3, or 5, and neces-
be utilized in more than one domain, such as transform- sarily one from group 4), they were considered relevant for
ers, relays, batteries, high voltage cables, coolers, and so this research and followed for further analysis.
on, were labeled as “Multiple”; In Fig. 7, graph (a) shows the amount of papers fitting into
– Power plant papers, regardless of the energy source, were each established category before the practical screening and
labeled as “Generation”; graph (b) shows the same, but after the quality screening,
where the previously explained quality criteria had to be full
filled, resulting in a higher quantity of papers on the distri-
Following the same approach, considerations were made bution domain.
to the actors found for the generation and distribution The digital twin use cases categorization enabled a quanti-
domains: tative analysis similar to the one performed at [52], contribut-

123
A review on digital twins for power generation and distribution 1183

implementation details are not always explained, limiting the

understanding of the tools and techniques that may leverage
the application of digital twins.
Regarding stakeholders, only 8 (9,8%) of the studies
approached them and their roles. Some just mention them
briefly [53, 54] as sources for data gathering and sharing,
while when considering larger-scale digital twins, as can be
the case for the distribution domain involving several actors,
we share the opinion of [20] and [55], who support includ-
ing both stakeholders and end-user during the designing and
Fig. 7 Statistics on NIST smart domains after the practical screening implementation of digital twins.
and after the quality screening
Corroborating this rationale, the papers that broadly
approached stakeholders were mainly about the distribution
domain, emphasizing how important it is to enable analy-
sis based on different perspectives and priorities that each
involved stakeholder may have [56]. Similarly, [57] argue
that the digital twin environment must enable these test sce-
narios from different viewpoints, such as induration and
carbon emission. The study [58] was performed by an electri-
cal distribution company in Brazil aiming to develop a power
system digital twin prototype to enable services of anomaly
detection, predictive maintenance, and automated network-
Fig. 8 Use cases statics for the distribution domain
ing modeling. As part of the solution, they built a third-party
platform to share information with stakeholders. An example
of stakeholders was discussed in [59], a use case of beyond
fifth generation (B5G) enabled smart grid, for which stake-
holders range from microgrid owners, resident operators,
utility administrators, prosumers, governments, utilities, and
regulators.
Regarding the generation domain, where most studies
focus on the component level and do not involve many actors,
from an internal perspective stakeholders may be from differ-
ent departments of the same entity [60], such as engineering,
Fig. 9 Use cases statics for the generation domain maintenance, management, system architects, and process
engineers.

ing to identify the most relevant use case for two scenarios. 5.2 Reference architecture
The first scenario in Fig. 8 groups the categories of “distri-
bution,” “multiple,” and “all” NIST domains, resulting in The findings of reference digital twin architectures from the
“distribution management” as the most relevant use case, fol- energy sector review and the generic architectures discussed
lowed by “fault diagnosis.” But in Fig. 9, only use cases from in the related work Sect. 2.3 are very similar, relying on the
“generation” domain papers were categorized, changing the common ground of the three basic dimensions: virtual entity,
most relevant use case to “fault diagnosis” followed by “per- physical entity, and the link between them. There may be
formance optimization” and “predictive maintenance.” discrepancies in the designations given to the dimensions,
with some seeing data and services as individual dimensions,
5.1.1 Studies design overview and stakeholders while others view them as subsections of the virtual entity.
For [61], we can express it as a five-dimension equation with
We observed that most selected studies present similar meth- the physical entity, the virtual entity, services, data, and the
ods for developing their research work, following a path that connection. An extra dimension is also considered by [62],
establishes the background on previous works, conceptual- named as the display layer.
izing a use case relevant for the paper, and performing a To avoid repetitiveness with the previous section on this
practical study case as a proof of concept. Although the qual- topic, only the differences will be discussed, and the most rel-
ity criteria were developed to collect more technical studies, evant one is that some studies associate the term digital twin

123
1184 J. B. Heluany, V. Gkioulos

Fig. 10 Digital twin architecture that differentiates from the definition by considering the “digital twin” block mainly as modeling and simulation,
including AI algorithms [34]

only with the layer of modeling and simulation, not with – “Field Level” corresponds to Purdue Level 0, showing
the physical entity. Interestingly, NASA’s definition does not the sensors and actuators at production process.
clearly mention it as well, pointing to an additional find- – “Control Level” corresponds to Purdue Levels 1 and 2,
ing of our work about the boundaries of the digital twin, showing the Programmable Logic Controllers for each
and whether the physical entity should or not be one of the system, the switches, computers, gateways, among oth-
architecture blocks. In Figs. 10 and 11, it is possible to see ers.
examples of architectures where the block “digital twin” is – “Enterprise Level” for Purdue Levels 3 and 4, where the
associated with modeling and simulation, including AI algo- computers are labeled according to their role in the busi-
rithms. ness planning, logistics, and manufacturing management.
Many studies depict the architecture and start defining the
functional blocks within the dimensions still considering as We believe that the data flow model of the Purdue Refer-
part of the architecture description. In this paper, however, ence Model, which has been available since the 1990s, is still
the tools and building blocks detailing is considered as part relevant and enough to detail the main blocks and their cor-
of the framework, which, as defined in a previous section, responding components and roles in each level. It is possible
is composed of reusable components for application-specific to map this model also to modern IIoT (Industrial Internet
features that enable the implementation of an architecture. of Things) environments, even considering technical features
The reason for this decision (building blocks within the such as time constraints, latency and availability, as explained
framework, not architecture) is based on how the ISA-95 by AWS [66]. They argue that the use of external resources
standard and its corresponding Purdue Reference Model should be placed on Purdue Level 2 systems or higher, not for
are translated to industrial automation architectures. This Level 0–1 systems. This is possible as the response time in
standard defines a conceptual model for industrial control Level 2 is within minutes, compared to seconds in Level 1 and
systems (ICS) with five layers, as recalled in Fig. 12. The milli/micro seconds in Level 0. Summarizing, we believe that
levels range from field equipment (sensors and signals) to an architecture as exemplified in Fig. 13 is in accordance with
business related level (Enterprise Resource Management). the definition considered in this paper, giving room to detail
Based on this standard, most industrial automation archi- the building blocks, software, and services in the framework
tectures present these same five layers, showing the main description.
components of each level without detailing the hardware,
software, and services, but showing how the components are
connected and labeling their role. As an example, a refer- 5.3 Framework and its tools and techniques
ence automation architecture was extracted from the PCS7
Siemens catalog [65], where we can see in Fig. 13, the corre- The architecture is a hierarchical system that relies on the
spondence between the levels of the Purdue Reference Model coordination of requirements between its layers. As stated
and this reference automation architecture from PCS7: by [67], a digital twin depends on requirements that range
from computational and communication platforms, both with

123
A review on digital twins for power generation and distribution 1185

Fig. 11 Digital twin architectures that differentiate from the definition simulations are made outside of this block, in the “data learning center,”
in the following senses: (a) the block “digital twin grid” receives fore- thus not included in the digital twin [63]. Contrasting with (b), different
cast information and returns the states, being possible to infer that it is models, simulation and optimization are included in the “digital twin”
basically a static model/simulation tool for the “actual grid” and other block [64]

and Table 5 for the set labeled as Distribution, Multiple, and

All.
Attempting to establish a link between the physical enti-
ties for each evaluated domain and the engineering systems
hierarchy, it is possible to infer from Table 4 that, for the
generation domain, most entities are component or subsys-
tem/system level, usually the main generation unity. For
nuclear power the reactor, for solar the photovoltaic con-
Fig. 12 ANSI/ISA-95 hierarchy pyramid of Purdue Reference Model version unit, for thermal the steam turbine, for wind power
the wind turbine, and for hydro the generating unity com-
posed of turbine and generator. These entities are aligned
hardware and software components. We can then see the dig- with the most relevant use case identified as fault diagnosis
ital twin as the set of all its components and their respective for equipment.
configurations within and across the layers, which here are For the distribution domain, the physical entities observed
assumed as part of the framework. in Table 5 hierarchy are system level, or, as would be called
Going back to the framework of Fig. 4, we will further in other literatures, system of systems, once the distribution
explore the four main functional blocks in this section, focus- includes all energy sources. Here, the ability to accurately
ing on similarities and differences found during the relevant estimate power demand and generation to balance the grid
analysis for the energy sector. is the main feature of the digital twin, also aligning with the
main use case, which is distribution management. Another
5.3.1 Physical entity interesting finding specific to the energy sector is the use of
benchmarks for case studies. Some of the identified bench-
This section’s goal is to provide information for the types marks are:
of physical entities from the energy sector that have been
addressed in the NIST domains selected for this study. Table 4 – IEEE 39-bus [95, 96].
summarizes the physical entities for the Generation domain – IEEE 9-bus [96].

123
1186 J. B. Heluany, V. Gkioulos

Fig. 13 Automation architecture example extracted from Siemens PCS7 catalog

Table 4 Physical entities modeled in typical actors from generation-related domain

Actor Physical entities

Hydro Pressure chamber [68]; generating unity [69]

Nuclear Reactor core [70]; functional modules of netronic calculations, temperature dynamics, process subsystems, steam generators,
pumps, secondary circuits, and auxiliary systems [71]; microreactor [72];
Solar Photovoltaic energy conversion unit (PVECU) composed of panel, power converter, and electrical sensors [73]; set of PV
inverters [63]; single-diode model of a single battery [61]; inverter system and weather data [74]; photovoltaic power station
with several panels [75]; photovoltaic solar farm [76]; PV power plant [77]
Thermal Boiler island, steam turbine island and emission control equipment [78];coal mill and feedwater pump and other key equipment
[79]; key equipment from main subsystems (mechanical, lubrication, water cooling, fuel, exhaust, air intake) [80]; gas turbine
system [81]; high-temperature thermal energy storage device [82]; steam turbine system [60]; gas turbine system [83];
combustion process, air and flue gas system [84]
Wind Wind turbine power converter [40]; WT with gearbox failure [85]; turbine’s physical system, and weather sensors [86];
direct-drive wind turbine [87]; onshore wind turbines [53]; wind turbine mechanical components [88]; generators, main
shafts, gearboxes, yaw systems, pitch systems, hydraulic systems, racks, and wind turbine control systems [89]
Power plant Large electrical generators [90]; battery energy storage [91]; auxiliary systems, thermodynamic properties, unit models (heat
exchanger, pump, compressor, boiler, etc.) [92]; boiler system [57]; four generators connected with the grid (Cigré
benchmark) [93]; CEP engine (complex event processing) [94]

– IEEE 33-bus [97]. is implicitly included in the service platform. An important

– Cigré benchmark [93]. aspect to highlight about the virtual entity is that modeling
and simulation, despite being different functionalities when
represented in a reference architecture/framework, are very
5.3.2 Virtual entity modeling often developed and implemented by the same tools, and
that is the reason why the new tools and techniques iden-
The intention of this section is to enrich the assessment sum- tified during the systematic literature review are grouped
marized in Table 2. In the reference framework of Fig. 4, there together:
is a functional block for modeling the virtual entity, but not
for simulation. However, we can assume that the simulation

123
A review on digital twins for power generation and distribution 1187

– Modeling and Simulation tools and techniques: Ter- analysis and assessment, system planning, model validation,
moflowTM [78], semiautomatic modeling from formal and disturbance analysis.
specification [73], OpenFAST (for wind turbine) [40],
DSL (Domain-Specific Language) [69], gradient boost-
– Energy sector common services: the services coincide
ing regressor and random forest [80], modeling based on
with what was called previously as use cases. Here, some
state machine [98], METHONTOLOGY approach [82],
are repeated to highlight energy-related services such as
convolution neural network [99], Scikit-learn toolbox
load balancing [56, 63, 64], demand forecasting [56, 59,
and EnergyPlus [56], Cigré benchmark [93], OpenDSS
118, 126], power flow analysis [55, 55, 63, 97, 101, 103,
[100], KNIME analytics platform [68], AnyLogic [101],
113, 116, 119, 129], three-phase short-circuit diagnosis
IEEE 39-bus benchmark [95];
[129], among others that may be more generic to different
market verticals, such as fault diagnosis, state estimation,
Additionally, the engineering hierarchy and the require-
operational planning, and so on.
ment for updates during all lifecycle phases interfere with
how the actual entity is modeled and why modeling and sim-
ulating are increasingly intertwined, which might be one of 5.3.5 AI and ML as enabling technologies
the essential differences between a digital twin and conven-
tional solutions. Modeling and simulation have been in the A search for terms related to AI and ML on all papers pro-
market for a lengthy period, but before, the simulations were vided results in 76 studies out of 81, which corresponds to
just static models over time. But the need of the market for 93,8%. 16 of these (19,7%) approach the topic directly as a
predictions to foresee scenarios and improve decision making key part of the study.
has created a space for dynamic models. This way, the previ- According to [122], a digital twin is a dynamic virtual
ous physical, mathematical, or logical models are becoming model and, for this reason, they propose a solution that
hybrid and increasingly include data-driven algorithms that is updated periodically by inputting new data to a neural
can be regularly updated. Existing component level models network model used for security assessment purposes. An
based on equations are still applied. However, when it comes important aspect of their work is that there is a bidirectional
to system level and its intrinsic high complexity to link all communication between the virtual and the real entities and
knowledge domains among its components, hybrid models one parameter of the solution is the time resolution. In this
gain strength. In this paper, it was noticed that the dynamic case, a sub-second delay. Google Tensor Flow was used for
models usually rely on AI and ML for their implementation. the offline training, what was also observed in other studies
[56, 59, 68, 80, 91, 96, 117].
5.3.3 Data management Similar to [122] system level approach, but with a different
objective, [56] proposed a management tool for multi-vector
The examination of the related work showed a greater energy systems which evaluates demands, supplies, and stor-
focus on data management than the systematic literature age from different sources, using machine learning to predict
reviews that typically center around single use cases. From the energy demand when there is a pattern, and artificial neu-
Table 2, where the tools were categorized, three groups were ral networks in cases where no patterns are found. This was
about data, comprising processing, storage, and analysis. The demonstrated by an example of a government building, show-
framework taken as a reference from Fig. 4 shows two sub- ing that it is feasible to optimize energy consumption by
functional blocks within the Data Management Platform, calculating the cost for different scenarios and selecting the
which are Data Model and Data Management Method. Other option that best suits the user-defined parameters.
categories found in a nonstructured way throughout the ana- Generic and more theoretical approaches, such as those
lyzed papers were related to data acquisition, pre-processing, proposed by [82] and [114] highlight the trend for utiliz-
storage, and processing. ing neural networks and machine learning for the simulation
functionality of digital twins. While [82] proposed a gray
5.3.4 Services box, defining as partially theoretical with data to complete
the model, and modeled in MATLAB, [114] designed a pure
Services make up another relevant block, also identified as theoretical modeling engine based on several algebraic equa-
one of the architecture’s layers and framework’s functional tions. Both proposed the comparison between the simulated
block. results and the real entity measurements, so that the system
Services can be performed online or offline, depending on is continuously updated to be as realistic as possible.
the real-time need for the outcome. Some examples of online When dealing with forecasting situations, many scholars
services are visualization, monitoring, alarming, state esti- study a broad range of AI/ML techniques applied to dif-
mation, protection, and control. Offline services may include ferent actors of the power generation domain. A solution

123
1188 J. B. Heluany, V. Gkioulos

Table 5 Physical entities modeled in typical actors from distribution-related domain

Actor Physical entities

DERs 5 DERs response power [102]; prosumers facility [103]; Battery Energy Storage Systems (BESS) [104]; DER and ESS
(Energy Storage Systems) resources [101]; PV inverters [100]
Electrical Power Distribution transformers [105]; 2 level inverters [106]; Gas-insulated Switchgear (GIS) [107]; converter station [99];
Systems communication network [108]
Energy Management Power line communication [109]; heating system electrification in part of the social housing stock and interventions in
System (EMS) the transport sector through increased EV charger installation [56]
Microgrid IEEE 39-bus benchmark [95]; generic generators from several sources [110]; energy units (solar, wind or conventional
generator) [111]; distribution controllers [112]
Power Flow Carbon neutrality-focused systems [113]; control room EMS (Energy Management System) [114]; general power
system components of single or multiple loads and their sensors [115]; multi-energy flow including electricity, heat,
gas, and hydrogen [116]; control center system [55]; IEEE 9-bus and IEEE 39-bus benchmark power systems [96]
Pricing Transformers, wind turbines, solar panels, CHP systems, TESs, boilers and Evs [117]
Smart Grid Smart meter devices, controller units, storage units, and analytics applications for network management and monitoring
[64]; distribution transformer, smart meters [118]; transmission line, transformer, controllers, circuit breakers [119];
medium voltage feeder, transformer, measuring devices [120]; energy providers [121]; dispatching control center
SCADA [122]; energy providers and stakeholders [59];generic power equipment [123]; DERs and DSOs [124];
synthetic NEM (S-NEM2300) composed of power stations, substations, wind farms, and transmission lines [54];
end-users platform of consumption monitoring [52]; control center system [67]
Substation Substation module, equipment supervision module, and personnel positioning module [125]; medium and low voltage
distribution network [126]; converter system [127]; regional multi-energy system integrating hydroelectric, wind
power, photovoltaic, combined cooling, heating and power, energy storage, and electric vehicle charging facilities
[128]; substation components [129]; MV/LV transformers, and grid cables [58]; primary equipment and sensors of the
substation [98]; transfer switches, PMUs, power transformers, energy meters and sensors [130]; IEDs, RTUs, control
room [131]; IEEE 33 node distribution power system with photovoltaic arrays, wind generators, and circuit breakers
[97]; terminal equipment of distribution network [62]; PV loads, transformers, cables [132]

with Microsoft Azure platform based on deep learning was from other traditional unidirectional solutions that are man-
developed by [53], using temporal convolution and k-nearest ually or automatically connected from the physical entity
approaches to predict the generated power given the wind to the virtual one. From [64] “a bidirectional communica-
speed estimation for a wind turbine. Following a similar con- tion platform for real-time continuous data management with
text, [61] developed a digital twin to support decision making high availability is essential,” but the authors gave no details
for power grid dispatching of wind power farms, using data explaining what exactly is shared, what are the operational
mining and artificial intelligence technology to achieve out- requirements, and how this channel affects the solution. In
put forecast of new energy field groups. In the case of hydro [107], the authors affirm that the process is bidirectional and
source, [68] predicted the pressure of the oscillating water that the real-time communication allows the synchronization
column using machine learning. between the virtual and real entities.
Other applications may include anomalies detection of According to computer science terminology, this intercon-
different natures. [76], for example, studied solar farms and nection that couples data sources and synchronizes them is
proposed a digital twin based approach to detect anoma- called binding, a term that is not common in the literature
lies of the physical system. The technique used for the time (used only in 4 out of 81 papers), but the term “synchroniza-
series simulation was deep learning, more specifically auto- tion” and stemmed words is present in many studies, 41 out
encoders. of 81. This data sharing for synchronization relies on the goal
of updating the digital twin throughout the lifecycle of the
5.4 Network and binding real system, allowing the services to be more reliable, thus
improving decision making.
The aim of the research questions focusing on network is Attempting to comprehend the necessary features of this
to better understand the bidirectional communication that is communication, some of the use cases were assessed to iden-
described in most definitions. Given that one of the quality tify the nature of the data exchange. For [56], the model states,
screening inclusion criteria was that the paper should nec- set-points and forecasts are updated. Similarly, in [78] the
essarily mention the communication, all studies analyzed operating performance of a thermal power plant is extracted
approach the networking to some degree. The need to state and shared with the virtual platform, and in [111], the live
the communication as bidirectional may be to differentiate states are updated when a change happens. An example

123
A review on digital twins for power generation and distribution 1189

refers to solar power plants transmitting irradiance, voltage, is mentioned, it is not possible to answer the four questions.
frequency, and settings back to the real system at every sim- Naturally, if the query for the papers collection focused on
ulation cycle [100]. security, many other studies would have appeared for evalua-
For the physical implementation of the communication tion, but the energy restriction on our query limits the results
channel, a wide range of technologies can be applied, as to provide a feasible reading work and restrict the market
shown previously in Table 2. The technologies explored here vertical. For this reason, the findings about security were
are complimentary and may guide to more energy-related grouped regardless of the research questions to provide an
implementations: overview of how it is being approached and what are the
gaps, being the first identified gap the lack of focus on secu-
– FPGA (Field Programmable Gate Array): in [73] they rity.
justify the use of an Artix-7 FPGA due to higher speed Using the words from [64], although “a digital twin can
and resources performance for the digital twin estimator both provide and undermine security based on its deploy-
of a PV cell. We found other applications in [106], where ment,” from the 81 papers under analysis, only three approach
an FPGA was utilized to emulate the physical hardware security directly [64, 95, 106], while others discuss briefly or
firmware, and to program the behavioral model of wind just mention. From an application perspective, when secu-
turbines in [87]. rity is the main topic, it is deployed as a security layer not
– Raspberry Pi: similarly to FPGA, some studies applied focusing on the security of the digital twin itself, as can be
Raspberry Pi to implement the RTUs (Remote Terminal seen in the following studies description.
Unity) between the physical and virtual entities [111, 112, Approaching smart grid security with digital twins, [64]
132]; argue that this sector lacks of security standardization and
– SCADA: Interestingly, 31 studies mention some sort of propose a digital twin structure with a framework consisting
connection with records from the Supervisory Control of a virtual entity, cyber-threat intelligence database for grid-
and Data Acquisition (SCADA) system. SCADA sys- specific attacks, simulation of attacks, and data analysis to
tems can provide multivariate time series data that allows perform vulnerability and risk assessment. All these layers
the modeling of the system, and continuous parameters would be continuously updated, including the latest attack
update [119]. According to [80], any industrial sector vectors, to provide better security evaluation.
can adapt SCADA supervisors to establish data exchange In a smaller scale, [95] focused on microgrids in a project
between the physical and virtual entities. entitled ANGEL (Automatic Network Guardian for ELectri-
– Industrial Fieldbus: getting data from the field usually cal systems) where the digital twin framework sophistication
requires compatibility features where communication is stated as level III, with physics-based simulation, phys-
protocols play an important role. Examples given by ical system, and adaptive GUI, but not machine learning,
[107] include Modbus, Industrial Ethernet, optical fiber which would characterize it as level IV. For their proof of
networks, among others for the collection, and MQTT, concept, the IEEE 39-bus benchmark was modeled in MAT-
HTTPS, RPC for transmission. Adding to the collection LAB/Simulink and tested with a three-phase fault in the
protocols, besides Modbus, [100] communicates with physical system that is alarmed in the digital twin when a
DERs using also DNP3 and proprietary protocols, [81] divergence is noticed. Based on this result, the authors defend
established a communication interface with Profibus to that the solution could also be applied for security cases such
integrate a gas turbine DCS with a PC based digital twin as false data injection, denial-of-service, topology attack, and
platform. In the context of the distribution domain, other vulnerabilities in general.
common protocols are IEC 61850 [54, 55, 95, 107, 114, A more specific application still focused on grid devices
130]. was developed by [106] with embedded security in the con-
trol and hardware layers. The aim was to emulate and verify
5.5 Security hardware patchings in the digital twin before applying them
to the real devices. An FPGA was utilized together with
From the research questions in the group entitled Security, it LabVIEW to allow the interface during the Hot-Patching
is possible to see that the initial goal was to understand this operation, adding a protection layer before the real imple-
topic in different layers of the digital twin, starting from the mentation.
security of the digital twin itself, then of the communication
intra and inter systems, and any other layer and/or applica-
tion that could be approaching security. During the papers 6 Discussion and contributions
analysis of the systematic literature review, the finding from
the first review discussed in Sect. 2.5 was reinforced, that This section will address the contributions provided by this
security is not a main topic in the use cases and, even when it work by discussing them according to the groups of the

123
1190 J. B. Heluany, V. Gkioulos

research questions (use cases, architectures, frameworks and discussion is that the papers evaluated lacked details of imple-
tools, network and binding, and security). By first approach- mentation, decreasing the understanding of the technical
ing the digital twin conceptualization regardless of the challenges for tools selection and utilization. Interestingly,
market, we could compare it with energy-related studies AI and ML are mentioned by 90% of the papers, which may
showing how it is progressing in this critical infrastructure be related to the dynamic nature of modeling in digital twins
sector. when compared to traditional static models.
Domain-specific use cases: one of the gaps identified in Network and Binding: the evaluation of the questions
the related studies and also from our review studies selec- related to the network and bidirectional communication,
tion was the lack of smart grid domain-specific use cases. resulted in showing that the term “synchronization” is applied
Even though the application domains are common sense, than “binding,” appearing in approximately 50% of the
few studies address it in a quantitative way. For this rea- studies. This was an unexpected finding given that the bidi-
son, we categorized the use cases for the generation and rectional communication is part of the digital twin definition
distribution domains for all the 81 papers and quantified the and yet not well explored. Even when mentioning synchro-
amount resulting in the most relevant ones. The criteria used nization it is not often discussed in terms of shared values,
to categorize the actors was explained in the “SLR Analy- latency issues, security issues, and so on. Thus, our con-
sis,” 5.1, being easily reproducible in case other researches tribution was limited, but it was possible to identify some
want to perform this statistics. By doing the statistical analy- variables specific to the energy sector such as states, set-
sis, we were able to indicate that the progress of digital twins points, weather forecasts, performance indicators, voltage,
in the energy sector regarding distribution and generation frequency, current, among others.
domains are, respectively, fault diagnosis, and distribution Security: no significant information was found about the
management. Moreover, we further investigated what were security of the digital twin itself or its application as a security
the physical entities of each actor, providing a comprehensive layer, showing a gap in this field for energy-related studies.
list in Tables 5 and 4. Even being aware of our research limitation due to the marked
Architectures, Frameworks, and Tools: from the com- filter that could exclude papers only about the cybersecurity
parison with the broad conceptualization, we identified a of digital twins, taking into consideration that smart grids are
difference between the most accepted definition and how critical infrastructures that might require security measures
the energy-related studies establish the digital twin bound- by law, it was expected to find more information on how it is
aries represented in architectures: while NASA’s definition being addressed.
includes the real entity, services, data management, and vir-
tual entity, most analyzed studies label as digital twin only
the virtual entity, which is translated most of the times as the 7 Challenges and future work
modeling and simulation block. Therefore, it shows the need
for consensus regarding the digital twin boundaries when Before stating challenges and future work, it is important
representing it in an architecture or a framework. to highlight some of the limitations that may influence how
We noticed that the issue presented in the related studies the research questions were answered. For our systematic lit-
about the interchangeable use of the terms “architecture” and erature review, we considered only studies from the energy
“framework” repeated itself in our data set. Thus, we con- sector that followed a combination of the research questions.
tributed with a new perspective based on the well-established This decision may have limited or even excluded papers that
Purdue model applied to reference automation architectures detail digital twins in regard to the technical topics covered
showing how it relates major suppliers, e.g., Siemens, high- (architecture, framework, stakeholders, network, and secu-
lighting how a framework is expected to differ from an rity). During the analysis, we noticed that many layers of the
architecture by detailing tools and reusable blocks. digital twin still lack more consistency in definitions, tools,
Turning now to the tools, it was performed a compre- and applications to allow more real-world applications. This
hensive compilation divided into categories: bridging, data starts with the lack of consensus for the term definition itself,
processing, data storage, data analytics, modeling, and simu- which may lead to a misuse of terminology and impacting
lation. Naturally there are overlaps of tools that are applicable how other layers are defined and implemented. Moving to the
to more than one category, but the idea is to provide an implementation, the gaps found about the lack of focus on
overview and serve as consulting material for the choice detailed architecture and framework, security, stakeholders,
of market solutions. It is also worth noticing that the set of and network integration would be a natural progression of this
tools might change according to the use or the researchers work. The next step of this research will focus on proposing
approach. For example, a use case of wind turbines can an architecture and framework that takes into consideration
make use of OpenFAST, or utilize MATLAB or Labview for the specific needs of the energy sector.
the modeling and simulation. One of the limitations of this

123
A review on digital twins for power generation and distribution 1191

8 Conclusion Acknowledgements *We would like to express our gratitude to Lama

Amro, the laboratory assistant at IIK, NTNU, who assisted on the steps
of practical and quality screening.
On the question of the digital twin definition, this study rein-
forced that there is a misconception. We discussed some Author Contributions JBH and VG were involved in writing. VG was
reasons like the utilization of broad terms such as “ultra- involved in supervision. All authors have read and agreed to the pub-
realistic,” “mirror,” and “system” and proposed using the lished version of the manuscript.
concept of systems engineering design hierarchy to better Funding Open access funding provided by NTNU Norwegian Univer-
translate what is the real entity and what knowledge domains sity of Science and Technology (incl St. Olavs Hospital - Trondheim
will be addressed in the virtual entity. University Hospital). This work has received funding from the Research
When evaluating use cases according to smart grid Council of Norway through the SFI Norwegian Centre for Cybersecu-
rity in Critical Sectors (NORCICS) project no. 310105.
domains as defined by NIST, the collaboration comes from
finding the most relevant use case for the two domains chosen Open Access This article is licensed under a Creative Commons
as scope, being distribution management for the distribu- Attribution 4.0 International License, which permits use, sharing, adap-
tion domain studies, and fault diagnosis for the generation tation, distribution and reproduction in any medium or format, as
long as you give appropriate credit to the original author(s) and the
domain. Still, for the generation domain, another finding that source, provide a link to the Creative Commons licence, and indi-
stands out from the results is that few studies address hydro cate if changes were made. The images or other third party material
power plants. Given the energy transition and considering in this article are included in the article’s Creative Commons licence,
that this energy source is renewable and composes a great unless indicated otherwise in a credit line to the material. If material
is not included in the article’s Creative Commons licence and your
percentage of power generation, it was expected to be the intended use is not permitted by statutory regulation or exceeds the
main focus in more papers. According to the World Eco- permitted use, you will need to obtain permission directly from the copy-
nomic Forum [133], countries such as China, Brazil, Canada, right holder. To view a copy of this licence, visit http://creativecomm
USA, Russia, and Norway are among the ones who produce ons.org/licenses/by/4.0/.
most hydroelectric power.
Taking into consideration the recent cyber-attack cases in
critical infrastructures, one unanticipated result was that no References
focus is given to the security of the digital twin itself. Few
1. CISA, Energy sector. https://www.cisa.gov/topics/critical-
papers address security, and when they do, the digital twin
infrastructure-security-and-resilience/critical-infrastructure-
role is an additional security layer, without further discus- sectors/energy-sector
sions on the vulnerabilities that it may add to the system. 2. Gopstein, A., Nguyen, C., O’Fallon, C., Hastings, N., Wollman,
There are additional aspects that could possibly be advan- D.A.: Nist framework and roadmap for smart grid interoperability
standards, release 4.0 (2021-02-18 00:02:00 2021). https://doi.
tageous in discovering any deficiencies for digital twin
org/10.6028/NIST.SP.1108r4
execution, pertaining to both technical and human elements. 3. Ketter, W., Collins, J., Saar-Tsechansky, M., Marom, O.: Infor-
The bidirectional communication was stated to be present in mation systems for a smart electricity grid: emerging challenges
all selected papers, but requires deeper exploration of what and opportunities. ACM Trans. Manage. Inf. Syst. (TMIS) 9(3),
1–22 (2018)
is shared and how the outcomes of the digital twin go back
4. Nguyen, S., Abdelhakim, M., Kerestes, R.: Survey paper of digital
to the real system. From the studies, the data exchange from twins and their integration into electric power systems, In: 2021
the real entity to the virtual one is clear, and usually used IEEE Power and Energy Society General Meeting (PESGM), pp.
to create and update the models that are further used for the 01–05. https://doi.org/10.1109/PESGM46819.2021.9638011
5. Ardeshiri, A., Lotfi, A., Behkam, R., Moradzadeh, A., Barzkar,
simulations. In turn, the simulations result provide informa-
A.: Introduction and literature review of power system challenges
tion that guide better decision making, regardless of the use and issues, Application of machine learning and deep learning
case. However, the human–machine interface that displays methods to power system problems, 19–43 (2021)
the results is not necessarily the same as in the real system, so 6. SmartGrid.gov, Ethe smart grid. https://www.smartgrid.gov/the_
smart_grid/smart_grid.html
the communication could be unidirectional only. None of the
7. GT17, Technology trends 2017 (2017). https://www.gartner.com/
studies stated an autonomous feedback that impacted the sys- smarterwithgartner/gartners-top-10-technology-trends-2017
tem automatically. Moreover, given that the use cases allow 8. GT18, Technology trends2018 (2018) https://www.gartner.com/
better decision making, it was expected that also the stake- smarterwithgartner/gartner-top-10-strategic-technology-trends-
for-2018
holders would be more present in the discussions, but who
9. GT19, Technology trends 2019 (2019) https://www.gartner.com/
are the stakeholders, what are their roles, and how each of smarterwithgartner/gartner-top-10-strategic-technology-trends-
them have access to the information is commonly discussed. for-2019
The present results show interesting findings not only for 10. GT20, Gartner20 (2020) https://www.gartner.com/
smarterwithgartner/gartner-top-10-strategic-technology-trends-
the energy sector, where use cases and gaps were identified,
for-2020
but also for gaps related to the architecture, framework, net- 11. Uhlenkamp, J.F., Hauge, J.B., Broda, E., Lütjen, M., Freitag, M.,
work, and human aspects involving decision making. Thoben, K.D.: Digital twins: a maturity model for their classi-

123
1192 J. B. Heluany, V. Gkioulos

fication and evaluation. IEEE Access 10, 69605–69635 (2022). Engineering, Technology and Innovation (ICE/ITMC), pp. 1–7
https://doi.org/10.1109/ACCESS.2022.3186353 (2021). https://doi.org/10.1109/ICE/ITMC52061.2021.9570114
12. Vasiljevska, J., Gangale, F., Covrig, L., Mengolini, A.: Smart 28. Glaessgen, E., Stargel, D.: The digital twin paradigm
grids and beyond ? An EU research and innovation perspective, for future nasa and us air force vehicles, In 53rd
Scientific analysis or review KJ-NA-30786-EN-N (online), Lux- AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics
embourg (Luxembourg) (2021). https://doi.org/10.2760/705655 and Materials Conference 20th AIAA/ASME/AHS Adaptive
(online) Structures Conference 14th AIAA, p. 1818 (2012)
13. Grieves, M., Vickers, J.: Digital twin: mitigating unpredictable, 29. Kuehner, K.J., Scheer, R., Strassburger, S.: Digital twin: finding
undesirable emergent behavior in complex systems. In: Transdis- common ground—a meta-review. Proc. CIRP 104, 1227–1232
ciplinary perspectives on complex systems, pp. 85–113. Springer, (2021). https://doi.org/10.1016/j.procir.2021.11.206
Berlin (2017) 30. Kritzinger, W., Karner, M., Traar, G., Henjes, J., Sihn, W.: Digital
14. Negri, E., Fumagalli, L., Macchi, M.: A review of the roles of twin in manufacturing: a categorical literature review and classi-
digital twin in cps-based production systems. Proc. Manuf. 11, fication. IFAC-PapersOnLine 51(11), 1016–1022 (2018)
939–948 (2017). https://doi.org/10.1016/j.promfg.2017.07.198 31. Kossiakoff, A., Biemer, S.M., Seymour, S.J., Flanigan, D.A.: Sys-
15. Barricelli, B.R., Casiraghi, E., Fogli, D.: A survey on digital twin: tems Engineering Principles and Practice. John Wiley & Sons,
definitions, characteristics, applications, and design implications. New Jersey (2020)
IEEE Access 7, 167653–167671 (2019). https://doi.org/10.1109/ 32. Melesse, T.Y., Pasquale, V.D., Riemma, S.: Digital twin models in
ACCESS.2019.2953499 industrial operations: a systematic literature review. Proc. Manuf.
16. Semeraro, C., Lezoche, M., Panetto, H., Dassisti, M.: Digital 42, 267–272 (2020). https://doi.org/10.1016/j.promfg.2020.02.
twin paradigm: a systematic literature review. Comput. Ind. 130, 084
103469 (2021). https://doi.org/10.1016/j.compind.2021.103469 33. Zhang, H., Ma, L., Sun, J., Lin, H., Thürer, M.: Digital twin in
17. Perno, M., Hvam, L., Haug, A.: Implementation of digital twins services and industrial product service systems: review and analy-
in the process industry: a systematic literature review of enablers sis. Proc. CIRP 83, 57–60 (2019). https://doi.org/10.1016/j.procir.
and barriers. Comput. Ind. 134, 103558 (2022). https://doi.org/ 2019.02.131
10.1016/j.compind.2021.103558 34. Aghazadeh Ardebili, A., Longo, A., Ficarella, A.: Digital
18. Liu, M., Fang, S., Dong, H., Xu, C.: Review of digital twin about twins bonds society with cyber-physical energy systems: a lit-
concepts, technologies, and industrial applications. J. Manuf. erature review, In 2021 IEEE International Conferences on
Syst. 58, 346–361 (2021). https://doi.org/10.1016/j.jmsy.2020. Internet of Things (iThings) and IEEE Green Computing and
06.017 Communications (GreenCom) and IEEE Cyber, Physical and
19. Sjarov, M., Lechler, T., Fuchs, J., Brossog, M., Selmaier, A., Social Computing (CPSCom) and IEEE Smart Data (Smart-
Faltus, F., Donhauser, T., Franke, J.: The digital twin concept Data) and IEEE Congress on Cybermatics (Cybermatics), pp.
in industry - a review and systematization, In 2020 25th IEEE 284–289 (2021). https://doi.org/10.1109/iThings-GreenCom-
International Conference on Emerging Technologies and Factory CPSCom-SmartData-Cybermatics53846.2021.00054
Automation (ETFA), Vol. 1, pp. 1789–1796 (2020). https://doi. 35. van Dinter, R., Tekinerdogan, B., Catal, C.: Predictive main-
org/10.1109/ETFA46521.2020.9212089 tenance using digital twins: a systematic literature review. Inf.
20. Jones, D., Snider, C., Nassehi, A., Yon, J., Hicks, B.: Charac- Softw. Technol. 151, 107008 (2022). https://doi.org/10.1016/j.
terising the digital twin: a systematic literature review. CIRP J. infsof.2022.107008
Manuf. Sci. Technol. 29, 36–52 (2020). https://doi.org/10.1016/ 36. Core concepts (2023). https://pubs.opengroup.org/togaf-
j.cirpj.2020.02.002 standard/introduction/chap03.html
21. Ardebili, A.A., Longo, A., Ficarella, A.: Digital twin (dt) in smart 37. Cavaness, C., Keeton, B.: Jakarta Struts pocket reference,
energy systems-systematic literature review of dt as a growing “O’Reilly Media, Inc.” (2003)
solution for energy internet of the things (eiot), In E3S Web of 38. Josifovska, K., Yigitbas, E., Engels, G.: Reference framework for
Conferences, Vol. 312, EDP Sciences, p. 09002 (2021) digital twins within cyber-physical systems, In 2019 IEEE/ACM
22. Tao, F., Zhang, H., Liu, A., Nee, A.Y.C.: Digital twin in industry: 5th International Workshop on Software Engineering for Smart
state-of-the-art. IEEE Trans. Ind. Inf. 15(4), 2405–2415 (2019). Cyber-Physical Systems (SEsCPS), pp. 25–31 (2019). https://doi.
https://doi.org/10.1109/TII.2018.2873186 org/10.1109/SEsCPS.2019.00012
23. Fuller, A., Fan, Z., Day, C., Barlow, C.: Digital twin: enabling 39. Heluany, J.B., Galvão, R.: IEC 62443 standard for hydro power
technologies, challenges and open research. IEEE Access 8, plants. Energies 16(3), 1452 (2023)
108952–108971 (2020). https://doi.org/10.1109/ACCESS.2020. 40. Sivalingam, K., Sepulveda, M., Spring, M., Davies, P.: A review
2998358 and methodology development for remaining useful life pre-
24. Mylonas, G., Kalogeras, A., Kalogeras, G., Anagnostopoulos, C., diction of offshore fixed and floating wind turbine power con-
Alexakos, C., Muñoz, L.: Digital twins from smart manufacturing verter with digital twin technology perspective, In 2018 2nd
to smart cities: A survey. IEEE Access 9, 143222–143249 (2021). International Conference on Green Energy and Applications
https://doi.org/10.1109/ACCESS.2021.3120843 (ICGEA), pp. 197–204 (2018). https://doi.org/10.1109/ICGEA.
25. Balijepalli, V.M., Sielker, F., Karmakar, G.: Evolution of power 2018.8356292
system cim to digital twins-a comprehensive review and analysis. 41. Löcklin, A., Müller, M., Jung, T., Jazdi, N., White, D., Weyrich,
In 2021 IEEE PES Innovative Smart Grid Technologies Europe M.: Digital twin for verification and validation of industrial
(ISGT Europe), pp. 1–6 (2021) automation systems—a survey, In 2020 25th IEEE International
26. Rathore, M.M., Shah, S.A., Shukla, D., Bentafat, E., Bakiras, S.: Conference on Emerging Technologies and Factory Automation
The role of AI, machine learning, and big data in digital twin- (ETFA), Vol. 1, pp. 851–858 (2020). https://doi.org/10.1109/
ning: a systematic literature review, challenges, and opportunities. ETFA46521.2020.9212051
IEEE Access 9, 32030–32052 (2021). https://doi.org/10.1109/ 42. Perno, M., Hvam, L., Haug, A.: Enablers and barriers to the imple-
ACCESS.2021.3060863 mentation of digital twins in the process industry: A systematic
27. Pronost, G., Mayer, F., Marche, B., Camargo, M., Dupont, L.: literature review, In IEEE International Conference on Industrial
Towards a framework for the classification of digital twins and Engineering and Engineering Management (IEEM), pp. 959–964
their applications, In 2021 IEEE International Conference on (2020). https://doi.org/10.1109/IEEM45057.2020.9309745

123
A review on digital twins for power generation and distribution 1193

43. Nguyen-Huu, T.A., Tran, T.T., Tran, M.Q., Nguyen, P.H., cal distribution system-an application case. Energies 15(8), 2836
Slootweg, J.: Operation orchestration of local energy commu- (2022). https://doi.org/10.3390/en15082836
nities through digital twin: a review on suitable modeling and 59. Lopez, J., Rubio, J.E., Alcaraz, C.: Digital twins for intelligent
simulation approaches, In: 2022 IEEE 7th International Energy authorization in the b5g-enabled smart grid. IEEE Wirel. Com-
Conference (ENERGYCON), pp. 1–6. https://doi.org/10.1109/ mun. 28(2), 48–55 (2021). https://doi.org/10.1109/MWC.001.
ENERGYCON53164.2022.9830264 2000336
44. Fu, Y., Huang, Y., Hou, F., Li, K.: A brief review of digital twin in 60. Ghita, M., Benhadou, S., Hicham, M., Mounaam, A.: Ht-tpp: a
electric power industry, In 2022 IEEE 5th International Electrical hybrid twin architecture for thermal power plant collaborative
and Energy Conference (CIEEC), pp. 2314–2318. https://doi.org/ condition monitoring. Energies 15, 5383 (2022). https://doi.org/
10.1109/CIEEC54735.2022.9846081 10.3390/en15155383
45. Digital twin (dt) in smart energy systems - systematic literature 61. Shen, R., Wang, Y., Ma, M., Zhou, Q., Lyu, Q., Zhang, J.: Appli-
review of dt as a growing solution for energy internet of the things cation of digital twin technology in auxiliary decision-making
(eiot) (2021) https://doi.org/10.1051/e3sconf/202131209002 system for grid-connected dispatching of new energy. J. Phys.
46. Balijepalli, V. M., Sielker, F., Karmakar, G.: Evolution of power Conf. Ser. 2202(1), 012045 (2022). https://doi.org/10.1088/1742-
system cim to digital twins - a comprehensive review and 6596/2202/1/012045
analysis, In 2021 IEEE PES Innovative Smart Grid Technolo- 62. Zijian, Z., Jiaxin, L., Tie, G., Guanyu, W., Defu, W., Shiqing,
gies Europe (ISGT Europe), pp. 1–6. https://doi.org/10.1109/ W.: Fault prediction of distribution terminal equipment based
ISGTEurope52324.2021.9640174 on entropy weight vague matter-element under the digital twin
47. Brosinsky, C., Kummerow, A., Richter, M., Naumann, A., Wiest, framework, In 2021 6th International Conference on Robotics
P., Nicolai, S., Westermann, D.: The role of digital twins in power and Automation Engineering (ICRAE), pp. 195–199. https://doi.
system automation and control: necessity, requirements, and ben- org/10.1109/ICRAE53653.2021.9657779
efits. In ETG Congress, pp. 1–6 (2021) 63. Jung, Y., Han, C., Lee, D., Song, S., Jang, G.: Adaptive volt-var
48. Cui, Y., Xiao, F., Wang, W., He, X., Zhang, C., Zhang, Y.: Digital control in smart pv inverter for mitigating voltage unbalance at pcc
twin for power system steady-state modelling, simulation, and using multiagent deep reinforcement learning. Appl. Sci. 11(19),
analysis, In 2020 IEEE 4th Conference on Energy Internet and 8979 (2021). https://doi.org/10.3390/app11198979
Energy System Integration (EI2), pp. 1233–1238. https://doi.org/ 64. Atalay, M., Angin, P.: A digital twins approach to smart grid
10.1109/EI250167.2020.9346850 security testing and standardization. In 2020 IEEE International
49. Onile, A.E., Machlev, R., Petlenkov, E., Levron, Y., Belikov, J.: Workshop on Metrology for Industry 4.0 and IoT, pp. 435–440.
Uses of the digital twins concept for energy services, intelli- https://doi.org/10.1109/MetroInd4.0IoT48571.2020.9138264
gent recommendation systems, and demand side management: 65. Simatic pcs7 (2018) https://cache.industry.siemens.com/dl/files/
A review. Energy Rep. 7, 997–1015 (2021). https://doi.org/10. 406/109766406/att_981146/v1/br_simatic_pcs7_en_2017_Web.
1016/j.egyr.2021.01.090 pdf
50. Fink, A.: Conducting Research Literature Reviews: From the 66. Mapping to the ISA-95 model (2023) https://docs.aws.amazon.
Internet to Paper. Sage publications, Thousand Oaks (2014) com/whitepapers/latest/industrial-iot-architecture-patterns/
51. Okoli, C.: A guide to conducting a standalone systematic literature mapping-to-the-isa-95-model.html
review. Commun. Assoc. Inf. Syst. 37(1), 43 (2015) 67. Semenkov, K., Promyslov, V., Poletykin, A., Mengazetdinov, N.:
52. Martin Lopo, M.M., Boal, J., Sanchez Miralles, A.: Transitioning Validation of complex control systems with heterogeneous digital
from a meta-simulator to electrical applications: an architecture. models in industry 4.0 framework †. Machines (2021). https://doi.
Simul. Model. Pract. Theory 94, 177–198 (2019). https://doi.org/ org/10.3390/machines9030062
10.1016/j.simpat.2019.02.007 68. Seo, D., Huh, T., Kim, M., Oh, J.W., Cho, S.G., Jeong, S.C.: A pre-
53. Fahim, M., Sharma, V., Cao, T.V., Canberk, B., Duong, T.Q.: dictive model for oscillating water column wave energy converters
Machine learning-based digital twin for predictive modeling in based on machine learning. ICIC Express Lett. Part B Appl. 12(8),
wind turbines. IEEE Access 10, 14184–14194 (2022). https://doi. 733–740 (2021). https://doi.org/10.24507/icicelb.12.08.733
org/10.1109/ACCESS.2022.3147602 69. Zhao, Z., Li, D., She, J., Zhang, H., Zhou, Y., Zhao, L.: Construc-
54. Arrano-Vargas, F., Konstantinou, G.: Modular design and real- tion and application of digital twin model of hydropower plant
time simulators toward power system digital twins implementa- based on data-driven, In 2021 3rd International Workshop on Arti-
tion. IEEE Trans. Ind. Inf. (2022). https://doi.org/10.1109/TII. ficial Intelligence and Education (WAIE), pp. 60–64. https://doi.
2022.3178713 org/10.1109/WAIE54146.2021.00020
55. Marot, A., Kelly, A., Naglic, M., Barbesant, V., Cremer, J., 70. Jharko, E.: Digital twin of npps: Simulation systems and
Stefanov, A., Viebahn, J.: Perspectives on future power system verification. In 2021 International Russian Automation Con-
control centers for energy transition. J. Modern Power Syst. Clean ference (RusAutoCon), pp. 852–857. https://doi.org/10.1109/
Energy 10(2), 328–344 (2022). https://doi.org/10.35833/MPCE. RusAutoCon52004.2021.9537546
2021.000673 71. Jharko, E.: Digital twin of the technological process of the npp
56. O’Dwyer, E., Pan, I., Charlesworth, R., Butler, S., Shah, N.: power unit. In 2021 14th International Conference Management
Integration of an energy management tool and digital twin for of large-scale system development (MLSD), pp. 1–5. https://doi.
coordination and control of multi-vector smart energy systems. org/10.1109/MLSD52249.2021.9600212
Sustain. Cities Soc. 62, 102412 (2020). https://doi.org/10.1016/j. 72. Garcia, H.E., Aumeier, S.E., Al-Rashdan, A.Y., Rolston, B.L.:
scs.2020.102412 Secure embedded intelligence in nuclear systems: framework and
57. Amar, B., Subhrojyoti, R.C., Barnali, B., Dhakshinamoorthy, methods. Ann. Nucl. Energy 140, 107261 (2020). https://doi.org/
R., Seenivasan, A., Naveenkumar, S.: Knowledge driven rapid 10.1016/j.anucene.2019.107261
development of white box digital twins for industrial plant sys- 73. Jain, P., Poon, J., Singh, J.P., Spanos, C., Sanders, S.R., Panda,
tems, In IECON 2021 - 47th Annual Conference of the IEEE S.K.: A digital twin approach for fault diagnosis in distributed
Industrial Electronics Society, pp. 1–6. https://doi.org/10.1109/ photovoltaic systems. IEEE Trans. Power Electron. 35(1), 940–
IECON48115.2021.9589912 956 (2020). https://doi.org/10.1109/TPEL.2019.2911594
58. Fernandes, S.V., João, D.V., Cardoso, B.B., Martins, M.A.I., 74. Tuomiranta, A., Horvath, I., Schils, A., Brabandere, K., Voroshazi,
Carvalho, E.G.: Digital twin concept developing on an electri- E., Bertrand, E., Aldalali, B., Gordon, I., Wabbes, A., Scheerlinck,

123
1194 J. B. Heluany, V. Gkioulos

S.: Auto-Parametrizing the Digital Twin of Photovoltaic Power in wind turbines of wind power farm, In Advances in Transdisci-
Systems (2021). https://doi.org/10.4229/EUPVSEC20212021- plinary Engineering, Vol. 17, pp. 80–92. https://doi.org/10.3233/
5DO.1.4 ATDE210263
75. Liu, J., Lu, X., Zhou, Y., Cui, J., Wang, S., Zhao, Z.: Design of 90. Ebrahimi, A.: Challenges of developing a digital twin model of
photovoltaic power station intelligent operation and maintenance renewable energy generators. In 2019 IEEE 28th International
system based on digital twin. In 2021 6th International Conference Symposium on Industrial Electronics (ISIE), pp. 1059–1066.
on Robotics and Automation Engineering (ICRAE), pp. 206–211. https://doi.org/10.1109/ISIE.2019.8781529
https://doi.org/10.1109/ICRAE53653.2021.9657759 91. Tang, X., Sun, Y., Zhao, Y., Pei, W., Li, N., Kong, L.: Digital twin
76. Berlanga, R.: Digital twins in solar farms: an approach through based bess state estimation and operating opimization. In 2021
time series and deep learning. Algorithms 14(5), 156 (2021). IEEE 5th Conference on Energy Internet and Energy System Inte-
https://doi.org/10.3390/a14050156 gration (EI2), pp. 3402–3405. https://doi.org/10.1109/EI252483.
77. Livera, A., Paphitis, G., Pikolos, L., Papadopoulos, I., Montes- 2021.9713587
Romero, J., Lopez Lorente, J., Makrides, G., Sutterlueti, J., 92. Sleiti, A., Kapat, J., Vesely, L.: Digital twin in energy industry:
Georghiou, G.: Intelligent Cloud-Based Monitoring and Control proposed robust digital twin for power plant and other complex
Digital Twin for Photovoltaic Power Plants (2022) capital-intensive large engineering systems. Energy Rep. 8, 3704–
78. Xu, B., Wang, J., Wang, X., Liang, Z., Cui, L., Liu, X., Ku, A.Y.: 3726 (2022). https://doi.org/10.1016/j.egyr.2022.02.305
A case study of digital-twin-modelling analysis on power-plant- 93. Song, X., Cai, H., Jiang, T., Schlegel, S., Westermann, D.: Parame-
performance optimizations. Clean Energy 3(3), 227–234 (2019). ter tuning for dynamic digital twin of generation unit in power grid
https://doi.org/10.1093/ce/zkz025 01–06 https://doi.org/10.1109/ISGTEurope52324.2021.9640105
79. Yan, X.: Construction of digital twin ecosystem for coal-fired 94. Huang, J., Zhao, L., Wei, F., Cao, B.: The application of digital
generating units. J. Phys. Conf. Ser. (2021). https://doi.org/10. twin on power industry. IOP Conf. Ser. Earth Environ. Sci. (2021).
1088/1742-6596/1748/5/052037 https://doi.org/10.1088/1755-1315/647/1/012015
80. Deon, B., Cotta, K.P., Silva, R.F.V., Batista, C.B., Justino, G.T., 95. Danilczyk, W., Sun, Y., He, H.: Angel: An intelligent digital
Freitas, G.C., Cordeiro, A.M., Barbosa, A.S., Loução, F.L., twin framework for microgrid security. In 2019 North Ameri-
Simioni, T., Morais, A.M., Medeiros, I.E.A., Almeida, R.J.S., can Power Symposium (NAPS), pp. 1–6. https://doi.org/10.1109/
Araújo, C.A.A., Soares, C., Padoin, N.: Digital twin and machine NAPS46351.2019.9000371
learning for decision support in thermal power plant with com- 96. Danilczyk, W., Sun, Y. L., He, H.: Smart grid anomaly detection
bustion engines. Knowl. Based Syst. (2022). https://doi.org/10. using a deep learning digital twin. In 2020 52nd North Ameri-
1016/j.knosys.2022.109578 can Power Symposium (NAPS), pp. 1–6. https://doi.org/10.1109/
81. Panov, V., Cruz-Manzo, S.: Gas turbine performance digital twin NAPS50074.2021.9449682
for real-time embedded systems. In Proceedings of the ASME 97. Zhao, Y., Zhou, Y., Li, S., Zhao, M., Zheng, Y., Fu, C.: Distri-
Turbo Expo, Vol. 5. https://doi.org/10.1115/GT2020-14664 bution network reconfiguration digital twin model based on bi-
82. Steindl, G., Stagl, M., Kasper, L., Kastner, W., Hofmann, R.: level dynamical time division. In 2021 International Conference
Generic digital twin architecture for industrial energy sys- on Power System Technology (POWERCON), pp. 2178–2187.
tems. Appl. Sci. 10(24), 8903 (2020). https://doi.org/10.3390/ https://doi.org/10.1109/POWERCON53785.2021.9697854
app10248903 98. Huang, L., Liang, Y., Huang, H., Zhou, J.: Digital twin mod-
83. Tsoutsanis, E., Hamadache, M., Dixon, R.: Real time diagnostic eling and operating state assessment of substation equipment.
method of gas turbines operating under transient conditions in In 2021 4th International Conference on Energy, Electrical and
hybrid power plants. In Proceedings of the ASME Turbo Expo, Power Engineering (CEEPE), pp. 159–163. https://doi.org/10.
Vol. 5. https://doi.org/10.1115/GT2020-14748 1109/CEEPE51765.2021.9475674
84. Lei, Z., Zhou, H., Hu, W., Guo-Ping, L., Guan, S., Feng, X.: 99. Zhou, J., Chen, Y., Ran, L., Fang, H., Zhang, Y., Zhu, X., Jaidaa,
Toward a web-based digital twin thermal power plant. IEEE Trans. A.: Hybrid data-driven modeling for an ac/dc power system con-
Ind. Inf. 18(3), 1716–1725 (2022). https://doi.org/10.1109/TII. sidering renewable energy uncertainty. Front. Energy Res. (2022).
2021.3086149 https://doi.org/10.3389/fenrg.2022.830833
85. Xiangjun, Z., Ming, Y., Xianglong, Y., Yifan, B., Chen, F., Yu, 100. Darbali-Zamora, R., Johnson, J., Summers, A., Jones, C.B.,
Z.: Anomaly detection of wind turbine gearbox based on digital Hansen, C., Showalter, C.: State estimation-based distributed
twin drive, In 2020 IEEE 3rd Student Conference on Electrical energy resource optimization for distribution voltage regulation
Machines and Systems (SCEMS), pp. 184–188. https://doi.org/ in telemetry-sparse environments using a real-time digital twin.
10.1109/SCEMS48876.2020.9352321 Energies 14(3), 774 (2021). https://doi.org/10.3390/en14030774
86. Sahal, R., Alsamhi, S.H., Breslin, J.G., Brown, K.N., Ali, M.I.: 101. Feng, Z., Wu, Y., Gao, H., Zhu, S.: Digital twin framework for adn
Digital twins collaboration for automatic erratic operational data flexible resources assessment. In 2020 IEEE Eurasia Conference
detection in industry 4.0. Appl. Sci. 11(7), 3186 (2021). https:// on IOT, Communication and Engineering (ECICE), pp. 209–212.
doi.org/10.3390/app11073186 https://doi.org/10.1109/ECICE50847.2020.9302004
87. Jeroen, D.M.D.K., Stockman, K., De Maeyer, J., Jarquin-Laguna, 102. Han, J., Hong, Q., Syed, M.H., Khan, M.A.U., Yang, G., Burt, G.,
A., Vandevelde, L.: Digital twins for wind energy conversion sys- Booth, C.: Cloud-edge hosted digital twins for coordinated con-
tems: a literature review of potential modelling techniques focused trol of distributed energy resources. IEEE Trans. Cloud Comput.
on model fidelity and computational load. Processes 9(12), 2224 (2022). https://doi.org/10.1109/TCC.2022.3191837
(2021). https://doi.org/10.3390/pr9122224 103. Kovalyov, S.P.: Design and development of a power system digital
88. Olatunji, O.O., Adedeji, P.A., Madushele, N., Jen, T.C.: Overview twin: a model-based approach. In 2021 3rd International Confer-
of digital twin technology in wind turbine fault diagno- ence on Control Systems, Mathematical Modeling, Automation
sis and condition monitoring, In 2021 IEEE 12th Interna- and Energy Efficiency (SUMMA), pp. 843–848. https://doi.org/
tional Conference on Mechanical and Intelligent Manufacturing 10.1109/SUMMA53307.2021.9632191
Technologies (ICMIMT), pp. 201–207. https://doi.org/10.1109/ 104. Han, J., Hong, Q., Feng, Z., Syed, M., Burt, G., Booth, C.:
ICMIMT52186.2021.9476186 Design and implementation of a real-time hardware-in-the-loop
89. Li, F., Li, L., Peng, Y.: Research on digital twin and collaborative platform for prototyping and testing digital twins of distributed
cloud and edge computing applied in operations and maintenance

123
A review on digital twins for power generation and distribution 1195

energy resources. Energies 15, 6629 (2022). https://doi.org/10. 119. Zhang, Y., Luo, J., Zhu, W., Wu, Y., Zhang, X.: Application
3390/en15186629 of digital twins in smart grids (2022). https://doi.org/10.1109/
105. Tiago Rabelo, C., Marcos Aurélio Izumida, M., Kennedy Alves, ICPICS55264.2022.9873758
M., Amadeu Fernandes de, M., de Francisci, S.: Application study 120. Wagner, T., Mehlmann, G., Richter, M.: Application of the dig-
in the field of solutions for the monitoring distribution transform- ital twin concept for a distribution network. In IEEE Power and
ers of the overhead power grid. Energies 14(19), 6072 (2021). Energy Student Summit, pp. 1–5 (2020)
https://doi.org/10.3390/en14196072 121. Mourtzis, D., Angelopoulos, J., Panopoulos, N.: Development of
106. Farnell, C., Soria, E., Jackson, J., Mantooth, H.A.: Cyber protec- a pss for smart grid energy distribution optimization based on
tion of grid-connected devices through embedded online security. digital twin. Proc. CIRP 107, 1138–1143 (2022). https://doi.org/
In 2021 IEEE Design Methodologies Conference (DMC), pp. 1– 10.1016/j.procir.2022.05.121
6. https://doi.org/10.1109/DMC51747.2021.9529935 122. Zhou, M., Yan, J., Feng, D.: Digital twin framework and its appli-
107. Jiang, Z., Guo, Y., Wang, Z.: Digital twin to improve the virtual- cation to power grid online analysis. CSEE J. Power Energy
real integration of industrial IoT. J. Ind. Inf. Integr. (2021). https:// Syst. 5(3), 391–398 (2019). https://doi.org/10.17775/CSEEJPES.
doi.org/10.1016/j.jii.2020.100196 2018.01460
108. Chen, J., Deng, R., Guo, Y., Lin, W., Cao, W., Guan, K.: Research 123. Darian, L., Kontorovych, L.: Electrical power equipment digital
on network management technology of power line carrier commu- twins. Basic principles and technical requirements (2021). https://
nication in low-voltage distribution network based on digital twin. doi.org/10.1051/e3sconf/202128801029
In 2021 7th International Conference on Computer and Com- 124. Sellitto, G.P., Aranha, H., Masi, M., Pavleska, T.: Enabling a zero
munications (ICCC), pp. 2112–2116. https://doi.org/10.1109/ trust architecture in smart grids through a digital twin. In Com-
ICCC54389.2021.9674420 munications in Computer and Information Science, Vol. 1462, pp.
109. Chen, Y., Chen, Q., Gao, J., Chen, X.: Hardware-in-loop based 73–81. https://doi.org/10.1007/978-3-030-86507-8_7
digital twin technology for integrated energy system: a case study 125. Hu, C., Shi, W., Jiang, L.: Application case of digital twin tech-
of guangyang island in chongqing, copyright - Copyright The nology in electric power system. IOP Conf. Ser. Mater. Sci. Eng.
Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2022 (2020). https://doi.org/10.1088/1757-899X/788/1/012083
Last updated - 2022-08-18 (2022 2022). https://doi.org/10.1109/ 126. Ma, Q.: Application of digital twin and hologram technology to
CIEEC54735.2022.9846475 achieve distribution network reliability forecast (2022). https://
110. Li, C., Yang, D.: Construction of power grid digital twin model doi.org/10.1109/ACPEE53904.2022.9784043
based on gan. In 2021 China Automation Congress (CAC), pp. 127. Ruhe, S., Nicolai, S., Jiang, T., Cai, H., Sayed, N. E., Geithner,
7767–7771. https://doi.org/10.1109/CAC53003.2021.9728190 T., Frueh, N., Ulbig, A., Schoenfeld, B., Prinz, S.: Approach of a
111. Saad, A., Faddel, S., Mohammed, O.: IoT-based digital twin for dt based automatic grid operation for distribution grids, In ETG
energy cyber-physical systems: design and implementation. Ener- Congress, pp. 1–6 (2021)
gies 13(18), 4762 (2020). https://doi.org/10.3390/en13184762 128. Bai, H., Yuan, Z., Tang, X., Liu, J., Yang, W., Pan, S., Xue, Y.,
112. Saad, A., Faddel, S., Youssef, T., Mohammed, O.A.: On the imple- Liu, W.: Automatic modeling and optimization for the digital twin
mentation of IoT-based digital twin for networked microgrids of a regional multi-energy system. In 2022 Power System and
resiliency against cyber attacks. IEEE Trans. Smart Grid 11(6), Green Energy Conference (PSGEC), pp. 214–219. https://doi.org/
5138–5150 (2020). https://doi.org/10.1109/TSG.2020.3000958 10.1109/PSGEC54663.2022.9881075
113. Xie, J., Guo, J., Sun, M., Su, D., Li, W., Chen, S., Wang, S.: 129. Pan, H., Dou, Z., Cai, Y., Li, W., Lei, X., Han, D.: Digital twin
A digital twin five-dimensional structural model construction and its application in power system. In 2020 5th International
method suitable for active distribution network. In 2022 2nd Conference on Power and Renewable Energy (ICPRE), pp. 21–
International Conference on Electrical Engineering and Mecha- 26. https://doi.org/10.1109/ICPRE51194.2020.9233278
tronics Technology (ICEEMT), pp. 418–426. https://doi.org/10. 130. Jose, F.V., Norma, G.M., Ivan, R.F., Sebastian, R.D., Arnaldo,
1109/ICEEMT56362.2022.9862649 G.O., Cesar, A.C.: Digital twin of the medium voltage grid of
114. Brosinsky, C., Song, X., Westermann, D.: Digital twin - concept university city at unam. In 2021 IEEE PES Innovative Smart Grid
of a continuously adaptive power system mirror. In International Technologies Europe (ISGT Europe), pp. 1–6. https://doi.org/10.
ETG-Congress 2019; ETG Symposium, pp. 1–6 1109/ISGTEurope52324.2021.9640160
115. Huxoll, N., Aldebs, M., Baboli, P.T., Lehnhoff, S., Babazadeh, 131. Kummerow, A., Nicolai, S., Brosinsky, C., Westermann, D., Nau-
D.: Model identification and parameter tuning of dynamic loads mann, A., Richter, M.: Digital-twin based services for advanced
in power distribution grid: digital twin approach. In 2021 Inter- monitoring and control of future power systems. In 2020 IEEE
national Conference on Smart Energy Systems and Technolo- Power and Energy Society General Meeting (PESGM), pp. 1–5.
gies (SEST), pp. 1–6. https://doi.org/10.1109/SEST50973.2021. https://doi.org/10.1109/PESGM41954.2020.9354468
9543095 132. Schäler, C., Strasser, K., Damböck, R., Schwefel, H.P.: Increased
116. Xing, J., Sun, S., Yu, P., Li, Y., Cheng, Y., Wang, Y., Li, S., renewable hosting capacity of a real low-voltage grid based on
Zhu, J.: Multi-energy simulation and optimal scheduling strat- continuous measurements—results from an actual pv connection
egy based on digital twin. In 2022 Power System and Green request, In: Communications in Computer and Information Sci-
Energy Conference (PSGEC), pp. 96–100. https://doi.org/10. ence, Vol. 1462, pp. 90–98. https://doi.org/10.1007/978-3-030-
1109/PSGEC54663.2022.9881079 86507-8_9
117. You, M., Wang, Q., Sun, H., Castro, I., Jiang, J.: Digital twins 133. Which countries produce the most hydroelectric power? (2023).
based day-ahead integrated energy system scheduling under load https://www.weforum.org/agenda/2015/10/which-countries-
and renewable energy uncertainties. Appl. Energy 305, 117899 produce-the-most-hydroelectric-power/
(2022). https://doi.org/10.1016/j.apenergy.2021.117899
118. Tzanis, N., Andriopoulos, N., Magklaras, A., Mylonas, E., Birbas,
M., Birbas, A.: A hybrid cyber physical digital twin approach for
Publisher’s Note Springer Nature remains neutral with regard to juris-
smart grid fault prediction. In 2020 IEEE Conference on Industrial
dictional claims in published maps and institutional affiliations.
Cyberphysical Systems (ICPS), Vol. 1, pp. 393–397. https://doi.
org/10.1109/ICPS48405.2020.9274723

123