IEEE Systems, Man and Cybernetics Magazine - July 2022 - 11

network. Thus, a platform is required to facilitate the
integration of a multitude of heterogeneous devices
through a unified protocol architecture and using the
existing electric grid infrastructure.
◆ Heterogeneity of applications: Besides providing
invaluable features, the smart grid concept has created
novel problems or introduced further intricacies to
the existing set of problems. For instance, realt
ime dynamic pr icing schemes, coordinated
charging of EVs, and more
sophisticated extensions of
demand response, such as
transactive energy, contribute
to the heterogeneity of smart
grid applications.
The intrinsic characteristics of
clouds, particularly the scalability
and the pay-as-you-go payment
method, can effectively address
many obstacles heightened by the
heterogeneity of devices and applications
in smart grids and, hence,
have introduced several cloudbased
solution concepts, which we
discuss in the following sections.
capacity offered by available and future RERs. Cloud
resources can advance power system planning by accelerating
the extensive computations.
Fog computing offers
cloud capabilities
closer to end users
by making a cloudto-thing
continuum,
while edge
computing offloads
the computational
burden of the cloud to
edge devices.
Power Dispatching
Cloud facilities can serve powerdispatching
automation systems to
improve the limitations caused by hardware resources in
power systems where the information infrastructure cannot
keep up with the expansion pace of these systems.
Accordingly, cloud platforms can transform the way largescale
power system simulations are managed. Security,
reliability, and fast recovery in response to extreme natural
disasters are crucial considerations for power-dispatching
systems and, if compromised, can lead to
wide-area blackouts. The vulnerability of on-premise systems
to natural disasters and cyberattacks has triggered
the idea of using virtual clouds and networks for powerdispatching
automation purposes.
Power System Planning
Cloud computing can assist independent system operators
with planning problems. The widespread use of distributed
energy resources (DERs) in electric grids has added further
uncertainty to the expansion planning of these systems.
Specifically, transmission network planning imposes more
of a computational burden since various criteria, e.g., thermal,
voltage, short-circuit, and electromagnetic transient
analyses, should be examined. Thus, system operators
need to handle these challenges besides the nonlinearity of
power system components, uncertainties on load forecasts,
forced outages of generation and transmission facilities,
and emergency operating procedures. Moreover, the retirement
of generation units will be tied to the generation
Demand-Side Management
The smart grid concept has significantly served demandside
management (DSM) applications, creating tremendous
business opportunities for enterprises such as the
NASDAQ-traded companies Opower and C3Energy, providing
value-added services to utilities
and consumers. However,
despite the computational needs of
network analysis applications,
which tend to have small fluctuations
under a steady network configuration,
DSM applications call
for scalable computation resources
since the associated computational
needs rely highly on the number of
resources to be scheduled, i.e.,
flexible loads. Furthermore, in
most DSM applications, computation
facilities need to grow in a
polynomial manner with respect to
the number of resources. Considering
the fluctuating computational
needs associated with DSM applications,
cloud computing can substitute
for expensive central
information processing facilities, enabling small-size businesses,
start-ups, and even individuals to contribute to this
field without incurring the high provisioning and maintenance
costs of computing resources.
Moreover, the data-sharing feature of cloud-based
infrastructures can create a transparent platform where
data can be deidentified and then made available to the
public. This characteristic helps businesses and third parties
to identify the specific needs of consumers and develop
customized intelligent applications accordingly.
Utilities can deploy cloud platforms to forecast the energy
demand based on real data coming from the consumer
end. Data-driven knowledge on future demand assists utilities
in scheduling the optimal operation of energy
resources and avoiding rolling blackouts. Cloud computing
offers software platforms needed to incorporate the realtime
and historical records of electricity usage data in programming
models that entail extensive computation.
Nevertheless, some smart grid applications necessitate
more stringent requirements that cloud-based platforms
fail to satisfy. We discuss the challenges arising in cloud
systems in the following section.
Cloud Computing Challenges
Despite numerous recognized advantages, cloud computing
still may not be a viable solution for many use cases. Transferring
data over long distances between users and the cloud
July 2022 IEEE SYSTEMS, MAN, & CYBERNETICS MAGAZINE 11
IEEE Systems, Man and Cybernetics Magazine - July 2022

Table of Contents for the Digital Edition of IEEE Systems, Man and Cybernetics Magazine - July 2022
