IEEE Systems, Man and Cybernetics Magazine - July 2019 - 29

response algorithms. Performance can be further tailored
by also including data that pertains to related phenomena
that may affect demand less directly; examples include the
forecasted and actual values of renewable energy production, of weather conditions, and of energy market prices.
Big data and related cloud technologies have proven their
worth to collect, process, and manage this abundance of
data in a timely, reliable, scalable, and secure manner.
Figure 11 presents such a cloud-based approach-
SmarThor-that is being developed at EnergyVille.
Besides serving in data collection, management, and provisioning roles, SmarThor provides a platform-as-a-service
application environment to host long-running demand
response and optimization applications, and, crucially in
smart grids, a control interface to various systems, including the building management systems at the local building.
While designed and built as a generic and reusable data
platform, the first application to make use of the SmarThor platform implements demand response to maximize
self-consumption of locally produced renewable energy to
charge a fleet of (hybrid) electrical vehicles.
Considerations
A number of interesting paradoxes pop up if one considers
the broader infrastructural perspective of demand response
within smart grids. A first paradox is that the flexibility
added to the electricity system by demand response makes
electricity generation (i.e., supply) less flexible. When
demand response becomes more widely available to better
accommodate electricity generation from renewables,
incentives to modulate the less-flexible classical power
plants (such as nuclear or coal-fired power plants) become
less appealing, effectively leaving them in the market.
A second paradox focuses on the increased resilience
and increased vulnerability of the electrical grid infrastructure. With mounting bottom-up control and distributed intelligence, the resilience of the grid can be increased:
outages can be detected more quickly and be covered
more locally. However, this decentralized control can also

start interfering with the higher-level top-down control
mechanisms, leading to oscillations in set points or to cascading effects of failures. Additionally, the need of information and communication technology for operating the
grid brings in an additional point of failure.
The different objectives for which demand response is
used (market, technical, or prosumer objectives) can also
lead to conflicting incentives relating to the devices that
provide flexibility. Low electricity prices could lead to an
increase in demand by some demand response applications, while the resulting under voltage would call for a
decrease in demand by the same flexibility providers. The
value of flexibility still is a hot research topic [14], [16].
This increase in demand response opportunities also
brings in new actors in the energy field (e.g., aggregators,
energy service companies, prosumers) while prodding
existing actors (e.g., system operators, utilities) to take on
new roles. Such developments have implications for the regulatory and legal frameworks that need to be in place. In
Europe, these changes have been driven since the late
1990s by directives concerning market openings, and in the
fourth package of directives (the so-called winter package,
"Clean Energy for All Europeans," outlined at the end of
2016 and expected to be implemented by 2020), Europe
fully envisages an active role of the consumer/prosumer
interacting directly with neighbors in trading energy.
In this context, the individual customer, engaged in
energy efficiency and economic delivery of electricity,
plays an ever more important role. Bottom-up groups in
society, cooperatively working together to increase their
sustainability and decrease their carbon footprint, are
eager to take on this path of being an active customer.
They are seeking alternative measures that do not require
central control. Instead, they want to remain in control
themselves and are sensitive to privacy issues and related
concerns. In this context, such new technologies as peerto-peer energy trading and smart grid control [17], [18], or
smart contracts and blockchain technology (e.g., Energy
Web Foundation, www.ewf.org) are becoming prominent.

Control

Web

Captor 2
Captor 3
Captor 4

Relational
DB
Table
Storage

Project
X

Project
Y

Project
Z.1

Project
Z.2

Project API

Smart Grid

Application Environment

Real Time

Data API

Captor 1

Project
DB

Figure 11. A diagram showing the cloud-based SmarThor data platform. DB: database; API: application

programming interface.

Ju ly 2019

IEEE SYSTEMS, MAN, & CYBERNETICS MAGAZINE

http://www.ewf.org

IEEE Systems, Man and Cybernetics Magazine - July 2019

Table of Contents for the Digital Edition of IEEE Systems, Man and Cybernetics Magazine - July 2019