IEEE Systems, Man and Cybernetics Magazine - July 2019 - 27

Challenges for Demand Response
Many technical and nontechnical challenges remain for
the widespread adoption of demand response, and they
require additional research. Let us look at four challenges
in more detail: scalability, distribution of control, uncertainty, and aggregation.
To address scalability, a data-driven approach is needed. If
a certain aggregated demand response (in terms of required
power reduction/increase for a determined period) is fulfilled
with large industrial loads, then often one or two industrial
loads are enough and the control can be dedicated, taking all
process features into account. If commercial loads are used,
probably some tens of them are needed to cover the request,
but still the problem remains tractable and detailed process
information can be considered to reach a global optimum.
However, if residential loads are considered, then thousands
or more of these are required to fulfill the request and a model-based approach becomes too cumbersome. Thus, an
approach is required whereby the flexibility of the devices is
learned from the data delivered by them. This is more scalable than the a priori definition of device models.
To address the distribution-of-control problem, the most
appropriate paradigm must be determined (Figure 9). In a
case with only local control, there is no communication
between devices, and control decisions are only taken based
on locally measured parameters (voltage, power, known
price profiles). Often it is beneficial, however, to coordinate
control among entities, such that it is not only based on local
sensor information, but also on information communicated
from other entities. This can be done in different ways.
◆ In direct control, there is a direct interaction between
two entities. For instance, an intelligent EV charger
interacts with the car to determine the optimal
moment to charge the car.
◆ With centralized control, a central agent controls all
flexibility. This would occur in the case of a distribution system operator that sends an interruption signal
to specific loads to be shed.

Power (kW)

Energy (kWh)

electricity. In fulfilling its first objective,
the battery offers a guaranteed response. In
Energy Content
Battery Power
10
achieving its second objective, the battery
1
8
provides an additional benefit to custom6
ers. The varying power and energy bands in
0
4
Figure 8 indicate the remaining flexibility
2
for the self-consumption objective, while
-1
0
the rest is reserved for the frequency sup5
10 15
20
5
10 15
20
Time (h)
Time (h)
port. The different light-gray lines result
(a)
(b)
from the stochastic scenarios underlying
the optimization methodology.
Beside data-driven approaches toward Figure 8. Graphs showing (a) varying energy and (b) power
f lexibility aggregation for combined constraints to deliver frequency support, and associated battery
behavior for maximizing self-consumption [13].
objectives (e.g., market and technical), it
is also possible to rely on game-theoretic
analysis to determine where it is most beneficial to use
the flexibility [14].

Local

Direct

Centralized

Hierarchical

Peer to Peer
Figure 9. An illustration showing relevant control

paradigms for local and coordinated control.

◆ With a hierarchical approach, intermediate levels

ensure some scalability, such as the concentrator
agents shown in Figure 4.
◆ Within a peer-to-peer approach, components only
interact with some physical or logical neighbors in a
flat hierarchy.
Determining which one of these control paradigms is best
depends on the type of application for which the flexibility
will be used and on the actors involved (Table 1).
A next challenge is the uncertainty of the response to
the control signal. The ability of individual devices to deliver the required flexibility can depend on user behavior, grid
behavior, local circumstances, and external (e.g., weather)
parameters. In addition, energy markets (day-ahead, intraday, or balancing markets) are sources of uncertainty.
Ju ly 2019

IEEE SYSTEMS, MAN, & CYBERNETICS MAGAZINE

IEEE Systems, Man and Cybernetics Magazine - July 2019

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