IEEE Systems, Man and Cybernetics Magazine - July 2019 - 23

However, if this additional electricity use can be moved to
valley periods when excess electricity is available, the
peak power will not increase. The capacity of the grid will
not have to be expanded to meet brief periods of high
demand. This is where demand response comes in.
Demand response (also called active demand or
demand-side management) enables changes in the size
or the timing of the demand for electricity. This might be
used, for example, to better follow a variable supply. This
is not new. Demand response has been applied for more
than 40 years. Demand response initiatives have steered
residential demand toward the night so that power plants
can run at highest efficiency. Examples of these efforts
includes electric appliances with a night demand for electricity (e.g., electric accumulation heating) and day/night
tariff schemes that give consumers an incentive to turn
on white appliances only in off-peak periods.
In moving toward a carbonless society, with a varying
supply from renewables and a decrease of classical conversion technology in the overall energy mix, demand-side
management must be reactivated to keep the balance.
Hence, we need to shift from a supply-follows-demand to a
demand-follows-supply paradigm. This becomes more feasible if there is more flexibility in the demand side, which is
(due to increased electrification) increasingly the case.
Flexibility in electricity consumption is enabled by
energy reservoirs, such as heat buffers (storage tanks or
vessels connected to electric boilers or heat pumps), thermal mass (driven by heating, ventilation, and air-conditioning systems), cold buffers (refrigerators, deep
freezers, cooling cells), and batteries [from EVs, including
plug-in hybrid EVs (PHEVs), and stand-alone batteries].
These reservoirs make it possible to decouple the time
when electricity is demanded from the time that energy is
used (hot water for a shower, keeping the building within
a comfortable temperature range, have a charged car
before departure, and so on). By using the heat pump to
heat water electrically and storing it in a vessel, a consumer can make use of the hot water later to keep the room
warm; the heat does not have to be generated when the
heating service is needed. Hence, it is decoupled. A similar story can be told about electricity use in commercial,
office, and industrial environments, mutatis mutandis,
where the amounts of power and energy are larger.
Demand response will use this flexibility to move the
electricity use to a different time, such as when the energy
services are required. This flexibility can be used locally,
or in an aggregated way, and target economic, ecologic, or
technical objectives.
Most importantly, demand response, which will be essential for infrastructure of the future, must support the customers. They need to be engaged and willing to contribute,
without being hassled with technicalities or organizational
challenges. It should be as invisible as the electricity infrastructure itself, but provide the energy service in sustainable
way in line with the roadmap to a carbonless society.

Characteristics of Demand Response
The flexibility of individual appliances and their energy
buffers at a local, residential scale is mostly limited to
some kilovoltamperes of power modulation (e.g., 2-4 kW
for a monophasic electric boiler, 20 kVA for a three-phase
EV charging station), and up to some tens of kilowatt
hours of energy storage (e.g., 8 kWh for an electric boiler,
20 kWh for an EV's battery).
This flexibility is often sufficient to meet local objectives, such as to maximize consumption of electricity from
local renewables that are behind the same point of common coupling and metering equipment, or to minimize the
electricity injection into the local grid. It can also be used
for providing grid support, i.e., to mitigate voltage issues
on weak distribution feeders. Incentives driving such local
objectives can be economic (minimizing cost of electricity), ecologic (maximizing the use of electricity from local
renewables), or technical (avoiding power-quality problems or minimizing grid losses).
This electrical flexibility from individual households
can also be used in an aggregated way, i.e., together with
the flexibility of other households or other actors. While
this requires that a communication and control infrastructure be in place, the accumulated flexibility can grow several orders of magnitude and reach megawatts of power
and megawatt hours of energy. This allows for flexibility
trading at wholesale markets, both for energy services (on
the day-ahead or intraday market), as well as ancillary services (markets for balancing or frequency support, for congestion avoidance, and so on). In the latter case, the
flexibility from the demand side has to compete with the
classical power plants, but many transmission system
operators are opening up their markets for these types of
smaller-scale demand response (e.g., http://www.elia.be/en/
products-and-services/product-sheets).
Before elaborating on demand response coordination,
it is important to understand the asymmetry of the flexibility, its longevity, and its uncertainty. While standalone batteries can be kept at an average state of charge
(SoC) of 50%, as to enable them to both take electricity
from the grid and inject it into the grid as needed, the
physical properties and usage characteristics of many
appliances imply that it is easier to increase electricity
consumption than to decrease it. For example, consider
an electric boiler for domestic hot water that is kept at a
certain average temperature (corresponding to a particular SoC). Depending on consumer behavior (hot tapwater profile), such a boiler might be turned off 80% of
the time. When it is off, it can be turned on (until the
maximal temperature is reached), but it can only
decrease its electrical consumption in the 20% of time
that it is actually running. Additionally, the temperature
(or comfort) limits determine how long this flexibility
can be provided. The Local Intelligent Networks and
Energy Active Regions project (www.linear-smartgrid.be)
analyzed the flexibility potential in a large-scale pilot
Ju ly 2019

IEEE SYSTEMS, MAN, & CYBERNETICS MAGAZINE

http://www.elia.be/en/products-and-services/product-sheets http://www.elia.be/en/products-and-services/product-sheets http://www.linear-smartgrid.be

IEEE Systems, Man and Cybernetics Magazine - July 2019

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