IEEE Power & Energy Magazine - May/June 2016 - 36

Classifying Distribution-Level Energy
Management Approaches

Top-Down Switching

in a distribution-level te system, mid- to small-sized electricity consuming or producing devices automatically negotiate
about their actions with each other, with devices in the physical network, and with dispatch systems of energy suppliers
through efficient and scalable electronic market algorithms.
to debate the advantages of this approach, and to classify
different approaches, the "smart energy management matrix"
has been introduced. this matrix classifies smart grid energy
management approaches into four main categories. the vertical matrix axis distinguishes if an approach makes decisions
on local issues either locally or centrally. the horizontal axis
plots whether the approach uses one-or two-way communications. figure 1 shows this matrix with four general classes of
energy management approaches filled in: top-down switching, price reaction, centralized optimization, and transactive
control and coordination.

this quadrant contains the classical demand-response
programs where, typically in a certain grid area, one
device group is switched simultaneously following a
broadcasted signal. this is the simplest demand-response
approach, and it has been used successfully for decades in
different parts of the world. in the typical setup, a signal
sent out through the power grid by the local utility company switches off systems such as water heaters, and airconditioning systems during peak load periods. although
the approach is simple and effective, it does not unlock
the full response potential of devices, as the device state
is not taken into account. the expected system reaction is only known by using statistics, and as a result the
operation is based on worst-case scenarios. Most of all,
the method ignores the consumer altogether. it does not
take user preferences into account and interferes with the
autonomy of energy consumers.

Centralized Optimization
Classifying Distribution-Level Energy Management Approaches
Decisions
on Local
Issues
Made
Locally

Price
Reaction

Decisions
on Local
Issues
Made
Centrally

Transactive
Control

Top-Down
Switching

Centralized
Optimization

One-Way
Communications

Two-Way
Communications
(a)

Decisions
on Local
Issues
Made
Locally
Decisions
on Local
Issues
Made
Centrally

Price Reaction
+ Full Use of Response Potential
- Uncertain System Reaction
- Market Inefficiency
+ No Privacy Issues

Transactive Control
+ Full Use of Response Potential
+ Certain System Reaction
+ Efficiency Market
+ No Privacy Issues

Top-Down Switching
- Partial Use or Response Potential
- Uncertain System Reaction
- Autonomy Issues

Centralized Optimization
+ Full Use of Response Potential
+ Certain System Reaction
- Privacy and Autonomy Issues
- Low Scalability

One-Way
Communications

(b)

Two-Way
Communications

figure 1. The energy management matrix: the four main categories of (a) smart grid
energy management and (b) their pros and cons.
36

ieee power & energy magazine

in the centralized optimization
quadrant, local decisions are still
made centrally, but communications are two way. Here, a complex optimization engine oversees
all flexible demand and supply
in the smart grid cluster under
consideration (such as a virtual
power plant or a local grid segment). Based on available information and taking into account
the global and, perhaps, local control goals, the optimizer searches
for the best solution for the whole
system. all relevant local data
need to be communicated to the
optimizer, which informs the central controller that communicates
control signals or schedules to
the field.
Having the relevant local data
available as input to the optimization, the method is able to fully
unlock the response potential of
the individual devices. further,
as the central system performs
a direct control on the local
devices, the system-level reaction
of the response cluster is known
when a response is triggered. the
autonomy issue of the top-down
switching approach remains, and
a privacy issue is added as detailed
local information is now communicated. further, communicating
may/june 2016

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - May/June 2016