IEEE Electrification - September 2021 - 14

TECHNOLOGY LEADERS
Kilo-megawatt Systems (SHARKS)
program proposed a CCD approach to
find completely new concepts of tidal
and riverine energy converters. The
highly coupled dynamics of these
systems make them ideal candidates
for the CCD approach (see Figure 7).
Similarly, the extremely coupled
dynamics present in microgrids
make them an ideal candidate for
the CCD approach. Microgrids are
composed of many subsystems that
interact dynamically, including conventional
and renewable generators,
power electronics, flexible loads, EVs,
ESSs, control, communications, and
protection systems. The higher the
subsystem dynamic interactions, the
more needed the CCD methodology.
The design of optimal microgrids,
composed of electromechanical systems,
cyberphysical systems, and
techno-economic solutions with disparate
mathematical descriptions,
require multiple areas of expertise in
a CCD framework (see Figure 5).
As shown, the three CCD methods
involve a concurrent and iterative
engineering effort that redesigns the
components, networks, and control
solutions of the microgrid at each
iteration of the optimization process.
The combination of the three methods,
with the engineering creativity
of the control-inspired paradigms
(A1), the mathematical co-optimization
techniques (A2), and the
co-simulation campaigns (A3), will
definitely open the door to radically
new optimal designs of microgrids.
A Metric Space for Microgrid
Design Guidance
Metrics play a key role guiding
research and technical innovation.
This section proposes a new metric
space to apply the CCD methodology
to microgrids. Metrics quantify the performance
of microgrids and facilitate a
graphical understanding that guides
optimal designs and operations.
Figure 8 depicts a conceptual picture
of the metric space. It is composed
of two orthogonal metrics: M1 and M2.
The first metric (M1) measures the
resiliency and reliability of the
microgrid under a set of tests. The
resiliency is computed as the availability
of the microgrid to maintain the
energy supply under some predefined
events, and the reliability is calculated
as a dynamic margin in terms of frequency
and voltage damping of the
microgrid to those events. The second
metric (M2) is a technoeconomical
evaluation of the microgrid, which
includes costs and grid services. Putting
both metrics in an orthogonal
space, it is easy to 1) estimate the performance
of existing microgrids; 2)
define the objectives with bounds that
represent a tradeoff between the two
metrics (see the dashed line), with the
areas of interest where an optimal
design should land; and (3) find a path
to guide the research and innovation
efforts to accomplish those objectives.
The metric space captures some
of the key aspects of the CCD philosophy
for the microgrids introduced in
the previous section. In particular, it
describes the dynamics and control
aspects with the M1 metric, and the
steady-state calculations and cost
analysis with the M2 metric.
The M1 metric is composed of two
functions, f1 and f2, as shown in (1).
The first function f1 is a second-order
polynomial of the availability Ay, and
the second function f2 is a secondorder
polynomial of the dynamics Dy,
as presented in (2) and (3).
Mf Af D
1 yy
=+
12
^^
fA AA^h=+ +aa a
2
12 10yy y
fD DD^h=+ +bb b .
2
22 10yy y
hh (1)
(2)
(3)
The availability Ay measures the
number of load losses under some
contingencies during the set of tests,
as shown in (4).
A =
y
nc
cc p
k
np
11 1
nn nncp dco
zk++
k
zkp
++
nd
bb b// /zk
1- == =
^ ^^
^
dd
k
h hh
h
,
(4)
M2 .
f(kWh/$)
With Grid
Services,
Renewable
Penetration, Size,
and so on
Objective
(≥ Line)
Area of Interest
where the load type (tiers) are classified
as critical " c, " priority " p, " and
discretionary " d, " being the number
of customers of each load type nc, np,
and nd, respectively, the number of
load losses zc, zp, zd, respectively, the
number of contingencies in the study
nco, and the weight of each respect
ive load
d
b
bb b++ =
b =d 031
..
Microgrids
State of the Art
M1 . f[Availability (Resiliency), Dynamics (Reliability)]
Figure 8. The metric space required for microgrid design and optimization.
14
IEEE Electrification Magazine / SEPTEMBER 2021
There are many practical ways to
define the set of tests and predefined
events or contingencies that evaluate
availability and damping. As an example,
a test composed of five periods in
sequence, with five contingencies per
period and 5 min between contingencies
was chosen. The five periods are 1)
(S1) microgrid connected to the grid for
c , bp and bd , with
cp 1 . A typical weight scenario
of each respective load can be
defined as:
b = .,
c 036 b = .,
p 033 and
IEEE Electrification - September 2021

Table of Contents for the Digital Edition of IEEE Electrification - September 2021
