IEEE Electrification - December 2022 - 31

Power ramping rate capacity defines the capability of a DER
to adjust its power input and output within a limited
amount of time. Power capacity defines the amount/interval/range
to which the power input/output of a DER can
change. For a producer DER, it is the maximum and minimum
power output of the generation. For a consumer DER,
it is the maximum reduction in power consumption. For a
prosumer DER, it is the maximum charge and discharge
rate. Energy capacity can be calculated by integrating power
changes over time. For an ESS, it is defined by the energy
capacity of the batteries. For producer and consumer DERs,
it is defined by the power capacity and the service duration.
Apart from these three main characteristics, many
other characteristics affect the quality of the flexibility service
of a DER, but they can be summarized as ramp-up/
ramp-down duration, service duration, rebound effect, and
respond duration. Figure 1 depicts some key features that
determine DER flexibility. The flexibility of grid-edge DERs
can be quantified if one knows the models of the DERs and
the operation preferences. Model-predictive control is an
effective approach to quantify the flexibility. Without direct
access to DER information, machine learning techniques
can be used to estimate DER flexibility by learning typical
consumption/generation patterns from historical data.
Aggregating and Optimizing Grid-Edge Flexibility
Control strategies are required to efficiently aggregate and
use the flexibility of grid-edge DERs. As shown in Figure 2,
five types of control architectures can be employed: centralized
control, hierarchical control, decentralized control,
distributed control, and hybrid control.
1) Centralized control is the most common control structure
applied in the power industry today. A centralized
control strategy involves a central controller that gathers
all of the relevant information about various DERs,
conventional loads, and the network connecting
them. The information could be generation cost functions,
DER operating constraints, customer preferences,
network parameters, and the forecasts of grid-edge
DERs. This information is then processed by the central
controller, and the optimal set points are dispatched
to the grid-edge DERs
to be implemented. Although
centralized control strategies
theoretically can provide global
optimal solutions for a given
problem, they are not likely to
efficiently handle a large number
of control variables introduced
by the vast number of
grid-edge DERs due to computational
and communication
bottlenecks.
Power
Capacity
(kW)
2) In a hierarchical control strategy,
there are two or more layers
of controllers, wherein the
Control Signal Activation
Response
Duration
RampUp
Duration
Figure
1. Key features shaping DER flexibility.
IEEE Electrification Magazine / DECEMBER 2022
31
Service
Duration
RampDown
Duration
controllers
in a particular layer have some level of
autonomy but need to coordinate their operation with
the controllers in the upper or lower levels. In a hierarchical
control strategy, grid-edge DERs will communicate
only with their corresponding controllers at the
higher control layer, and the upper-level controllers
solve the problems with a reduced number of variables,
which provides some level of relief from computational
and communication bottlenecks compared
with the centralized strategy.
3) A decentralized control strategy performs in an antithetical
manner to a centralized control strategy in that
there is no central controller involved to coordinate the
operation of the grid-edge DERs. Instead, each DER will
be assigned to its own local controller, which could be
a DER aggregator, a building energy management system,
or a microgrid controller. Each local controller will
optimize and control the DERs within its jurisdiction
using local data and measurements.
4) A distributed control strategy is similar to a decentralized
strategy in that there is not a central controller that
controls every DER. However, distributed control goes
one step further, in that local distributed controllers
can communicate and coordinate with each other to
achieve globally optimal solutions. Distributed control
strategies provide the highest level of autonomy to the
controllable entities while also preserving their privacy
and providing a high level of redundancy. A utility control
center can also be integrated into the distribution
control structure to monitor grid operating constraints.
5) A hybrid control strategy employs a mix of the four
other control strategies mentioned. The example
shown in Figure 2 combines hierarchical and distributed
strategies, where a hierarchy of controllers is used
to provide the set points to the DERs, and the DERs
themselves can coordinate with each other. Depending
on the nature of the legacy control structure and the
ownerships of grid-edge DERs, a hybrid control strategy
can be tailored to incorporate different control
structures to achieve the most efficient grid-edge DER
control with a seamless control structure transition.
Power
Ramping
Capacity
(kW/h)
Energy
Capacity
(kW)
Rebound
Effect
Time (h)
Power (kW)
Adjustment
Direction
IEEE Electrification - December 2022

Table of Contents for the Digital Edition of IEEE Electrification - December 2022
