IEEE Electrification Magazine - March 2015 - 53

types-individual loads and communal (shared) loads,
such as street lights. The total load is defined as the community-aggregated load. The DGs in a community
microgrid are either dispatchable or nondispatchable. The
dispatchable units can be controlled by the microgrid
master controller and are subject to technical constraints
depending on the type of unit, such as capacity limits,
ramping limits, minimum on/off time limits, and fuel and
emission limits. The nondispatchable units, on the contrary, cannot be controlled by the microgrid master
controller since the input source is uncontrollable. The
nondispatchable units are mainly renewable energy
resources, typically solar and wind, which produce a volatile and intermittent output power. The intermittency
indicates that the generation is not always available and
the volatility indicates that the generation is fluctuating in
different timescales. These characteristics negatively
impact the nondispatchable unit generation and increase
the forecast error; therefore, these units are commonly
reinforced with DES. The primary application of DES is to
coordinate with DGs to guarantee the microgrid generation adequacy. They can also be used for energy arbitrage,
where the stored energy at low-price hours is generated
back to the microgrid when the market price is high. This
action is analogous to shifting the load from high-price
hours to low-price hours. The DES also plays a major role
in microgrid islanding applications. The microgrid community-aggregated load minus the local generation from
DGs and DES could be identified as the utility load, i.e., the
amount of load that should be supplied by the utility grid.
Smart switches and protective devices manage the connection between DERs and loads in the microgrid by connecting/disconnecting line flows. When there is a fault in
part of the microgrid, smart switches and protective devices
disconnect the problem area and reroute the power to
other loads, preventing the fault from propagating in the
microgrid. The switch at the point of common coupling performs microgrid islanding by disconnecting the microgrid
from the utility grid.
The microgrid scheduling in interconnected and islanded modes is performed by the microgrid master controller
based on economic and security considerations. The master controller determines the microgrid interaction with the
utility grid, the decision to switch between interconnected
and islanded modes, and the optimal operation of local
resources. It also controls and maintains the frequency and
voltages within permissible ranges. Communications, control, and automation systems are used to implement these
control actions and to ensure constant, effective, and reliable interaction among microgrid components.

Architecture
The common microgrid control architectures are either
centralized or distributed. The centralized model collects
all of the required information for the microgrid operation
and performs centralized control. In the distributed model,

however, each component is considered an agent with the
ability for discrete decision making. The optimal schedule
is obtained using iterative data transfers among the
agents. Both control schemes offer benefits and drawbacks, but the centralized model is emerging as a more
desirable approach as it ensures a secure microgrid operation and is more suitable for the application of optimization techniques. The main drawbacks of the centralized
scheme are reduced flexibility in adding new components
and extensive computational requirements.
In community microgrids, a hybrid model, which benefits
from both centralized and distributed models, is expected to
be adopted. Three levels of control can be identified for the
microgrid, i.e., the device level, the building (or consumer)
level, and the microgrid level. Adjustable loads would provide
the first level of control. These loads would receive real-time
electricity prices from the building controller as well as scheduling time interval, including operation start and end times,
from the consumer. Once the time interval is set, the adjustable load would autonomously start and complete the operation cycle. Adjustable loads are important components of a
community microgrid, which will offer load management
capabilities. Examples of these loads for residential consumers are washers, dryers, dishwashers, and pool pumps.
The second level of control will be performed in the
building. The building controller would coordinate the
schedule of adjustable loads and the DES. If the building
already includes backup generation, which is the case for
critical loads, the building controller would also control
and operate the backup generation. A building controller
is an intermediate controller between adjustable loads
and the microgrid master controller. Building controllers
act as agents that schedule their own loads while at the
same time communicating with the microgrid master
controller to reach the overall community microgrid goals.
(Facilities with several buildings that are themselves part
of a larger community microgrid may include an additional control layer between the master and individual building controllers. The objective of this facility controller is to
coordinate the dispatch of loads and DERs within its own
service boundaries.)
The last level of control, which monitors and controls
the entire microgrid, is the microgrid master controller. As
mentioned, the microgrid interaction with the utility grid
(i.e., the amount of power to be purchased/sold to the utility grid), the decision to switch to islanded mode (in the
case of upstream network faults or voltage variations),
and optimal operation of local resources (including load
schedules received from the buildings, dispatchable DGs,
and DES) will be performed by the microgrid master controller. It also controls and maintains the frequency and
voltages within permissible ranges. The ultimate operational objective of the controller is to optimally schedule
loads to maximize the utilization of local resources, manage power transfer with the utility grid, and minimize the
community-aggregated electricity payment considering
IEEE Electrific ation Magazine / March 2 0 1 5

Table of Contents for the Digital Edition of IEEE Electrification Magazine - March 2015