IEEE 2001 WM Panel;
Role of Distributed Generation in Reinforcing the Critical Electric Power
Infrastructure.
Title : Microgrids
Panelist; Bob Lasseter
Introduction
Environmentally friendly renewable energy technologies such as photovoltaics and clew
efficient, fossil-fiels technologies such as micro-turbines and fbel cells are among new
generating systems driving the demand for distributed generation of electricity. If
combined heat and power at resident~ industrial plants or commercial building can be
achieved the efficiencies can be higher than conventional central generation plants. The
smallness of these new distributed generation along with the low voltages at the interface
create a new class of problems which requires innovative approaches to managing and
operating the distributed resources. One such concept is the microgrid.
Emerging Generation Technologies
The trends in technology points toward smallness, under the 500kW level. An excellent
example are the small gas fired micro-turbines in the 25-100 kW range that can be mass
produced at low cost. They are designed to combine the reliabfity of on board
commercial aircrafl generators with the low cost of automotive turbochargers. These
systems are high speed turbines (50,000-90,000 rpm) with air foil bearings. They are
small and use power electronic to interfiwe with the load. Examples include Honeywell’s
75-kW Turbogenerator, AllisorI/GE. 50-kW generator and Capstone’s 30 kW system.
Fuel cells are also well suited for distributed generation applications. They offer high
efficiency and low emissions, but today’s costs are high. Phosphoric acid cell are
commercially available in the 200 kW range, while solid oxide and molten carbonate cell
are in a pre-commercial stage of development. In October 1997 the U.S. Depart of
Energy and Arthur D. Little unveiled the “first-ever on-board gasoline powered fiel cell
for the automobile”. The possibility of using gasoline as a fbel for cells has resulted in a
major development effort by the automotive companies. This work is focused towards the
polymer electrolyte membrane (J?EM) fiiel cells.
In 1997 Ballard Generation Systems formed a strategic alliances with Daimler- Chrysler
and Ford to develop the next generation of efficient and clean engines for the world’s
vehicles using the Ballard’s PEM fhel cell. These engines are expected to meet the
performance and range between refieling requirements of transportation vehicles at a
projected fuel cell cost of $200 per kW. This together with much higher efficiency and
without the significant polluting emissions associated with the internal combustion
engine are expected to result in fiel cells capturing emerging markets. Many other
leading international automobde companies are making major investments in fiel cells
including General Motors, Hond~ Nis~ Volkswage~ Volvo, and Matsushita Electric.
0-7803-6674-3/00/$10.00
0-7803-6672-7/01/$10.00
(C)(C)
2000
2001
IEEE
IEEE 144
146
0-7803-6674-3/00/$10.00 (C) 2000 IEEE
If fhel cells for transpiration become a reality they will also provide a stationary power
source under the 100 kW range at a very competitive price.
Mixed fhel cell and micro-turbine systems will also be available as distributed
generation. In a joint DOE Westinghouse project a solid oxide fuel cell has been
combmed with a gas turbine creating a combined cycle power plant. It has expected
electrical efficiency of greater than 70 percent with low and and virtually zero . The
expected power levels range from 250-kW to 2.5-MW
Distributed resources include more than small generators and fbel cells. Storage
technologies such as batteries, ultracapacitors and flyWheelplay an important role.
Combining storage with micro-sources provide peak power and ride-through capabilities
during disturbances. Storage systems are fiu more efficient than five years ago. Flywheel
systems can deliver 700-kW for 5 seconds while 28-cell ultracapacitors can provide up to
12.5 kW for a few seconds.
Micro-source Issues
These technologies have very dflerent issues than normally found in traditional power
sources. For example the are applied very close to the customers load at voltage of 480
volts or less, require power electronics and different methods of control and dispatch. All
of theses energy technologies provide a dc output which requires power electronics to
interface with the power network and its loads. In most cases the conversion will be
petiormed using a voltage source converter with a possibility of phase width modulation
to provide f~ control of voltage magnitude.
Power electronic interfaces introduces new control issues and new possibfities. A system
with clusters of micro-generators and storage could be designed to operate in both an
island mode and connected to the power grid. One large class of problems are related the
fhct that micro-turbmes and fbel cell have slow response and are inertia-less. It must be
remembered that the current power systems have storage in generators’ inertia. When a
new load comes on line the initial energy balance is satisfied by the system’s inertia. This
results in a slight reduction in system frequency. A system with clusters of micro-
generators could be designed to operate in an island mode provide some form of storage
to provide the initial energy balance.
The control of inverters used to supply power to an AC system in a d~tributed
environment should be based on information available locally at the inverter. In a system
with many micro-sources, communication of information between systems is impractical.
Communication of information maybe used to enhance system performance, but must
not be critical for system operation. Essentially this implies that the inverter control
should be based on terminal quantities.
In many case there are also major barriers to connecting small generation to the electrical
network. These barriers range from technical issues, to business-practices and regulatory
issues. In a National Renewable Energy Laboratory report NRELBR-200-28053 it was
0-7803-6674-3/00/$10.00
0-7803-6672-7/01/$10.00
(C)(C)
2000
2001
IEEE
IEEE 145
147
0-7803-6674-3/00/$10.00 (C) 2000 IEEE
found that interconnection cost ranged from zero to over $1000/kw. These highest
number would more than double the cost of most micro-turbines.
Microgrid Concept
To overcome the issues discussed above a microgrid concept is presented as a way to
bring value to both the utility and the customer. A microgrid is a cluster of micro-sources,
storage systems and loads which presents itself to the grid as a single entity that can
respond to central control signals. The heart of the microgrid concept is the notion of a
flexible, yet controllable interface between the microgrid and the wider power system
This intefice essentially isolates the two sides electricall~ and yet connects them
economically. On the tilde, the conditions and quality of service are determined by the
rnicrogrid, while flows across the dividing line are motivated by the prevailing need of
the transmission system.
From the customer side of the interfhce, the microgrid should appear as an autonomous
power system ibnctioning optimally to meet the requirements of the customer. Such
issues as local voltage, reliability, losses and quality of power should be those that
support the customers’ objectives. From the wider power system side, however, the
microgrid should appear as a good citizen or even model citizen to the power grid.
The microgrid structure stratifies the current strictly hierarchical centralized control of
the power system into at least two layers. The upper layer is the one with which current
power engineers are f-, that is, the high voltage meshed power grid. A centralized
control center dispatches a limited set of assets in keeping with contracts established
between electricity and ancillary services buyers and sellers, while maintaining the
energy balance and power quality, protecting the systen-qand ensuring reliability.
The customer views the microgrid as a low voltage neighborhood of the power system
that obeys the upper layer central command center only to the extent that its behavior at
the node is in keeping with the rigorous requirements of the grid. Within the microgrid,
standards of operatio~ and methods of control could diverge significantly from the
norms traditional power systems to meet the needs of the user.
Microgrids have unique internal protection requirements. It is evident that the protection
requirements are dif%erentthan tridtiorud utility considerations and the protection
requirements for microgrids, both at the grid intetiace and internally. While traditional
protection schemes may be adequate at the interface, a difllerent set of considerations may
govern the protection internal to the rnicrogrid which is highly dependent upon the short
circuit currents provided by the micro-sources and power grid. For example the micro-
source can be rapidly isolated from the microgrid followed by a long restoration process.
Whh storage on the dc bus the inverter can control the short circuit current without
shutting down the microgrid while the fiudt is cleared. Equally important is how can we
design the inverter’s control along with the protection to minimim the contribution of the
rnicrogrid to fault current sensed in the transmission grid?
0-7803-6674-3/00/$10.00
0-7803-6672-7/01/$10.00
(C)(C)
2000
2001
IEEE
IEEE 146
148
0-7803-6674-3/00/$10.00 (C) 2000 IEEE
The control of the microgrid also have unique requirements. The bade assumption is that
of plug-and play for the fast response needs and a control/communication system to
provide the “acceptable characteristics” to the grid for slower responses.
The success of microgrids depends on how well each micro-source integrates into a
cluster of similar sources. In fact the preferred model is one of “plug-and-play” or
“appliance” like. In such a model, techniques for control of distributed resources that
depend on SCADA-like centralized communication and coordination will not be effective
for microgrid. The control of the micro-source should be based on information available
locally. In a system with many micro-sources, iiast communication of information
between sources is irnpracticaL Communication of tiormation may be used to enhance
system performance, but must not be critical for system operation
At the same time, the microgrid should appear as an autonomous power system which
meets the requirements of the customer. Voltage, reliabtity performance, and quality of
power should be those that support the customers’ objectives. From a control prospective,
techniques need to be developed which significantly lower the system complexity
encountered with the addition of extra micro-sources to a microgrid. The presence of
inverter interfaces in fiel cells, photovoltaics, micro-turbmes and storage technologies
creates a dtierent situation when compared to more conventional synchronous generator
sources in power sources and standby emergency power systems. Taking advantage of
the properties of the power electronic interface to provide additional functionality to the
microgrid and localized inverter control technology, along with a minimal amount of
short energy storage at the dc bus forms the basis for the approach. Various features of
the inverter control could include: plug-and-play features, seamless connection and
isolation from the electric grid, independent control of reactive and active power, ability
to correct voltage sags and system imbalances are all critical for the creation of a
microgrid.
0-7803-6674-3/00/$10.00
0-7803-6672-7/01/$10.00
(C)(C)
2000
2001
IEEE
IEEE 147
149
0-7803-6674-3/00/$10.00 (C) 2000 IEEE
