Sioshansi F.P. Smart Grid: Integrating Renewable, Distributed & Efficient Energy
Подождите немного. Документ загружается.
Moreover, the increasing capabilities of building energy
systems will make continuous commissioning more feasible
for residential consumers as well as large commercial and
industrial ones. Smart homes and buildings will be able to
monitor equipment performance, optimize comfort and
efficiency, and convey environmental performance [21]. They
will be able to target opportunities for efficiency with
stunning accuracy [22]. In the smart grid, energy efficiency
will become intrinsic to how we operate our buildings and
live our lives.
Operational Efficiency
Modernizing equipment and automating operations can
significantly improve performance, reduce losses, and reduce
the cost of operating the grid. A recent EPRI report [23]
indicates that most of the cost and benefits of the smart grid
will occur in the distribution network (Figure 1.5). While
most utilities have automated their business processes, many
have not automated their field operations. The oft-repeated
anecdote is that utilities know when the power is out at some
locations—especially at the end of radial lines—when
consumers call and tell them. They then often diagnose the
problem by dispatching crews to drive the distribution lines
with paper maps in hand until they spot the problem. At the
transmission level, of course, operators may have
sophisticated sensors, software, and controls at their disposal.
Nevertheless, there are significant opportunities for
improvement throughout the grid.
121
Figure 1.5
Total smart grid costs.
Source: EPRI, 2011
New technologies and systems are needed to implement
proactive maintenance and repairs, reducing failures and
increasing the life of aging equipment. Significant energy
savings are possible through techniques such as voltage
optimization [24] and dynamic line rating. Many utilities have
begun to address this issue by installing sensors,
communications, and controls at critical points in their
system. Simply knowing the status of various devices in the
field in near-real time can enhance the ability of operators to
optimize power flow and quality. Digital meters equipped
with the ability to communicate at each account location can
reduce the cost of reading the meters, automatically report
local outages, allow operators to remotely connect and
disconnect accounts, reduce theft, monitor voltage levels, and
122
provide many other benefits. More progressive utilities have
installed automated systems with dynamic digital maps and
sophisticated management tools. These features not only
improve the efficiency of utility operations, but also open up
new levels of service and can make working conditions safer
for field workers.
Given their different starting points and driving motivations,
the full benefit of automating operations will vary
substantially from utility to utility, with highly customized
applications for each. Some, for example, may emphasize
automatic fault detection because of the high cost of repairing
complex underground systems. Others may emphasize
automatic switching to reduce the duration of an outage to
remotely located consumers. Eventually, the smart grid may
provide ancillary services like frequency regulation by
automatically adjusting demand with storage devices or by
varying certain loads. In this vein, the Pacific Northwest
National Laboratory, through its parent laboratory manager,
recently licensed a frequency sensor/controller chip that can
be installed inexpensively in various appliances and used to
adjust loads in real time, thereby moderating frequency
exertions on the grid automatically with little or no central
control [25].
While full and complete grid automation is a noble goal, in
isolation it is neither mission critical nor will it likely be cost
effective. In designing and building the smart grid, it will be
important not only to optimize performance but also to
optimize the cost of operating the grid.
123
Demand Management
When consumers decide to buy more computers, heaters, air
conditioners, or more and larger televisions, they expect their
electric utility to make sure that electricity is available to
power the devices. This is true whether it's the hottest day of
summer or in the middle of a winter storm. Consumers do not
perceive the grid-related consequences or costs of their
decisions. As further described in Chapter 2 by Healy and
MacGill, the industry essentially disengaged the consumers
from the activities up-stream of the meter.
The easiest way to please the most consumers in the past has
been to build the system big enough and robust enough to
meet all of their needs, especially at times of peak
consumption. According to the Edison Electric Institute, in
2008 U.S. consumers used 4,110,259,000,000 kWh [26], just
over 4 trillion kWh, of electricity [27] (Figure 1.6). That's
nearly 14 MWh per person per annum, much above the global
average. For more than 100 years, we've built an
infrastructure to serve essentially any amount of electricity
that consumers request. We ask utilities to estimate the future
consumer demand for electricity and build the production and
delivery infrastructure—power plants, regional transmission
systems, and local distribution systems—to accommodate this
growth. And since we cannot store electricity—at least not
much—production has to match consumer demand at any
instant. Moreover, we ask utilities to build the system
10–20% larger than the anticipated demand to make sure we
can accommodate unexpected growth or loss of critical
generation or transmission lines [28].
124
Figure 1.6
U.S. Electricity Flow, 2009 (Quads).
EIA Website
In North America, the Federal Energy Regulatory
Commission (FERC) and the North American Electric
Reliability Corporation (NERC) and its affiliated regional
reliability groups oversee system plans and operations to
ensure that the lights stay on. Among many other things, they
track trends in electricity use and ensure that the industry is
properly planning for and building the power production and
transmission capacity that are needed to meet demand. In
2009, NERC predicted that production capacity for electricity
in the United States will grow from just over 1,000,000 MW,
or 1 Billion kW, in 2009 to just over 1,400,000 MW in 2018,
a 40% increase in nine years [29].
NERC also projects the corresponding consumed energy to
grow from just over 4 TWh at present to just over 4.5 TWh in
2018—a total increase of about 15%. There are at least two
reasons why production capacity is expected to grow more
than twice as fast as electricity demand.
• The first is that more than 50% of the expected capacity
increase is attributed to wind power, the availability of
which is rarely coincident with peak demand (typically less
than 25% of the time), whereas baseload plants—generally
coal-fired or nuclear—are typically available during peak
demand.
125
• The second reason is that, since the 1960s, our electricity
use has been increasingly weather-dependent and peaky.
Businesses tend to use more electricity in the mid-to-late
afternoon, while residences use more electricity in
mornings and evenings [30]. More than 10% of our total
national production capacity is required less than 100 hours
each year (Figure 1.7).
Figure 1.7
Typical demand response potential in CA.
Source: Data derived from California ISO (CAISO)
The net result is that in many regions, peak demand is
growing much faster than average demand, a trend that is
unhelpful if costs are to be kept low and plant factors high.
Similar trends are experienced globally. This explains why
reducing consumer electricity use during peak
times—commonly referred to as demand response—has
become a high priority.
126
There are varied technologies and programs that can make
demand response relatively easy and cost effective for
consumers. In 2009, FERC assessed current demand response
programs state-by-state to project the improvements that are
possible nationally [31]. The report identified a range of
possible options that could reduce overall peak demand 4%,
or roughly 38 GW, to as much as 20%, or roughly 188 GW,
over time, as further described in the book's Introduction. The
latter figure would require that utilities be much more
aggressive in seeking improvement in load factor and that
consumers be induced to participate more widely in various
demand response programs. In 2010, FERC issued a
companion report that lays out a national plan for making
demand response a priority [32]. Most recently, in March
2011 FERC released Order 745, which establishes rules by
which demand response resources should be paid for
participating in organized wholesale markets. While industry
reaction to FERC's initiative is mixed at best, the ensuing
debate will no doubt clarify many remaining questions about
the true benefits of demand response [33].
In the smart grid, the exact amount of demand response is less
important than the concept of creating a grid that allows the
electricity system to reduce loads readily, when necessary,
diverting electricity from noncritical to more critical and
essential loads. This must be done both automatically—such
as lights dimming in warehouses during peaks or refrigerator
temperatures or thermostats shifting a few degrees—and with
substantially enhanced consumer participation. Such elegant
energy management systems might involve consumers setting
their preferences and varying them when prices or conditions
change.
2
New technologies and systems might provide ways
to induce demand to follow changes in production—and
127
production costs—instead of production responding to
changes in load.
3
This approach should logically incorporate
dynamic rates so that consumer behavior is linked to the cost
of delivering energy at any particular time.
4
One of the
important benefits of this change to the grid would be the
potential to sharply reduce the reserve margin requirements
for utilities and grid operators. This requirement, absolutely
critical in our current grid system, could eventually become
obsolete, with huge savings in capital investments that are
reduced or rendered unnecessary.
2
Harper-Slaboszewica et al. in Chapter 15 titled “Set and Forget” describe
relatively convenient ways for consumers to become active participants
in balancing supply and demand.
3
Chapter 10 by Hanser et al. describes the role of load as a means of
balancing supply and demand as opposed to the traditional method of
relying entirely on adjustments on the supply side.
4
Chapter 3 by Faruqui describes the role of dynamic pricing in the future
smart grid.
We Would Want the Smart Grid to Be Clean
In spite of the growing political and societal concern about
climate change, renewable energy in the United States until
now has been a small contributor to power production,
approaching just 4% of the total generation in 2008,
excluding hydropower. There are many reasons for this:
historically high costs for renewable energy, inconsistent
federal and state incentives, and a lack of consumer
acceptance. For various reasons, this is now changing. Even
without clear federal mandates, dozens of states and cities
128
have made carbon-reduction commitments that continue to
drive markets and more will likely follow suit. Renewable
energy and clean tech companies are at last able to forecast
both stability and growth in long-term markets for their
products and services. New technologies, new companies, and
new business models are emerging.
While the expectations for national or global commitments to
carbon reduction targets are not currently encouraging, other
factors, such as the prospect for more stringent regulations,
such as those proposed by the U.S. Environmental Protection
Agency (EPA), will continue to put pressure on the industry
to switch to a cleaner generation mix. Figure 1.8 shows the
forecasted growth in renewable energy from the EIA (Energy
Outlook 2011). On January 29, 2010, President Obama issued
an executive order mandating certain reductions in carbon
emissions by federal facilities—a 28% reduction over 2005
levels by 2020 [34]. Even without clear federal mandates,
dozens of jurisdictions have made commitments that continue
to drive markets [35]. Almost no matter what the final
specific targets are and what form national and international
policy takes, it is clear that the smart grid must accommodate
the integration of massive amounts of carbon-free production
technologies.
5
5
The chapter by Lilley et al. (Chapter 7), for example, examines the
implications of a global push towards a cleaner energy mix in the electric
power sector and its projected benefits.
129
Figure 1.8
Projections of renewable energy generation capacity in the United States,
2009–2035 (GW).
Source: Annual Energy Outlook 2011, EIA, Reference Case
Renewable Resources
While the amount of solar generation in the United States
more than tripled between 2000 and 2008 [36], it still
represents only a fraction of 1% of the total U.S. generation.
Specific utilities have recently launched projects to explore
the impact of high-penetration renewable scenarios by
installing 30% or more solar PV in a particular neighborhood
and assessing the grid impacts. These projects are an attempt
to anticipate the market impact of very inexpensive solar
equipment: while predictions on the future cost of PV systems
are difficult, there is general agreement that costs will
130