Sioshansi F.P. Smart Grid: Integrating Renewable, Distributed & Efficient Energy
Подождите немного. Документ загружается.
Others believe that non-traditional cost
recovery would be essential to accelerate
deployment of smart grid technologies.
Despite these challenges, smart grid technology deployment
is proceeding apace. The FERC's 2010 report on AMI and
Demand Response (FERC 2011) shows that AMI has been
deployed to 8.7% of total U.S. electric meters as of 2010, up
from 0.7% in 2006 and 4.7% in 2008. The FERC report,
however, also shows substantial unevenness in the penetration
of AMI technology:
• Investor-owned utilities (IOUs) lagged the national
average, showing a total of 6.6% AMI penetration,
compared to penetration rates above 20% for rural electric
coops and public power districts.
• States showed a wide disparity of AMI deployment: only
five states (AZ, OR, ID, WI, PA) exceeded 20% AMI
penetration, while the majority of states (29) showed AMI
penetration at less than 5%.
The fact that IOU AMI penetration lags the national average
is a concern, because IOUs serve some 75% of total
customers. And the unevenness of state action indicates
widespread resistance to universal AMI deployment, even as
several states move ahead rapidly. These data suggest that
unless some of the issues restraining AMI deployment are
resolved, the rapid growth of AMI in the last 5 years may
slow or even plateau. In fact, the FERC data indicate a
slowing in the growth rate in AMI deployment in the
2008–2010 period compared to the 2006–2008 period, from a
nearly ninefold increase to just under a doubling. While the
more than $4 billion in federal stimulus funding is estimated
691
to support the deployment of some 18 million additional
advanced meters, even the availability of federal grant funds
has not fully overcome state-level concerns about the net
benefits of AMI (see Maryland case below).
Three recent smart grid/AMI deployment case studies serve to
highlight the regulatory, perceptual, cost, and process issues
that can stymie otherwise promising projects:
Maryland. Baltimore Gas & Electric (BGE's) 2010 proposed
AMI deployment plan was initially rejected by the Public
Service Commission (PSC); after responding to PSC and
stakeholder concerns, a modified proposal was ultimately
approved [3]. The original proposal cost $835 million for
deployment of over 2 million meters, with an immediate cost
recovery surcharge mechanism, and mandatory TOU pricing
for all customer classes. Part of the urgency for this proposal
was a federal stimulus grant BGE had been awarded, which
would have covered some $136 billion of AMI costs; the
Department of Energy had asked for an August 2010 approval
from the PSC as a condition for grant funding.
The issues raised by the Maryland Office of People's Counsel
(OPC) summarize the concerns relating to demonstration of
benefits to customers [4]:
• About 20% of the benefits in the original proposal were
based on BGE operational savings in meter reading and
operations, and in distribution management costs.
• The other 80% of benefits were projected in terms of
capacity revenues from wholesale market forward-capacity
markets, reduced wholesale prices that would reduce future
standard offer rates, reduced energy bills as an effect of
692
dynamic pricing and customer feedback, and reduction in
T&D infrastructure costs.
• OPC pointed out that these 80% of projected benefits
were subject to considerable uncertainty, and they
depended heavily on customer response to price signals and
on future wholesale energy and capacity prices. In 2010,
wholesale market prices in the PJM power market had
shown signs of softening due to the recession; this shift
tended to support the OPC argument that such factors are
uncertain.
• OPC strongly objected to the cost recovery surcharge
element of the proposal, asserting that it places all of the
risk on ratepayers, with no comparable guarantee of
benefits.
• OPC rejected the mandatory TOU pricing element of the
proposal, stating that it imposes higher bills on some
customers, and that BGE did not include in its proposal any
customer-enabling technologies that would help them
respond to TOU prices, such as in-home displays or
automated appliance control devices.
• OPC also objected to the proposal on the grounds that it
would remove consumer protections by enabling remote
service disconnects, and also raised customer privacy and
security concerns that have not yet been fully addressed by
national efforts.
The PSC largely agreed with OPC's objections in its initial
order, stating:
The Proposal asks BGE's ratepayers to
take significant financial and
technological risks and adapt to
693
categorical changes in rate design, all in
exchange for savings that are largely
indirect, highly contingent and a long
way off. We are not persuaded that this
bargain is cost-effective or serves the
public interest, at least in its current
form…The Proposal is a “no-lose
proposition” for the Company and its
investors. (MD PSC Order No. 83410.
June 21, 2010, pp. 1, 3)
In response to the PSC's initial decision, BGE revised its
proposal to remove the automatic cost recovery surcharge,
treating the project's costs as a regulatory asset to be
recovered through subsequent proceedings, much like any
rate case. It also removed the mandatory TOU pricing
element, and included an expanded customer education and
outreach element. The PSC approved the revised proposal
later in 2010. For comparative purposes, the PSC approved
Pepco's AMI proposal, which also included a federal grant
cost offset, in large part because it did not include a special
cost recovery mechanism or mandate TOU pricing. This tends
to support the inference that guaranteed cost recovery without
risk or cost sharing between customers and shareholders, and
the imposition of mandatory time-based pricing, can be
significant obstacles to AMI deployment.
OPC also raised the issue that alternative ways to meet utility
peak load reduction and system operations goals can be less
expensive and less burdensome on customers. While this
issue did not become a central point in the BGE case, it has
been raised in other states, and can be a critical issue in some
situations. For example, in New Jersey the Department of
694
Public Advocate commissioned a study [5] on utility AMI
deployment, which focused largely on alternative approaches
to meeting the stated goals of the utility AMI deployment
proposals. Key points raised in this report include:
• Automated metering reading (AMR) is a less expensive
way to obtain a large fraction of the benefits of AMI.
Though limited in functionality (may not have hourly
reading capability or two-way communication potential),
AMR has been deployed by several utilities successfully as
based on avoided meter reading and operations costs. If
AMR is in place, the benefits of AMI deployment become
much harder to justify, as indicated in the Brattle Group's
analysis for the Institute for Electric Efficiency (IEE, 2011,
p. 13).
• Direct load control (DLC) and related utility load
management programs can serve to achieve the peak load
reduction benefits of AMI deployment. Utilities have been
conducting DLC programs cost effectively for many years.
• The Synapse report further analyzed the dependence of
forecast AMI benefits on three key factors: the average
load reduction per customer, percentage of customers
participating in dynamic pricing, and the long-term
persistence of savings. Synapse pointed out that utilities'
estimates for customer participant in dynamic pricing were
more than double historic participation rates for DLC
programs around the country.
Illinois. Commonwealth Edison planned one of the most
thoughtfully designed smart grid pilot projects yet conceived,
testing a variety of rate designs, information services, and
other features in a large sample of some 130,000 customers.
Com Ed filed its application with the Illinois Commerce
695
Commission (ICC) for a $360 million pilot program. In late
2008, the ICC approved $274 million in AMI cost recovery
charges for the pilot through a special rate rider. Because the
ICC had disallowed some of its costs from the 2007
application, the utility initiated an appeal with the Illinois
Appellate Court, mainly on matters related to labor costs.
This triggered a number of interventions in the appellate court
case, including the Illinois attorney general and Citizens
Utility Board
3
(CUB), objecting to the ICC's approval of the
special cost recovery rider. As was the case in Maryland,
consumer representatives objected to the accelerated,
automatic cost recovery features of Com Ed's cost recovery
mechanism. The appellate court case turned into an expanded
version of a full-blown ICC adjudicatory proceeding. In 2010,
the appellate court struck down the ICC's 2008 decision,
leaving Com Ed's pilot and its entire smart grid initiative in
regulatory limbo [6]. In its 2011 session, the Illinois
legislature sought to resolve some of these issues through
legislation. As of May, however, the governor was
threatening to veto the bill, and the situation remained
unresolved at this writing.
3
See Chapter 16.
While this case may reflect the nature of Illinois politics more
than the merits of smart grid technology, it does reinforce the
core point of the Maryland PSC's treatment of BGE's original
AMI application, in which the utility's original special rate
rider/surcharge mechanism was rejected as forcing ratepayers
to take all of the risk and front all of the costs for smart grid
deployment. The lesson that may lie at the core of both of
these case studies is that for regulated utilities, especially in
states with a history of vigorous consumer representation,
696
smart grid proposals are likely to be more successful if they
defer cost recovery to future rate cases or use other methods
that give stakeholders more opportunity to advance their
views.
Colorado. The Boulder SmartGridCity
™
project has been
widely heralded as an innovative smart grid experiment.
Working with the city of Boulder, one of the “greenest”
municipalities in the United States in its commitment to clean
energy and environmental protection, Xcel worked with its
business partners, including Accenture, Gridpoint, Landis and
Gyr, and Current Group, and other stakeholders to design a
state-of-the-art program, using fiber-optic technology as part
of its AMI deployment. Xcel conducted extensive customer
research via surveys and focus groups to help define the kinds
of offerings customers would prefer. Fifteen thousand Landis
and Gyr advanced meters were ordered for residential
installation beginning in 2008; the vision was ultimately to
include some 50,000 customers in the program [7].
SmartGridCity
™
was designed as an inclusive vision: it was
to include new renewable energy sources connected to the
grid, such that customers could choose low-emission power
pricing plans. It allowed for electric vehicle recharging, and
included utility-side upgrades for substations, feeders, and
distribution automation hardware and software. In the original
plan, customers were to be offered:
• Advanced meters that communicate with home appliances
for energy and cost savings
• Energy efficiency incentives integrated with the program
• Online tools for tracking energy use and making changes
to fit their lifestyle and efficiency goal
697
• The ability to automatically manage energy use based on
real-time price signals or green-power price signals
• Charging options for plug-in hybrid electric vehicles
• Automatic outage notification
• Automated load management options to reduce peak load
and prevent outages
• The ability to choose on-site renewable energy options
like wind, solar, and batteries
• Web-based tools to support the above features, including:
• A control portal: to set use preferences for major
appliances and household outlets.
• An analysis portal: for secure access to usage
information and analysis options.
• An information portal: to educate customers about
reducing costs and lowering their carbon footprint.
Customer choice was part of the program design, such that
customers would be able to (1) select a plan that works for
them, (2) specify their personal plan parameters, and (3)
monitor and control their energy use.
Xcel did not initially propose to recover costs from its
ratepayers using special rate riders or other mechanisms that
resulted in regulatory controversy in Illinois or Maryland. The
project was viewed primarily as a research and development
initiative, and the company did not feel compelled to require
official approval or cost recovery from the Colorado Public
Utility Commission (PUC). However, the project ran into cost
overrun problems: Xcel originally anticipated that capital
costs would be around $15.3 million in early 2008, but in
698
May 2009, it revised the number to $27.9 million, and in 2010
to $42.1 million. PUC staff estimate that total costs will
exceed $100 million. As the costs mounted, the PUC ruled
that Xcel would be required to apply for a Certificate of
Public Convenience and Necessity (CPCN), which will give
the PUC authority to regulate the project, including its cost
recovery. As that proceeding has unfolded, the City of
Boulder has asked to be removed from the proceeding, partly
to protect itself from exposure to cost recovery [8].
The PUC in December 2010 approved an Xcel rate increase,
primarily to pay for a new unrelated coal-fired facility. But
$11 million of this increase is to cover Boulder smart grid
costs. How the remainder of costs are to be recovered remains
to be soon; possibilities include the City of Boulder and its
customers, all Xcel customers, Xcel's business partners, and
Xcel shareholders. PUC staff were quoted in 2010 as saying
“… it is difficult to ascertain what portion ratepayers should
bear, if any.”
The program is in operation, with meters installed and many
of the planned services available. The company continues to
use it as an R&D project, However, the cost issue has placed
a shadow over the project. A significant part of the cost
overruns appears to stem from unexpected costs for
fiber-optic line installation. Xcel's PUC filing in May said that
more underground fiber had to be laid than expected, and that
installation costs rose due to “having to drill through granite
with diamond-tipped drill bits and remove large boulders with
cranes and dump trucks …”
SmartGridCity
™
turned into a major infrastructure
construction effort, using technologies with which there was
little local experience. This may in hindsight have been a
699
weakness in the program design. Difficulty in predicting and
controlling costs always creates risks in a new project. And
yet the project was deliberately intended to test new
technologies, so there may have been no other way to find out
what the challenges and the costs would be. The difficulty for
Xcel was that costs rose high enough to trigger regulatory
review. In hindsight, the company might have conducted the
project through an unregulated organization structure.
Perhaps the principal takeaway is that risks and costs for
smart grid technology, as for any new technology, can be
uncertain and substantial, and that risk management should be
part of the planning process.
Another takeaway from the Boulder project, again benefiting
from hindsight, is that many of the customer benefits could
have been tested and demonstrated prior to installation of the
core infrastructure. Enhanced information and feedback,
customer engagement and goal-setting, and even customer
device control can be piloted without AMI deployment.
Running a series of smaller pilots and field tests for such
customer-side elements prior to full construction mode for the
core infrastructure might have created an experience base that
could have built confidence in the benefits of AMI-based
customer offerings. Such a base might help weather later
challenges in the technology deployment process.
These three cases, in which AMI deployment has met
significant regulatory complications, were not selected to cast
a negative light on smart grid technology or its prospects.
They were chosen rather to illustrate the challenges that smart
grid deployment faces in traditional regulatory structures.
Like it or not, in many states these kinds of challenges are
likely to emerge, and utilities and other smart grid proponents
700