-
Comments of The General Electric Company
Carbon Pollution Emission Guidelines for Existing Stationary
Sources: Electric Utility Generating Units
DOCKET ID NO. EPA–HQ–OAR–2013–0602
RIN: 2060-AR33 79 FED. REG. 34,829 (JUNE 18, 2014)
SUBMITTED DECEMBER 1, 2014
-
1
TABLE OF CONTENTS
Introduction
.........................................................................................................................................
3
I. GE’s interest
........................................................................................................................................................
3 II. GE’s position on the climate issue
............................................................................................................
3 III. GE supports the goal of reducing carbon emissions, but
believes that several significant modifications to EPA’s proposal
are required
.......................................................................
5
Executive Summary
............................................................................................................................
6
General comments on EPA’s proposed guidelines
.......................................................................
8
I. Recommended modification to BSER definition
................................................................................
8 II. EPA must address stranded assets
.........................................................................................................
8
a. EPA should use its subcategorization authority to establish
standards for facilities that recently made substantial capital
investments
..............................................................................
9 b. EPA should acknowledge state authority to grant
facility-specific variances ................ 10
III. New Source Review
.....................................................................................................................................
13 a. Existing NSR requirements will impede the ability of EGUs to
improve efficiency and comply with the guidelines unless changes
are made to the proposal ...................................... 13
b. There are several approaches to NSR that would facilitate
compliance with Section 111(d)
.......................................................................................................................................................................
22
IV. Distinguishing 111(b) from 111(d)
..........................................................................................................
26 a. EPA’s final Section 111(d) guidelines should not impose dual
regulatory requirements on modified units
................................................................................................................................................
26 b. While states have the discretion to use new NGCC as an
emission mitigation strategy, a decision by EPA to include new
sources as part of a BSER determination under Section
111(d) would create significant legal risk.
................................................................................................
28 V. Accounting for CHP
......................................................................................................................................
29 VI. Fuel cells
............................................................................................................................................................
30
-
2
Comments on EPA Building Blocks
...............................................................................................
31
I. Building Block 1: Efficiency improvements
........................................................................................
31 a. EPA’s estimate of cost effective upgrades across the coal
fleet is unrealistic ................. 31 b. EPA should consider
operator experience, conditions, and the availability of new
technologies when calculating the impact of upgrades
...................................................................
32 c. Comments specific to best practices implementation
.............................................................. 34
d. Comments specific to equipment upgrades
..................................................................................
37 e. Estimate of total probable upgrade opportunity
.........................................................................
37 f. Emissions reductions through improvements in system
efficiency ..................................... 38
II. Building Block 2: Load shifting
................................................................................................................
39 a. A 70% NGCC utilization rate is achievable with adequate
phase-in time ........................ 39 b. EPA is correct to not
include existing NGCC Heat Rate Improvements as part of BSER 41 c.
EPA should clearly indicate that NGCC Heat Rate Improvements count
toward compliance in State Implementation Plans
............................................................................................
43 d. GE comments to request for comment in the NODA
.................................................................
43
III. Building Block 3: Renewable generation
............................................................................................
44 a. Proposed RPS Regional Approach is flawed
.................................................................................
45 b. NODA Regional Approach lacks sufficient detail for comment
............................................. 47 c. Detailed
comments on Alternative Approach
...............................................................................
47 d. Proposed RPS Regional Approach to quantifying RE
................................................................ 57
e. Credit for early action
..............................................................................................................................
59 f. Treatment of interstate renewable energy sales
.........................................................................
59 g. Increased RE should displace fossil generation in
goal-setting formula ........................... 60
IV. Building Block 4: Energy efficiency
........................................................................................................
60
Conclusion
........................................................................................................................................
63
-
3
Introduction
I. GE’s interest
The General Electric Company (GE) is actively involved in all
facets of the energy sector.
GE has long manufactured products that are designed to meet
stringent emissions standards, while also meeting customer
requirements for efficient, reliable generation. GE designs and
manufactures industry-leading gas, steam, and wind powered
turbines, gasification systems, nuclear power generation
technologies and transmission and distribution technologies,
including smart grid technologies, which are changing the
industry.
GE also invests in, owns, and operates power plants, so we
understand the perspective of plant operators and the challenges
they face in implementing controls while maintaining the
reliability and availability of their units.
GE also services the products we sell and offer upgrades to
products that we have sold that increase their efficiency and
availability.
No other company offers the depth or breadth of products and
services as does GE across fuels and technologies in the electric
generation and transmission and distribution sectors. This fact is
reflected in GE’s share of the installed base for electric
generating units in the United States:
64% (by MW installed) for gas turbines; 50% (by GW/h) for
coal-fired steam turbines: 39% (by MW installed) for wind turbines;
and 33% (by GW) for nuclear generation.
II. GE’s position on the climate issue
GE has long recognized and acknowledged that climate change is a
real problem and that production and use of fossil fuels contribute
to the problem. While all scientific issues relating to climate
change are not yet fully defined, particularly as to local or
regional impacts, we support international and national policies
charting a reasonable course to reduce emissions.
GE has acted consistently with our views on climate in the
operation of our businesses. GE has—―through our ecomagination®
initiative—―taken actions to increase the energy efficiency of our
operations and to reduce significantly our emissions of greenhouse
gases. Our commitments and performance are as follows:
-
4
Investment in clean-tech research and development: Since 2005,
GE has invested about $13B in ecomagination R&D, on track to
meet the commitment of $15B through 2015. In February, GE announced
the continuation of its ecomagination investment committing to
invest an additional $10B to reach a total investment of $25B by
2020.
Increase revenues from ecomagination products: In 2010, GE set
an ambitious goal of growing revenues from ecomagination offerings
at twice the rate of total company revenue in five years. In 2013,
ecomagination met this objective with revenue totaling $28
billion.
Reduce greenhouse gas (GHG) emissions 25 percent by 2015 and
improve the energy intensity of operations 50 percent by 2015: GE’s
energy efficiency improved 31 percent from the 2004 baseline year
(measured as energy/$ revenue). GHG emissions were reduced 32
percent from the adjusted 2004 baseline. Building off this success,
in February GE committed to reduce GHG emissions and freshwater use
by 20 percent, from the 2011 baseline, by 2020.
Reduce freshwater use by 25 percent and improve water reuse:
GE’s freshwater use was reduced 45 percent from the 2006
baseline.
While we support the need for national and international
policies and programs to reduce greenhouse gas emissions, GE
recognizes that climate change is a difficult and complex issue.
Climate change concerns are closely interconnected with the
production and use of fossil energy, and affordable, reliable, and
secure energy is fundamental to a healthy, growing, and competitive
economy. In addition, climate change is a global issue that
ultimately requires a global solution, and it is difficult to
address on a national basis as unilateral action is perceived to
result in a competitive disadvantage.
In our view, assuring affordable, reliable and secure energy is
as important as addressing the climate change issue for the simple
reason that a healthy, growing economy is essential to having the
wherewithal to address climate change. Therefore, we believe that
actions to address climate change must be balanced against their
potential economic impacts.
We believe strongly that an essential element for a reasonable
and cost-effective climate program is an adequate time period for
the transition to a lower carbon economy, particularly the power
generation sector, where the capital stock has a life measured in
decades and the cost of capital investments is measured in
billions. In these circumstances, a climate program for the power
generation sector must make every reasonable effort to avoid
stranding assets if the program is to be cost effective for
producers and consumers of electricity.
Finally, GE supports an “all of the above” approach to energy
and climate policy. We support increased reliance on renewables and
nuclear energy as a means of producing zero carbon electricity and
addressing climate change, but we also believe that fossil
-
5
fuels, particularly natural gas, must be a part of our energy
portfolio. Simply put, the continued use of low cost cleaner fossil
fuels, such as natural gas, is necessary if we are to have a
competitive US economy. While we see increased need to rely on
natural gas, we also believe that coal is also an abundant and low
cost fuel and will continue to be an important part of the global
and US energy mix for the foreseeable future.
III. GE supports the goal of reducing carbon emissions, but
believes that several significant modifications to EPA’s proposal
are required
Increasing the efficiency of coal-fired steam boilers and the
use of natural gas, renewable, and nuclear generation are critical
steps to reducing carbon intensity in the power sector. GE believes
that the proposed carbon pollution guidelines for existing electric
generating units represents a good faith effort to achieve these
goals, while providing the flexibility required by states and the
electric utility industry to reduce emissions in a cost-effective
manner without disrupting the reliability of the grid. However, we
believe that this rule should be improved to make it more flexible,
less burdensome, and more legally defensible.
-
6
Executive Summary
EPA’s proposed Best System of Emission Reduction (BSER) is
consistent with the actual operation of the power grid in Building
Blocks 1 through 3, but improvements are necessary. Compliance with
this system will promote lower emitting generation, and is
creative, but States and power-sector stakeholders need more
flexibility. EPA should base BSER on the first three proposed
Building Blocks, and should remain sensitive to consumer costs. EPA
must provide adequate time for the economy to transition to a
lower-carbon power generation sector. Capital assets in the power
sector have lifetimes typically measured in decades and costs
typically measured in billions of dollars. Providing adequate
transition time for asset owners to recover investment is critical
to avoid economic impingement. EPA must avoid stranding assets,
especially those facilities that have recently made substantial
capital investments to comply with new EPA programs, such as CSPAR
and MATS. Asset owners will not have the confidence to undertake
future environmental control retrofits if they are unable to trust
that they will be able to recoup their investments. Affording
federal relief to plants that have recently implemented
environmental controls in response to EPA mandates will prevent
consumer cost spikes and other economic fallout. EPA should ensure
that New Source Review (NSR) does not discourage efficiency
improvement projects necessary for compliance with state plans. EPA
should clarify in the proposed guidelines (and modify NSR rules or
interpretations accordingly) that efficiency projects undertaken
for compliance with State plans do not trigger NSR requirements.
Eliminating the NSR impediment to efficiency improvements would
encourage innovation and ensure that states and sources maintain
compliance flexibility under the proposed guidelines.
Compliance with State plans should not trigger
modification/reconstruction under CAA Section 111(b). EPA should
clarify that all physical and operational changes (including
upgrades) done to comply with Section 111(d) or that improve an
EGU’s energy efficiency are exempt from triggering NSPS
modification under the pollution control exclusion.
EPA should not impose dual regulatory requirements on units that
trigger modification under 111(b). Instead of the proposed dual
regulatory approach, EPA should confirm that a source can opt to
use the pollution control project exclusion for a
-
7
physical or operational change that improves efficiency or is
undertaken to comply with a state’s Section 111(d) plan.
EPA should not include new sources as part of a BSER
determination under Section 111(d). Including new sources as part
of a BSER determination under Section 111(d) significantly
increases the legal risk and uncertainty for the rulemaking given
that new and existing sources are regulated under separate
provisions in Section 111. Moreover, excluding new sources from the
BSER determination would create an important incentive for the
construction of new, higher-efficiency NGCC capacity.
EPA’s assumptions in Building Block 1 are unrealistic. A 6% heat
rate improvement on average across the fleet is technically
possible if costs are not a factor. A 4 to 5 percent gross heat
rate improvement on average is economically possible. Because we do
not know how many of these improvements are already implemented
across the fleet, we believe the actual potential may be less than
this range. Realistically, we expect a 1 percent gross heat rate
improvement to occur across the fleet. EPA’s assumptions in
Building Block 2 are achievable, given adequate phase-in time to
allow infrastructure to keep pace with increased gas demand. A 70%
utilization rate is technically achievable by NGCC plants today,
and will significantly reduce carbon emissions from the power
generation sector. Significant capital improvements to midstream
infrastructure may be necessary to achieve this level, so EPA
should allow for a reasonable phase-in time to accommodate this
re-dispatch. NGCC heat rate improvements should not be included in
BSER, or for purposes of calculating state emission rates, but
should count toward compliance in State Implementation Plans. EPA’s
should recalculate Building Block 3 reductions using its
alternative approach. GE supports the expanded use of renewable
generation to reduce the carbon intensity of the nation’s grid.
This is a proven method to reduce carbon pollution at reasonable
cost. EPA has underestimated actual potential for renewable
resources in many states, particularly wind energy potential. EPA
should recalculate Building Block 3 by undertaking a state-by-state
analysis of technical and market potential, with some changes to
its proposed methodology. EPA should base BSER on Building Blocks 1
through 3. The Agency could account for increased deployment of
energy efficiency measures even if it decides to base BSER on
Building Blocks 1 through 3. Reduction in demand functions
differently than increased lower- and carbon-free generation. Even
if energy efficiency is not used to calculate BSER, EPA can and
should factor those reductions directly in other parts of the
rulemaking package, including in the rule’s impact analysis.
-
8
General comments on EPA’s proposed guidelines
I. Recommended modification to BSER definition
GE agrees that cost-effective technology to control carbon
emissions from existing fossil-fired electric generating units does
not exist. Therefore, if emissions from the electric power
generating sector are to be reduced significantly the only way to
do so is to increase the efficiency of existing units; to encourage
greater utilization of lower or no carbon emitting technologies;
and to promote energy efficiency in the use of electricity. A
regulatory system as EPA proposes that seeks to align with the
actual operation of the power system holds potential as an
effective, workable solution to the problem of reducing emission of
carbon from the sector. However, we suggest that EPA base BSER on
the first three building blocks, with the revisions suggested
below, as such an approach is more consistent with how the power
system naturally works. We understand and support efforts to
increase demand side energy efficiency, but believe that this tool
is better used as a compliance option through State implementation
plans. In our view, demand side energy efficiency is outside the
“power system” and undermines the coherence of EPA’s “system” based
approach to Best System of Emission Reduction (BESR).
II. EPA must address stranded assets
GE believes that provision must be made in the final rule to
address the issue of stranded assets, particularly with respect to
coal-fired units that have recently installed expensive pollution
control equipment to meet other environmental requirements such as
the Mercury MATS and CSAPR. Utilities and others, including
subsidiaries of our company, have made such investments, and
opportunity must be provided to operate these plants profitably for
such time as is necessary to recoup those investments. We believe
that Section 111(d) of the Clean Air Act requires EPA to provide
relief for such facilities either directly in its guidelines or
indirectly by authorize states to do so in state implementation
plans. The President’s directions to EPA also require EPA to take
efforts to comply with environmental requirements into account in
developing rules under Clean Air Act Section 111(d). Allowing such
relief will also help lessen the costs of this program on consumers
of electric power in regions served by such facilities. Having
recognized the serious stranded cost implications of the proposed
rule, EPA’s recent Notice of Data Availability discusses and
requests comment on potential
-
9
alternative options to enable coal-fired facilities like GE’s
own Homer City subsidiary to recoup their pollution control
investments:
[T]o the extent that stakeholders are concerned that the tools
available to states under the proposal may, in some instances, be
inadequate to address concerns regarding stranded investments, an
additional way to address these concerns may be for the agency to
take account of the book life of the original generation asset, as
well as the book life of any major upgrades to the asset, such as
major pollution control retrofits. For example, in its modeling,
the EPA assumes a book life of 40 years for new coal-fired units.
The EPA requests comment on whether, and how, book life might be
either used as part of the basis for the development of an
alternative emission glide path for building block 2 or used to
evaluate whether other ways of developing an alternative glide path
(such as the phase-in approaches discussed above) would address
stakeholders’ stranded investment concerns. The EPA is providing
this additional information, arising from stakeholder concerns, to
allow additional continued engagement of stakeholders in the
comment process.
See 79 Fed. Reg. 64,543, 64,549 (Oct. 30, 2014). We appreciate
EPA’s attention to the stranded asset problem in the Notice of Data
Availability and the agency’s willingness to consider ways account
for the “book life” of major upgrades to covered facilities,
including expensive retrofits to achieve compliance with MATS or
CSAPR. To this end, GE echoes comments of its Homer City subsidiary
and strongly urges EPA (1) to consider subcategorizing and creating
separate standards for sources that have made recent capital
investments to comply with MATS and CSAPR, and (2) reaffirm states’
established variance authority and expressly identify variance
mechanisms by which facilities that have made substantial
investments to comply with EPA’s recently promulgated MATS or CSAPR
requirements may obtain the relief they need.
a. EPA should use its subcategorization authority to establish
standards for facilities that recently made substantial capital
investments
EPA could address the stranded asset issue by using its
authority under CAA section 111 to “distinguish among classes,
types, and sizes” and to consider costs when establishing standards
of performance, to develop separate 111(d) standards for existing
facilities that since 2010 have made substantial investments in
pollution control technologies to
-
10
comply with CSAPR and MATS.1 Standards for this subcategory of
sources could take several different forms. EPA could adopt an
emission rate standard that it determined sources within the
subcategory could reasonably achieve, or, consistent with one of
the variance options discussed below (cap-based variance), the
agency could establish an annual CO2 emissions cap for such
facilities, or criteria for states to use to develop such a cap,
that reflects a reduction (e.g., 20%) from 2005 levels. This type
of an approach would balance EPA’s dual goals of reducing CO2
emissions and not prematurely stranding assets, particularly those
just recently installed to comply with other EPA programs.
b. EPA should acknowledge state authority to grant
facility-specific variances
Homer City strongly recommends that EPA acknowledge and, in
certain scenarios endorse, states’ authority to issue variances
from EPA emission guidelines and to adopt alternative compliance
standards and deadlines for facilities meeting certain conditions.
CAA section 111(d)(1) provides that EPA “shall permit the State in
applying a standard . . . to take into consideration, among other
factors, the remaining useful life of the existing source.”
Similarly, 40 C.F.R. § 60.24(f) provides that states may apply less
stringent standards and longer compliance schedules to individual
facilities or “class of facilities” if the state demonstrates: (1)
unreasonable cost of control, (2) physical impossibility, or (3)
“other factors specific to the facility (or class of facilities)
that make application of a less stringent standard or final
compliance time significantly more reasonable.”2 It is worth noting
that the variance authorized by 40 C.F.R. § 60.24(f)(3) is distinct
from and broader than the other two categories, including the
concept of a facility’s remaining useful life. We believe that EPA
and Pennsylvania should readily rely on this specific regulatory
variance authority to allow facilities that have recently made
substantial capital investments time to recoup their investments
before being subject to the entirety of a state 111(d) plan. EPA
should embrace this authority, not shunt it aside. In the preamble
to the proposal, EPA takes the position that “no relief for
individual facilities would be needed” because EPA’s proposed
emission guidelines happen to take the form of “state emission
performance goals for the collective group of affected EGUs in a
state,” rather than a “presumptive standard of performance that
must be fully and directly implemented by each individual existing
source within a specified source category.”3 We disagree.
1 Similarly, 40 C.F.R. § 60.22(b)(5) provides that EPA
guidelines shall “specify different emission guidelines or
compliance times or both for different sizes, types, and classes of
designated facilities when costs of control, physical limitations,
geographical location, or similar factors make such
subcategorization appropriate.” 2 Emphasis added. 3 79 Fed. Reg. at
34,925.
-
11
However, the form of the emission goal has nothing to do with
the ability of a state, under CAA section 111(d)(1) to grant
variances which “take into consideration, among other factors, the
remaining useful life of the existing source to which such standard
applies.” EPA cannot circumvent this statutory provision merely by
changing how it may choose to prescribe state guidelines. Moreover,
we cannot accept the premise that the claimed flexibilities in
EPA’s portfolio approach obviate a state’s need to adjust its
emission target in account for any source-specific variances
granted. Unless states have the ability to adjust their emission
targets, variances would be of little utility, since any relief
granted to one source would have to be made up by another. To
provide meaningful relief and to help address the stranded asset
issue, states must have the ability to grant variances and to
adjust state emission targets commensurately.4 For its part, EPA
could codify criteria for when it would be appropriate for state
targets to be adjusted to reflect a variance. For example, EPA
could make clear that it supports the issuance of variances to
avoid stranding assets and the adjustment of state emission targets
to such variances. States could grant general variances to other
sources, but they would not be able to adjust their emission limits
in response to variances that do not satisfy EPA’s criteria (e.g.,
when a source has not made a capital expenditure since 2010). Below
we set forth two specific variance-based options that EPA should
authorize and endorse.
Option 1: Alternative Compliance Schedule
EPA should specify in the final rule that states should grant
variances pursuant to 40 C.F.R. § 60.24(f) to coal-fired
facilities, like Homer City, that made substantial investments in
pollution control technology since 2010 to comply with CSAPR or
MATS. A variance under this option would exempt the facility from a
state’s base 111(d) program for 20 years from the start of the
program. During the variance period, EPA could adjust their
applicable emissions target to exclude emissions from those sources
operating under the variance. At the conclusion of the variance
period, the state would have to readjust its emission target to
account for the source and the facility would be required to meet
all applicable state plan requirements.
4 See 40 Fed. Reg. at 53,343-44 (preamble to 1975 implementing
regulations for section 111(d) (“EPA’s emission guidelines will in
effect be tailored to what is reasonably achievable by particular
classes of existing sources, and States will be free to vary from
the levels of control represented by the emission guidelines [under
§§ 60.24(d) and (f)]. In most if not all cases, the result is
likely to be substantial variation in the degree of control
required for particular sources, rather than identical standards
for all sources . . . [I]t is up the States to decide whether less
stringent standards [under §60.24(f)] are to be applied permanently
or whether ultimate compliance will be required.”).
-
12
As noted above, such an alternative compliance schedule is
consistent with discussion of potential ways to consider of the
“book life” of major upgrades to a facility in EPA’s recent NODA.
The assumed book life for large environmental retrofits, such as
the dry scrubbers being installed at Homer City, has traditionally
been 30 years. Regulatory analysis underlying both CAIR and CSAPR
used a 30 year life for FGDs, as did the MATS Rule.5 While EPA has
more recently used 15 years for evaluating book life under CAA
section 111(d), EPA can certainly consider the “book life” of major
pollution control equipment to be 30 years given the large capital
outlays required and substantial operation and maintenance costs.
EPA’s endorsement of a 20-year alternative compliance schedule for
qualifying sources like Homer City is plainly consistent with the
statute and EPA regulations. As discussed above, states may
implement alternative compliance schedules pursuant to their §
60.24(f) variance authority, as supplemented by EPA guidance
setting forth specific eligibility criteria. Moreover, allowing
variances for those facilities that meet reasonable eligibility
criteria is consistent with EPA’s own authority to “distinguish
among classes, types, and sizes within categories of new sources
for the purposes of establishing such standards.”6 Although there
is not an explicit corollary to this language in CAA section
111(d), such may be fairly implied in light of EPA’s express power
to make such distinctions in establishing standards of performance
under section 111(b). Accordingly, EPA should consider such an
alternative compliance option and expressly allow states to (1)
grant eligible facilities 20-year variances, and (2) implement
corresponding adjustments in their state emission targets over the
variance period.
Option 2: Emission Caps
A second option would be for EPA to authorize and endorse states
to use their variance authority to subject sources to
site-specific, annual CO2 caps in lieu of the base 111(d) program.
Under this option, facilities would be subject to a CO2 emissions
cap that reflects a 20% reduction off of the source’s 2005
emissions. Qualifying facilities would again be those facilities
that have made substantial capital investments since 2010 and need
time to recoup those investments. Since many coal plants have and
will retire after 2005, the emission reductions from those plants
is 100 percent. Therefore, coal plants as a group will contribute
significant CO2 reductions to achieving supportable reduction
targets. The capped facility could operate pursuant to whatever
schedule it chooses, provided it stays below the cap. The facility
would also be able to run under a cap in any year and utilize the
“headroom” under the emission cap in subsequent years. The
5 See, e.g., Documentation Supplement for EPA Base Case
v.4.10_FTransport-Updates for Final Transport Rule. EPA
430-K-11-004, June 2011 at 51-51 (applying 30-year book life for
pollution control retrofits in CSAPR). 6 42 U.S.C. §
7411(b)(2).
-
13
obvious benefit to the source is that it would have the
flexibility to comply with state plan requirements through a permit
and operate at full load (when the plant has the lowest heat rate)
during select periods of time, as opposed to being subject to a
rate. The benefit to the state is that it would obtain a guaranteed
mass reduction of CO2 and would maintain the viability of EGUs
within the state. We believe that this approach would be
permissible under the CPP as proposed, as well as under the
variance authority of 40 C.F.R. § 60.24(f). However, we believe it
important for EPA to eliminate any confusion by specifically
recognizing this in the final rule or accompanying materials as a
viable option to avoid stranding assets. States are likely to have
more comfort adopting such a measure if EPA speaks to it directly
in the final rule.7
III. New Source Review
a. Existing NSR requirements will impede the ability of EGUs to
improve efficiency and comply with the guidelines unless changes
are made to the proposal
The Proposed Rule is premised on Building Blocks that will
require efficiency upgrades
EPA’s first category of approaches to reduce GHG emissions from
EGUs, “Building Block 1,” consists of changes to coal-fired steam
units that improve their heat rate and operating efficiency. As EPA
notes, these changes will increase the efficiency by which the unit
converts the energy in the fuel to electric energy, and will lower
the amount of carbon dioxide (CO2) emissions per unit of
electricity produced.8 EPA estimates suggest that these
improvements will yield 97 million ton/year fleet-wide reductions
in CO2 emissions,9 approximately 12 percent of the overall
reductions required by the proposed guidelines.
EPA also notes that while the potential for heat rate
improvements is greater for coal-fired steam EGUs, heat rate
improvements have the potential to reduce CO2 emissions from “all
types of affected EGUs”.10 GE agrees. EPA’s second category of
approaches to reduce CO2 emissions, “Building Block 2,” is based on
the re-dispatch of generation from higher emitting fossil
fuel-fired EGUs to less carbon intensive fossil fuel-fired EGUs, in
particular natural gas combined cycle (NGCC) units that were in
operation or
7 EPA recently issued a notice regarding conversion of state
emission rates to a state emission cap. See Carbon Pollution
Emission Guidelines for Existing Stationary Sources: Electric
Utility Generating Units, 79 Fed. Reg. 67,406 (Nov. 13, 2014).
Further explication of how “capped” facilities would be treated
within state plans – or potentially outside of state plans – would
be beneficial. 8 79 Fed. Reg. at 34,859/col. 1-2 9 EPA, TSD, GHG
Abatement Measures, at 2-39 10 79 Fed. Reg. at 34,859/col. 2
-
14
commenced construction as of January 8, 2014. As a result of the
increased role of natural gas units, EPA estimates that roughly
one-third of the overall CO2 reductions required will come from
this re-dispatch to gas.
The increased role of gas-fired generators will result in
increased demand for operational efficiency and reliability at
existing units. Many existing gas-fired units were put in service
years ago and, while highly efficient, have the potential for
additional heat rate improvements through the implementation of
technology advancements incorporated in more recent NGCC or natural
gas simple cycle turbines (SCGT). To facilitate this re-dispatch,
achieve optimal performance, and to generate electricity in the
most cost effective manner from these existing units, owners and
operators will look at upgrade opportunities to improve heat rate
and increase efficiency and reliability. In fact, given the
enhanced role of gas-fired EGUs will play as a result of the
proposed guidelines, the opportunities for equipment upgrades and
improved efficiency may be on par, and may even exceed the
opportunities available with coal-fired EGUs.
Equipment upgrades at existing fossil fuel-fired gas and steam
turbines can significantly improve performance, reliability, and
efficiency while reducing emissions
There are significant opportunities for improved performance at
each of the three main components of a gas turbine including: the
compressor that compresses or squeezes the incoming air; the
combustor that burns the fuel; and the turbine that extracts the
working energy from the exhaust gas. Increased efficiency will
lower CO2 emissions per MWH as well as the emissions of more
traditional air pollutants, such as nitrogen oxides (NOx). This is
an important objective of the proposed guidelines if the
re-dispatch of power envisioned in Building Block 2 causes
individual units to generate more power.
In addition to improving efficiency and lowering emissions per
megawatt of electricity produced, most if not all of the likely
equipment upgrades will also improve the reliability and durability
of the unit – an important aspect that cannot be overlooked given
their increased role in providing electricity. Reliability and
durability of gas units will be key in assuring that states can
develop plans that comply with the proposed guidelines without
risking service to residential, commercial, or industrial
customers. Equipment upgrades also often involve updating controls
and software packages that allow for a more responsive system.
Examples of specific upgrade options for gas turbines include
the following:
Advanced Gas Path Upgrade (AGP): GE’s AGP is an upgrade option
available for simple and combined cycle installations within the
footprint of the existing 7F gas turbine. The AGP upgrade lowers
fuel consumption while increasing maintenance intervals and
extending the life of parts. Located downstream from the combustor,
the gas turbine is exposed to extremely high temperatures, so
the
-
15
AGP upgrade also includes improved materials and cooling
technology that requires less air devoted to cooling. This allows
more airflow to be devoted to making energy, thereby improving
output and efficiency. The AGP upgrade will also include better
sealing technology so less air leaks around the turbine blades,
further improving efficiency. Overall, the AGP upgrade has the
potential to add about 10 MW to output (depending on the current
equipment configuration), while reducing CO2 by more than 2.6
percent. If the AGP upgrade is installed with GE’s OpFlex Suite of
software solutions, significant additional emission reductions can
occur as a result of up to a 50 percent faster warm/hot start-up
time. This results in up to 50 percent fewer emissions during
start-up periods. An existing 7F gas turbine with AGP technology
generating a net output of 525 megawatts of power can reduce its
CO2 emissions by around 11,400 tons per year, which is roughly
equal to removing 2,200 cars from U.S. roads.
DLN2.6+ combustor: GE’s Dry Low NOx (DLN) 2.6+ combustion system
for the GE 7F gas turbine incorporates advanced technology to
increase combustion stability, reduce emissions and improve
turndown, and extend outage intervals, thereby reducing the
frequency of maintenance outages and improving plant availability.
This improved combustor can combust natural gas more efficiently,
recovering more energy from the fuel with lower emissions. If
combined with an AGP upgrade (described above), the improved
combustor can add an additional 4 MWs for a total output gain of
approximately 14 MWs (total output approximately 186 MW) and reduce
CO2 by an additional 0.2 percent for a total of 2.8 percent.
Compressor Upgrade: The performance of existing NGCC units can
be improved with the installation of more advanced compressors. For
instance, the 7F.05 high-efficiency gas turbine can achieve
efficiency levels of 59 percent in combined cycle mode. This is due
in part to the fact that the 7F.05 turbine has a more advanced
compressor that is able to pass more air through the turbine and
more efficiently compress the incoming air. In addition to
improving efficiency, implementing the 7F.05 compressor upgrade on
earlier vintage 7F gas turbines will increase durability and
overall gas turbine availability. All of these features are
critical for the enhanced role of existing NGCC expected under the
proposed guidelines. If the 7F.05 compressor upgrade is coupled
with other upgrades, namely the AGP upgrade and the DLN2.6+
combustor upgrade described above (items 1 and 2), it can reduce
CO2 emissions by an additional 3.8 percent for a total emission
reduction of 6.6 percent per megawatt of electricity produced.
GE has also developed and deployed a number of significant steam
turbine upgrade technologies that have improved the performance,
operation and reliability of steam turbines while reducing
emissions. These include:
-
16
GE Energy’s Advanced Design Steam Path (ADSP): Because the
efficiency of a unit is largely dependent on the efficiency of the
energy conversion in the steam turbine, it is important to minimize
aerodynamic and steam leakage losses in the steam path. Introduced
in 1995, ADSP has resulted in steam path efficiency improvements
ranging from 1.5 to 3 percent. Along with ADSP, GE has also
developed advanced-vortex designs, integral cover buckets, and
brush seals that create additional improvements in efficiency.
Dense Pack Steam Path Redesign: The design goal of a Dense Pack
retrofit is
similar to ADSP; its aim is to put the most efficient steam path
into an existing high-pressure outer shell. The Dense Pack upgrade
inserts new rotor and stationary components inside the existing
turbine shell. The resulting high efficiency steam path produces a
lower heat rate and increases output for the same steam flow. The
Dense Pack design also has the benefit of reducing particle
erosion, such that the improved efficiency is more sustainable.
Installation of the Dense Pack Steam Path can be done during a
planned outage with minimal disruption.
Low Pressure Service Upgrades: The objective of low-pressure
service upgrades
is to recover performance losses attributed to unit aging by
installing advanced replacement components. These components
increase reliability due to their modern engineering and the
reduced likelihood of corrosion. Use of the low-pressure service
upgrades expand output and improve heat rate through improved
low-pressure section efficiency and reduced exhaust velocities. The
low-pressure service upgrades also have the advantage of reduced
maintenance costs by eliminating costly rotor repairs.
The mature age of many existing gas turbines and coal-fired
steam units suggests that repowering options may have significant
appeal and practicality as a compliance approach for the proposed
guidelines. Aging existing plants often suffer from reduced
reliability, limited OEM support, and rising cost of water and
wastewater disposal. Repowering addresses many of these challenges
by increasing engine reliability, significantly improving plant
efficiency, and reducing emissions. Repowering also allows existing
units to retain most of the existing plant infrastructure and can
be installed rapidly, with greater OEM support.
Without additional clarifications, NSPS modification
requirements may limit deployment of cost-effective compliance
options
As discussed in more detail below, despite increases in
efficiency, the increased output associated with upgrade projects
have the potential to increase emissions of pollutants on an hourly
basis, even though emissions per unit of electricity generated (per
MW/hr) decreases. Existing NSPS regulations recognize and address
concerns regarding the
-
17
potential application of additional requirements for projects
designed to control emissions. As explained in GE’s attached
October 16, 2014 comments on EPA’s proposed Carbon Pollution
Standards for Modified and Reconstructed Stationary Sources:
Electric Utility Generating (EGU), 79 Fed. Reg. 34,960 (June 18,
2014), GE supports the Agency’s decision to maintain the NSPS
pollution control project exclusion, which excludes from the
definition of “modification” the “addition or use of any system or
device whose primary function is the reduction of air pollutants.”
Although NSPS modification is triggered less frequently than NSR
modification, the energy efficiency focus of EPA’s proposed Section
111(d) guidelines could compel a larger number of upgrades at EGUs
that could lead to increases in a source’s maximum hourly emission
rate – the emissions trigger for NSPS modification.
Given the importance of this issue, GE recommends that the
Agency clarify in the preambles to the final Section 111
modification rule and in the final ESPS rule that not only
“traditional” pollution control projects—―e.g., installation of
scrubbers or baghouses—―qualify for the exclusion, but also
physical or operational changes undertaken to comply with CAA
Section 111(d) plans. This includes upgrade projects that improve
efficiency and, as a result, reduce the amount of CO2, and criteria
pollutants such as NOx, and PM2.5 emissions produced per MW of
electricity generated. In the case of compliance with Section
111(d), the definition of pollution control must be expanded to
encompass all actions aimed at compliance that improve efficiency.
While these projects do not always entail the installation of
specific pollution control “device,” they involve the “use” of a
pollution control “system,” the primary function of which is to
reduce CO2 pollution.
Without such an exclusion, the Section 111(d) plans could
effectively force existing EGUs to trigger NSPS modification
requirements merely through complying with an applicable state
plan. Such an approach would put both EGUs and state permitting
authorities in an untenable position. To eliminate this concern,
EPA should make clear in the final rule that any physical or
operational change undertaken to comply with a CAA Section 111(d)
plan qualifies as a “pollution control project” and is thereby
exempt from the NSPS definition of “modification.”
EPA is incorrect in stating that “few” sources will trigger
Prevention of Deterioration (PSD) and Non-Attainment New Source
Review (NNSR), collectively referred to as NSR, in complying with
the proposed guidelines
NSR is a preconstruction permit program that affects new sources
as well as the modification of existing sources. According to EPA
regulations, an existing source that undertakes a physical or
operational change that results in a “significant” emissions
increase will trigger NSR modification. The significance levels
vary by pollutant and can
-
18
be as low as an increase of 40 tons per year of NOx emissions or
10 tons per year of fine particulates (PM2.5).
Over the years, the NSR permitting requirements have discouraged
many owners and operators from undertaking energy efficiency
improvements at their existing units despite the availability of
new technologies to enhance performance and efficiency. As noted
above, equipment upgrades that would ultimately make the unit more
efficient – a 2 to 8 percent increase in efficiency – may also
elevate the position of the unit in the dispatch curve, which in
some circumstances could lead to emission increases measured on a
tons per year basis, thereby triggering NSR. As a result, many
projects aimed at improving efficiency, reliability and performance
were never undertaken, or were limited to ”like-kind” replacements
that did not obtain the full efficiency improvement available from
the upgrade but clearly avoided NSR.
EPA incorrectly concludes in the preamble that there will be
“few” instances where an NSR permit would be required in response
to the guidelines.11 As noted above, EPA’s proposed guidelines
require existing coal and oil-fired steam units to improve heat
rates – an action that will increase efficiency and output. For
many units, this may trigger NSR review unless operational limits
are taken which then has the perverse result of not allowing the
full effect of the efficiency improvements to be realized. For gas
turbines that are being required to dispatch at much higher levels,
many operationally and economically beneficial turbine upgrades to
improve reliability, performance, and efficiency may also trigger
NSR unless capacity limits are taken to restrict full deployment of
the efficiency upgrade. History, as well as the likely demands
imposed on existing sources under these proposed guidelines, will
significantly increase the number of units that could potentially
trigger NSR.
Existing NSR requirements will impede the ability of EGUs to
improve efficiency and comply with the Section 111(d) guidelines
and state plans
Unlike NSPS, NSR rules do not contain an applicability exemption
for pollution control projects. For this reason, NSR has
represented a substantial impediment to broad implementation of
otherwise feasible, cost-effective efficiency upgrade projects.
GE’s extensive experience working with utility customers suggests
that the fear of triggering NSR will discourage many existing EGUs
from making physical or operational changes that would improve
efficiency and facilitate compliance with Section 111(d).
For NGCC units, upgrades will generally increase gas turbine
output, improve efficiency, and reduce emissions per MW of
electricity produced. Overall, the upgrade will yield an
approximately 3 percent heat rate improvement resulting in a
reduced CO2 emission rate
11 79 Fed. Reg. at 34,928/col. 2-3
-
19
by 36 lbs./MWh. But the potential also exists for a unit to emit
more on an annual and even hourly basis whether or not the units is
dispatched more often. Thus, even though the emission concentration
will remain the same (e.g., 2 ppm NOx pre-upgrade and 2 ppm NOx
post-upgrade) and the rate of emission per unit of power will
decrease (e.g., by 0.02 lb./MWh), the mass emission rate will
increase slightly (e.g., by 0.5 lbs./hr. NOx). Thus, despite the
clear benefits, owners have forgone the energy efficiency benefits
from the upgrades due to the potential of triggering NSR permitting
requirements.
A key aspect of pre-reform NSR that discouraged sources from
considering upgrade projects include the current manner in which
EPA guidance required emission increases from a project to be
determined by comparing baseline actual emissions (highest actual
emissions over a 24-month period preceding a change) to future
potential emissions. As a result, if the facility operated at a
relatively low capacity factor, and the future potential emissions
are based on a higher capacity factor, the NSR significant emission
levels could easily be triggered. Under the NSR reform rules, which
apply in many, but not all, states, an actual-to-projected-actual
emission increase test applies. But even that test, as discussed in
more detail below, does not effectively and reliably exclude
emissions associated with increased utilization from an efficiency
upgrade.
Once triggered, the NSR permitting process can have many
negative consequences for a source, including construction delays.
Experience with the program suggests that major NSR permit reviews
can delay projects one to two years beyond the much-shorter minor
NSR permitting process. Moreover, because NSR requires case-by-case
review, there is often significant uncertainty regarding the likely
outcomes, including whether the project will be approved, and what
new permit conditions may be attached, including their potential
costs. EPA’s existing network of regulatory requirements is also
complex and constantly changing. Often customers are faced with the
problem of responding to new permit conditions from regulatory
changes that occurred following the permit application, creating a
difficult do-loop for permit seekers.
Many GE customers have historically expressed a general
reluctance to go through the NSR process due to the vagueness of
the program provisions and the extensive case law history around
the program. All of these reasons have lead customers to install
like or in-kind replacement parts instead of technology
improvements. Finalizing a Section 111(d) rule without including
regulatory changes that address NSR applicability would
significantly undermine the ability of EGUs to comply the proposed
guidelines.
-
20
Many of EPA’s proffered solutions for avoiding NSR are
ineffective and run counter to the objectives of the proposed
guidelines
In the preamble, EPA suggests that states will have
“considerable flexibility” to prevent the projects from causing an
emission increase:12
One of these flexibilities is the ability of the state to
establish the standards of performance in their CAA section 111(d)
plans in such a way so that their affected sources, in complying
with those standards, in fact would not have emissions increases
that trigger NSR. To achieve this, the state would need to conduct
an analysis consistent with the NSR regulatory requirements that
supports its determination that as long as affected sources comply
with the standards of performance in their CAA section 111(d) plan,
the source's emissions would not increase in a way that trigger NSR
requirements.
For example, a state could decide to adjust its demand side
measures or increase reliance on renewable energy as a way of
reducing the future emissions of an affected source initially
predicted (without such alterations) to increase its emissions as a
result of a CAA section 111(d) plan requirement. In other words, a
state plan's incorporation of expanded use of cleaner generation or
demand-side measures could yield the result that units that would
otherwise be projected to trigger NSR through a physical change
that might result in increased dispatch would not, in fact,
increase their emissions, due to reduced demand for their
operation. The state could also, as part of its CAA section 111(d)
plan, develop conditions for a source expected to trigger NSR that
would limit the unit's ability to move up in the dispatch enough to
result in a significant net emissions increase that would trigger
NSR (effectively establishing a synthetic minor limit). [317] p
34,928
EPA’s specific suggestions include having a state “adjust its
demand side measures or increase reliance on renewable energy as a
way of reducing the future emissions of an affected source” that
may trigger NSR.13 Alternatively, EPA suggests states could develop
“conditions” for a source expected to trigger NSR that would “limit
the unit's ability to move up in the dispatch enough to result in a
significant net emissions increase that would trigger NSR
(effectively establishing a synthetic minor limit).”14
While these proposed options have the potential to prevent a
source from triggering 12 Id. col. 3 13 Id. 14 Id.
-
21
NSR, they would also eliminate any incentive for an owner or
operator to make the required upgrade investment. The key reason to
invest in equipment upgrades in mature existing units is to recoup
the investment in increased efficiency, lower fuel costs, and
greater reliability. EPA’s proposal would ask sources to make the
investment but then limit the unit’s hours of operation,
effectively impeding the source’s ability to recover its
investment. This approach would effectively stop almost all
investments in existing units that could trigger NSR, thereby
eliminating many of the energy efficiency measures sources would
otherwise undertake to comply and increase system reliability.
On a more fundamental basis, the proposed solutions to limit
hours of operation also run counter to the objectives the proposed
guidelines to reduce CO2 emissions from electricity generation in
the most cost-effective manner possible. EPA notes repeatedly in
the proposed guidelines that states will have considerable
flexibility to design state programs, but then suggests limits on
major compliance options, such as improvements in energy efficiency
at existing units, even though these options will reduce emissions
per megawatt of electricity produced.
EPA’s proposed solution to have the state effectively limit the
dispatch of the unit is also problematic because it assumes
authorities that many states, as members of regional transmission
organizations, do not possess. Currently, the variable operating
cost of electric power generation is a key factor in determining
which units of a power system are dispatched. Plants with the
lowest variable operating costs are generally dispatched first, and
plants with higher variable operating costs are brought on line
sequentially as electricity demand increases.15 As yet, it is
unclear how the requirements of the proposed guidelines will impact
the structure and operation of electric markets. Any solution that
hinges on such an outcome is highly uncertain and unlikely to
encourage needed upfront investment in EGU upgrades.
Further, in the case of NGCC, restricting hours of operation run
directly counter to the objective of Block 2, namely re-dispatch to
lower CO2 emitting generation. An existing gas turbine EGU would
benefit from the increased output, efficiency, and cost
effectiveness of deploying an upgrade option. All of these factors,
along with the stated objectives of Block 2, would lead to the
desired result of increased cost effective gas turbine operation
leading to the ultimate goal of reduced grid-wide CO2 emissions. As
stated earlier in these comments, NSR permitting poses a direct
barrier to deployment of beneficial gas turbine upgrades, and
reducing operating capacity to avoid NSR permitting is directly
counter to every objective of the proposed rule.
15 EIA, “Electric generator dispatch depends on system demand
and the relative cost of operation” at
http://www.eia.gov/todayinenergy/detail.cfm?id=7590
http://www.eia.gov/todayinenergy/detail.cfm?id=7590
-
22
b. There are several approaches to NSR that would facilitate
compliance with Section 111(d)
Applying NSR to upgrades that improve efficiency or are being
implemented to comply with the proposed Section 111(d) guidelines
and resulting state plans is both anachronistic and
counterproductive. That said, the Agency has available several
practical, common sense approaches that are also legally supported
under the statute to prevent unintended results from the guidelines
and plans. With appropriate changes to the proposal, EPA can
implement the NSR program under the CAA and also maintain state and
source compliance flexibility, and unleash innovation, without fear
of triggering the costs, delays, and uncertainties associated with
NSR. EPA should move forward in the final Section 111(d) rule with
a clear preamble and corresponding regulatory provisions that would
allow sources to upgrade their equipment to improve the efficiency
of their overall power production (thus reducing CO2) without
triggering NSR.
The Proposed Rule seeks comment on a general approach to resolve
concerns about NSR applicability associated with projects
undertaken to comply with a state’s CAA section 111(d) plan:
We request comment on whether, with adequate record support, the
state plan could include a provision, based on underlying analysis,
stating that an affected source that complies with its applicable
standard would be treated as not increasing its emissions, and if
so, whether such a provision would mean that, as a matter of law,
the source's actions to comply with its standard would not subject
the source to NSR. We also seek comment on the level of analysis
that would be required to support a state's determination that
sources will not trigger NSR when complying with the standards of
performance included in the state's CAA section 111(d) plan and the
type of plan requirements, if any, that would need to be included
in the state's plan. P34,928-9.
GE believes that there are approaches that are logically
embedded in this general approach that could effectively address
concerns associated with NSR applicability. These approaches are
discussed below and can be used individually or in tandem to
accomplish the result:
Option 1: Interpret “increases” for EGUs on a pound-per-MWh
basis
Section 111(a)(4) of the CAA defines “modification” to mean “any
physical change in, or change in the method of operation of, a
stationary source which increases the amount of any air pollutant
emitted by such source or which results in the emission of any air
pollutant not previously emitted.” Because neither “increases” nor
“amount” is defined in
-
23
the Act, EPA retains discretion to determine what those words
mean in the context of the program at issue.16 This discretion
provides EPA, and states adopting plans pursuant to the guidelines,
with the flexibility to base NSR applicability on a pounds-per-MWh
basis, instead of or in addition to annual emissions.
EPA could use a pound-per-MWh emissions test as the sole
emissions applicability test for projects undertaken by EGUs. To
avoid unintended consequences with projects that do not have a
pounds per MWh decrease, EPA could also use a two-step emissions
test: EGUs would first determine whether or not a physical or
operational change results in an increase in emissions on a
pound-per-MWh produced basis; if pound-per-MWh are increased, EPA
could then revert to its traditional approach to assessing NSR
modification by determining whether the emissions increase in tons
per year exceeds de minimis levels for the regulated
pollutants.
Option 2: Use the routine maintenance, repair, and replacement
exclusion to address NSR applicability concerns associated with
efficiency improvement projects.
EPA could take the position that upgrades conducted pursuant to
an approved Section 111(d) project to improve efficiency are
covered by the routine maintenance, repair, and replacement
exclusion. Being required (or at least strongly encouraged) to
undertake efficiency improvements in Building Blocks 1 and 2
certainly suggests that the gatekeeper of “routine” has been
satisfied.
Option 3: Use state-wide netting (or a state-wide applicability
limit) to avoid a determination of emission increases.
Another way of achieving this same outcome could be for EPA to
state that sources that are in compliance with an approved
statewide plan are employing state-wide netting (or a state-wide
applicability limit) as a way of avoiding an emissions increase.
Given that EPA is relying on beyond-the-fence-line reductions as
part of the BSER under Section 111(d), EPA could also define
state-wide netting as a way to avoid triggering NSR. Thus, any
source in compliance with these statewide systems that has an
emissions increase by itself will not trigger NSR modification
because by definition of the state plan, the emission increase is
being offset by emission decreases elsewhere in the state. If this
approach is selected, it is critical that the netting exclusion be
able to apply on a pollutant-specific basis for all pollutants and
not be limited to CO2. As discussed above, despite reductions in
emissions per unit of electricity generated, efficiency improvement
projects have the potential to result in increases in emissions per
unit time of multiple pollutants, including NOx and PM. For this
reason, any NSR applicability solution, including state-wide
netting would need to apply across the range of pollutants
emitted.
16 [cite UARG Supreme Court decision on any air pollutant]
-
24
Option 4: Clarify that increases in emissions associated with
increased utilization of efficiency projects are properly excluded
under the demand growth exclusion.
While current PSD rules provide a regulatory mechanism for
excluding certain increases in emissions that result from increased
utilization due to product demand growth, this mechanism is
currently insufficient to provide regulatory certainty that
efficiency projects to comply with an approved state plan under
Section 111(d) will not trigger time-intensive and costly PSD
permitting requirements.
Current PSD rules provide that the calculation of “[p]rojected
actual emissions . . . [s]hall exclude . . . any increase in
emissions that results from the particular project, that portion of
the unit's emissions following the project that an existing unit
could have accommodated during the consecutive 24-month period used
to establish the baseline actual emissions under paragraph (b)(48)
of this section and that are also unrelated to the particular
project, including any increased utilization due to product demand
growth . . . .”17 This provision establishes two requirements for
excluding emission increases associated with demand growth. First,
the demand growth emissions must have been capable of being
accommodated during the baseline period. Second, the demand growth
must be “unrelated to the particular project.” In the first version
of EPA rules incorporating this concept, EPA highlighted the
causation requirement in the definition of modification for
providing the demand growth exclusion.18 However, in the 2002
rulemaking where EPA revisited the demand growth exclusion, EPA
explained:
On the other hand, demand growth can only be excluded to the
extent that the physical or operational change is not related to
the emissions increase. Thus, even if the operation of an emissions
unit to meet a particular level of demand could have been
accomplished during the representative baseline period, but the
increase is related to the changes made to the unit, then the
emissions increases resulting from the increased operation must be
attributed to the project, and cannot be subtracted from the
projection of projected actual emissions.19
Thus, while the 1992 WEPCO rule discussion focused on whether
the project caused the increase in emissions, the 2002 rulemaking
discussion focuses on whether the emission increases are related to
the project and, if so, they must be attributed to the project.
This difference in approach is reflected in EPA applicability
determination letters preceding and following the 2002 NSR reform
rule. For example, in a 2001 applicability letter, EPA stated,
“[p]rojected future actual emissions or representative actual
annual actual
17 40 C.F.R. § 52.21(b)(41)(ii)(c). 18 See e.g., 57 Fed. Reg.
32,314, 32,327 (July 21, 1992) (“demand growth can only be excluded
to the extent it—―and not the physical or operational change—―is
the cause of the emission increase.”). 19 67 Fed. Reg. 80,186,
80,203 (Dec. 31, 2002) (emphasis added).
-
25
emissions are determined by calculating only those emissions
increases that are caused by the modification.”20 In contrast, more
recent applicability determination letters have precluded the use
of the demand growth exclusion where the project is related to the
emission increases.21
In light of EPA’s interpretation of the current regulatory
language, the demand growth exclusion simply fails to provide
meaningful relief from PSD applicability concerns for efficiency
projects designed to meet improvements in efficiency associated
with the proposed ESPS. With amendments to its PSD rules and an
accompanying explanation in its rule preamble, however, GE believes
that the demand growth exclusion could be appropriately interpreted
(with a regulatory revision if necessary) to provide a useful
mechanism for clarifying non-applicability of PSD for
efficiency-improvement projects designed to meet improvements in
efficiency associated with the proposed ESPS. GE believes that this
relief could be provided with the following rule changes to EPA’s
PSD rule:
Change 40 CFR § 52.21(b)(41)(ii) to add new subsection
(b)(41)(ii)(e) as follows:
(e) For projects specifically undertaken to meet efficiency
improvements associated with [the proposed ESPS], emission
increases resulting from any increased utilization due to product
demand growth are deemed to not be related to the particular
project.
Change 40 CFR § 51.166(b)(40)(ii) to add new subsection
(b)(40)(ii)(e) as follows:
(e) For projects specifically undertaken to meet efficiency
improvements associated with [the proposed ESPS], emission
increases resulting from any increased utilization due to product
demand growth are deemed to not be related to the particular
project.
As these options suggest, EPA has many alternatives that it can
employ consistent with the CAA that would maintain state and source
compliance flexibility and unleash innovation without fear of NSR.
EPA should move forward in the final rule with a clear provision
that would allow sources to upgrade their equipment and improve
the
20 See Letter from Richard Long, Dir., Air and Radiation
Program, EPA Region 8 to Gary Helbling, Envtl. Eng’r, N.D. Health
Dep’t (Apr. 17, 2001) at Attach. A, 3 available at
http://www.epa.gov/region7/air/nsr/nsrmemos/otter.pdf (emphasis
added). 21 See e.g., Letter from Greg Worley, Chief, Air Permits
Section, EPA Region 4 to Mark Robinson, Plant Manager,
Georgia-Pacific Wood Products LLC (Mar. 18, 2002) at 1-2 available
at http://www.epa.gov/region7/air/nsr/nsrmemos/demandgrowth.pdf;
see also Letter from Dianne McNally, Acting Assoc. Dir., Office of
Permits & Air Toxics, EPA Region 3 to Mark Wejkszner, Manager,
Air Quality Program, Pa. Dep’t of Envtl. Protection (Apr. 20, 2010)
at 4 available at
http://www.epa.gov/region7/air/nsr/nsrmemos/psdanalysis.pdf (“the
facility must be able to demonstrate that excluded emissions are
completely unrelated to the project.”) (emphasis added).
-
26
efficiency of their overall power production without triggering
NSPS or PSD/NNSR. To the extent EPA believes any approach to
resolve this issue is not sufficiently tied to the proposal, we
note that the solicitation of comment was extremely broad such that
any of the above are reasonably viewed as a logical outgrowth of
the proposed action.22
IV. Distinguishing 111(b) from 111(d)
a. EPA’s final Section 111(d) guidelines should not impose dual
regulatory requirements on modified units
In EPA’s June 2014 proposed modification and reconstruction
rule, the Agency proposes to keep modified and reconstructed
sources in a state’s section 111(d) plan if at the time of
modification and reconstruction the source was subject to the
state’s section 111(d) plan:
For the reasons discussed in the “Legal Memorandum” . . . all
existing sources that become modified or reconstructed sources and
which are subject to a CAA section 111(d) plan at the time of the
modification or reconstruction, will remain in the CAA section
111(d) plan and remain subject to any applicable regulatory
requirements in the plan, in addition to being subject to
regulatory requirements under CAA section 111(b).23
Similarly, in the proposed Section 111(d) guidelines, EPA
restates this proposal, reiterating that the modified and
reconstructed unit would be subject to Section 111(d) and Section
111(b) requirements simultaneously:
The EPA is proposing that an existing source that becomes
subject to requirements under CAA section 111(d) will continue to
be subject to those requirements even after it undertakes a
modification or reconstruction. Under this interpretation, a
modified or reconstructed source would be subject to both (1) the
CAA section 111(d) requirements that it had previously been subject
to and (2) the modified source or reconstructed source standard
being promulgated under CAA section 111(b) simultaneously with this
rulemaking. It should be noted that this proposal applies to any
existing source subject to any CAA section 111(d)
22 Alternatively, EPA can address its proposed approach in the
preamble to the final rule and make the changes to rule language in
a separate rulemaking. For example, EPA recently announced that it
intended to change its PSD rules to facilitate the use of biomass
projects under the Proposed Rule. That rulemaking could be expanded
to fully implement any NSR remedy that cannot be fully implemented
as part of the Proposed Rule. 23 79 Fed. Reg. at 34,963/col. 1.
-
27
plan, and not only existing sources subject to the CAA section
111(d) plans promulgated under this rulemaking.24
While GE understands the initial appeal of this proposal for
purposes of ensuring continued emission reductions from existing
units, EPA’s proposal creates significant legal risk under the CAA.
Section 111(a) is clear in defining a “new source” as “any
stationary source, the construction or modification of which is
commenced after the publication of regulations (or, if earlier,
proposed regulations) prescribing a standard of performance under
this section which will be applicable to such source.” Section
111(b)(1)(B) further requires EPA to establish separate standards
for new sources within one year of listing the source category:
“Within one year after the inclusion of a category of stationary
sources in a list under subparagraph (A), the Administrator shall
publish proposed regulations, establishing Federal standards of
performance for new sources within such category.”
In contrast, existing sources are clearly regulated under
Section 111(d), which requires EPA to establish “standards of
performance for any existing source for any air pollutant” for
which air quality criteria have not been issued or which is not
included on a list published under Section 108 of the CAA or
emitted from a source category which is regulated under Section 112
of the CAA, but “to which a standard of performance under this
section would apply if such existing source were a new source.” EPA
cannot regulate the same sources under both provisions. The statute
is clear – either a source is “new” subject to standards under
Section 111(b), or a source is an “existing” source subject to
standards under Section 111(d). This separation is reinforced by
the different criteria guiding standard setting for new versus
existing sources in each provision. EPA ignores the plain meaning
and structure of section 111 by attempting to subject sources to
both existing and new standards at the same time.
Instead of proposing dual, simultaneous regulation of modified
and reconstructed units that would increase legal risk for the
rule, EPA should simply clarify that any physical or operational
change undertaken to improve efficiency or comply with a state’s
Section 111(d) plan represents a pollution control project and is
thus excluded from triggering modification. This would greatly
simplify the program and largely address the problem of dual
regulation.
While existing sources traditionally have not triggered
reconstruction as often as modification, GE anticipates that many
existing sources will opt to repower rather than build a new
greenfield plant. GE’s October 16, 2014 comments submitted in
response to
24 79 Fed. Reg. at 34,903/col. 3
-
28
EPA’s proposed rule for modified and reconstructed sources
recommends that the Agency finalize standards that are reflective
of what repowered sources can achieve.
If EPA insists on proceeding with the more legally vulnerable
approach of keeping reconstructed sources in Section 111(d), the
Agency could simplify compliance by establishing that a
reconstructed source’s compliance with a state’s 111(d) plan also
satisfies Section 111(b) standards for reconstructed units.
Alternatively, as noted above in the PSD/NNSR discussion, EPA has
similar discretion not to impose NSPS for modified or reconstructed
units. In fact, EPA has previously decided not to establish certain
NSPS for modified or reconstructed units in specific source
categories, including Asphalt Processing and Asphalt Roofing
Manufacture (which exempts modified sources from the visible
emission standard 40 C.F.R. § 60.472(a)(3)) and Synthetic Fiber
Production Facilities (that does not apply to modified facilities
40 C.F. R. § 60.600(c)). Deciding not to impose standards on these
units would prevent dual regulation.
b. While states have the discretion to use new NGCC as an
emission mitigation strategy, a decision by EPA to include new
sources as part of a BSER determination under Section 111(d) would
create significant legal risk.
While the Agency’s proposed guidelines do not include new NGCC
capacity as part of its BSER determination, EPA requests comment on
whether the Agency should consider new NGCC capacity as part of the
BSER and on ways to definite appropriate state-level goals:
While the EPA is not proposing that new NGCC capacity is part of
the basis supporting the BSER, we recognize that there are a number
of new NGCC units being proposed and that many modeling efforts
suggest that development of new NGCC capacity would likely be used
as a CO 2 emission mitigation strategy. Therefore, we invite
comment on whether we should consider construction and use of new
NGCC capacity as part of the basis supporting the BSER. Further, we
take comment on ways to define appropriate state-level goals based
on consideration of new NGCC capacity.25
Later in the proposed guidelines, EPA builds on this request for
comments by seeking additional input from the public on how the
emission changes from substitution of existing generation by new
NGCC should be calculated:
The agency requests comment on how emissions changes under a
rate-based plan resulting from substitution of generation by new
NGCC for
25 p 34,877/col 1
-
29
generation by affected EGUs should be calculated toward a
required emission performance level for affected EGUs.
Specifically, considering the legal structure of CAA section
111(d), should the calculation consider only the emission
reductions at affected EGUs, or should the calculation also
consider the new emissions added by the new NGCC unit, which is not
an affected unit under section 111(d)? Should the emissions from a
new NGCC included as an enforceable measure in a mass-based state
plan (e.g., in a plan using a portfolio approach) also be
considered?26
These are both important issues. As discussed above, Section 111
establishes separate regulatory programs for new units (including
modified and reconstructed units) under Section 111(b), and for
existing units under Section 111(d). In addition to being more
legally defensible, excluding new units from Section 111(d) plans
would also create an incentive to modernize and build new NGCC
capacity that can attain even higher efficiency levels than is
achievable with existing units. Requiring states to include new
units in existing state plans would increase the legal uncertainty
surrounding the rule and discourage investment in new, more
efficient capacity.
While the CAA is clear that newly constructed EGUs that commence
construction after January 8, 2014 (the date of publication of the
proposed NSPS rule for new EGUs under section 111(b)) are regulated
under subsection (b), EPA correctly recognizes that states may use
new NGCC capacity as a CO2 emission mitigation strategy. This
raises important questions regarding how emissions should be
calculated under a rate-base or mass-based approach.
If a state employs a rate-based approach under section 111(d),
the most appropriate approach would be to allow the megawatt hours
generated by these newly constructed units to be included in the
denominator for a state’s rate. This approach would be similar to
how renewable energy and new nuclear units are treated under EPA’s
Proposed Guidelines under section 111(d). For states that use the
mass-based approach under section 111(d), the emissions from newly
constructed units should not be included in the program at all.
V. Accounting for CHP
Combined Heat and Power (CHP) units are capable of providing
overall efficiencies of 60 to 80 percent, leading to improved
environmental performance, reduced energy
26 34,924, col.1
-
30
consumption, and improved grid reliability compared with
conventional generation. CHP creates less pollution per unit of
energy by using the energy potential of fuel inputs twice or three
times, yielding half to a third of the emissions that would
otherwise result from separate applications.
GE urges EPA to ensure that CHP facilities are not inadvertently
subject to regulation under this rule as “affected Electricity
Generating Units.” No existing CHP facilities should be subject to
regulation under the Clean Power Plan. We suggest excluding highly
efficient CHP units that provide greater than 65 percent overall
efficiency.
Because CHP units put waste heat to productive use, they can
increase a facility’s net energy output without a change to
emissions. We believe EPA should incentivize these highly efficient
units by allowing facilities to include 100 percent of useful
thermal output from CHP when calculating emission compliance. Local
applications of CHP also avoid transmission and distribution line
losses, which should be consistent with average line losses for the
state in which a particular CHP unit is located.
GE supports the Combined Heat and Power Association (CHPA)
comments to EPA, and we join with them in encouraging EPA to
recognize the benefits of encouraging the expansion and deployment
of CHP facilities to the greatest extent practicable.
VI. Fuel cells
EPA should include all-electric fuel cells as a compliance
option under this rule
EPA should specifically list all-electric fuel cells as
potential compliance options for states to use as they reduce their
overall carbon intensity. All-electric fuel cells are deployable as
demand-side resources to provide year-round power using natural gas
or biogas at efficiencies much higher than those of centralized
power plants. Additionally, fuel cells are typically situated close
to loads, where they avoid the transmission and distribution line
losses associated with centralized generation.
Twenty-two states, two territories, and the District of Columbia
already recognize the importance of fuel cells in supporting a
lower-carbon future. The California Air Resources Board (CARB) in
2007 qualified fuel cells as “ultra-clean” technology under their
existing standard. The New York Public Service Commission qualifies
them under their Renewable Portfolio Standard.
-
31
Comments on EPA Building Blocks
I. Building Block 1: Efficiency improvements
a. EPA’s estimate of cost effective upgrades across the coal
fleet is unrealistic
In building block 1, EPA suggests that a net heat rate
improvement (HRI) of 6% can be achieved in the existing coal-fired
steam boiler fleet as a means to reduce C02 emissions.27 According
to EPA, this 6% is achievable through implementing best practices
in operations and maintenance of these facilities (4% net) and
through installation of equipment upgrades (2% net).
We agree there is an untapped potential in the fossil power
sector to deploy upgrades and improve operations. For the purposes
of our comments, and in general, we consider heat rate improvements
from equipment and from operational improvements as part of one
category, and our estimates reflect the total impact of both
“types” of heat rate improvement.
GE is a major provider of equipment and services to the
coal-fired power sector. In fact, around half of the existing fleet
runs on GE technology, and over three-quarters of that subset have
already deployed upgraded heat rate technology. These upgrades not
only improve efficiency and lower CO2 emissions, they also drive
better economic performance of units and plants. This potential is
restrained by costs and the regulatory burdens triggered by
implementing certain measures (e.g. New Source Review).
GE analysis shows a gross 6% heat rate increase across the
existing fleet is perhaps technically possible, but only achievable
at an unreasonable cost. GE estimates the total economic potential
for heat rate improvements in the existing coal fleet to instead be
as high as ~4% gross, through a combination of operational and
equipment improvements. Because of anticipated market conditions
and potential regulatory barriers to implementation, most of these
improvements are likely to happen in a 42GW subset of the existing
fleet. As a result , we expect to see potential deployment of HRIs
to be closer to a 1% gross, on average, across the fleet.
27 We use the term net with regards to EPA’s 6 percent reduction
because environmental retrofits in response to other EPA
regulations, such as CSAPR and MATS, create a parasitic load on
generating plants that increase heat rates. Any heat rate
improvement target must first offset these increases in heat rate
from 2012 before they achieve the true net reduction in heat rate
called for by EPA. According to our estimate, the impact on heat
rate for the parasitic load is about 0.5% across the fleet from
2012. As we do not have individual plant data, parasitic load at
any individual plant could be higher or lower. Our estimates of
heat rate improvement potentials below are stated as “gross” terms
as they represent the gain from the heat rate projects and do not
offset parasitic load. For comparison purposes to EPA’s estimate,
our numbers should be reduced about 0.5%.
-
32
We have included a more in-depth discussion about how we arrived
at these conclusions in Appendix A. Effective NSR reform is
critical to unlock more potential from existing coal-fired power
plants.
We believe EPA should consider several important issues when
calculating the impact of heat rate improvements as part of the
Agency’s BSER determination.
b. EPA should consider operator experience, conditions, and the
availability of new technologies when calculating the impact of
upgrades
Advanced technologies to improve heat rate in coal plants have
been developed since the 2009 S&L study and should be
considered by EPA in its BSER determination
Advanced heat rate improvement strategies leverage technological
improvements from new units. New steam turbine units use advanced
3D airfoil and blade technology, enhanced sealing technology to
reduce leakages, and improved moisture removal from low pressure
turbines. Our experience with new units indicates that these
advanced technologies could offer up to an additional 2.0% gross
heat rate improvement, where feasibl