The recent EPA Good Neighbor Plan Action contains many implications for electric generating units and other processes in the industrial sector. Watch our summary in the first video, see the ways we can help units thrive in the wake of the action in the second video, and read below for additional details about the action.
The U.S. EPA recently released an action finalizing Federal Implementation Plan (FIP) requirements to address 23 states’ obligations to eliminate significant contribution to nonattainment, or interference with maintenance, of the 2015 ozone National Ambient Air Quality Standards (NAAQS) in other states, under the “good neighbor” or “interstate transport” provision of the Clean Air Act (CAA).
The action document itself comprises 935 pages of mostly small font text with few graphics. We at Taber have reviewed this action and attempted to break it down into its most important implications for the affected industries, and to provide a pathway for operating reliably in the action’s wake. We’ve started by describing the impact on Electric Generating Units (EGUs), and will soon follow-up with a similar summary for the non-EGU industries affected. Stay tuned!
Legal challenges to the action are anticipated, but as it stands now, the implications will go into effect for many of the affected states on May 1, 2023. Now, our condensed and summarized version of the plan, followed by several important facts and considerations.
Revised Group 3 Trading Program
Through various modeling and data analysis, the EPA identified that ozone-emitting sources within 23 states are contributing more than 1% of the NAAQS ozone threshold amount to states that are “downwind” of them (the primary combustion product contributing to ozone formation is nitrogen oxides, or NOx, which is the subject of regulation within the action). In order to be “good neighbors,” these states must lower their NOx emission rates to make it possible for their downwind neighbors to realize satisfactory air quality (e.g., reduce ozone). These 23 states will be required to participate in a revised version of the Cross-State Air Pollution Rule (CSAPR) NOx Ozone Season Group 3 Trading Program. The state map below available from the EPA shows the 23 identified states, and their findings for interstate transport of these emissions. One of these state (California) does not have any EGU’s which are affected by this action, so of the remaining 22 states, 12 currently participate in the Group 3 Trading Program (Illinois, Indiana, Kentucky, Louisiana, Maryland, Michigan, New Jersey, New York, Ohio, Pennsylvania, Virginia, and West Virginia), 7 currently participate in the CSAPR Group 2 Trading Program (Alabama, Arkansas, Mississippi, Missouri, Oklahoma, Texas, and Wisconsin) and 3 do not currently belong to any CSAPR program (Minnesota, Nevada, and Utah). All of these states will be transitioned to the revised Group 3 Trading Program and begin participating at the start of ozone season 2023 (May 1) except for the new states, Minnesota, Nevada, and Utah. These states will not be required to enter the program until after the effective date of the rule, if after May 1.
The EPA intends to “expeditiously review” the updated air quality modeling and related analyses to assess potential good neighbor requirements for six additional states – Arizona, Iowa, Kansas, New Mexico, Tennessee, and Wyoming – in a subsequent action. At that time these, and any other states found to contribute to downwind non-attainment, might be ruled on in some fashion. Other states, such as Oregon and Delaware, have had other deferrals or decisions made concerning them.
Rather than impose direct NOx emission rate limits on each of the identified states, the EPA has defined yearly NOx emissions “budgets” in total tons for each state, and is revising the existing CSAPR NOx Ozone Season Group 3 Trading Program a NOx allowance (allowed emitted tons) market, similar to allowances used in other programs. This group of states belong to “Group 3” and will participate in the “Group 3 NOx Allowance Trading Program.” The EGUs within these states will be able to trade allowable emission rates amongst one-another (as well as with non-EGU sources and non-emitters – brokers), such that a site that overperforms or expects to overperform would be able to sell their own unused allowances to a site that is struggling to stay within the budget at the current market price.
Budget Determination Assumptions
The year-to-year budgets were determined based on a handful of assumptions generated from available information across the industry.
- Coal-fired generators will be able to fully install Selective Non-Catalytic Reduction (SNCR) systems for post-combustion NOx treatment by the start of ozone season (May 1) 2026, and be well on their way to fully installing Selective Catalytic Reduction (SCR) systems which will be fully implemented by the start of ozone season 2027.
- Expected unit retirements and retrofits will affect the profile of EGUs during the 2023 – 2029 period and the necessary heat input to generate a given amount of electricity.
- Optimized coal-fired units >= 100 MW capacity with state-of-the-art controls without post-combustion control emit on average 0.199 lb/MMBtu
- Optimized coal-fired units >= 100 MW capacity with existing SCR control average NOx emission rates of 0.08 lb/MMBtu
- Optimized oil- and gas-fired steam units and simple cycle units with existing SCR control average 0.03 lb/MMBtu
- Optimized combined cycle units with existing SCR control average 0.012 lb/MMBtu
- Retrofitting a coal unit with a new SCR implementation and optimizing control will realize an average NOx emission rate of 0.05 lb/MMBtu
Dynamic Budget Determination
From these assumptions, the identified mitigation strategies expected effect on NOx emission rates are applied to each EGU according to their historical heat input in millions of BTUs. For the then-assumed current state of the unit, the highest three of the past five years of heat input are averaged for each unit, then summed to represent the state’s total heat input. Each unit is assigned its contribution percentage to this total. Then, the past three years of total state heat input are averaged, and each unit’s “normalized” heat input is equal to the state three year heat input average multiplied by the unit’s contribution percentage. The normalized heat input is multiplied by the unit’s expected emission rate condition, and the total of these creates the minimum level in determining the state’s NOx allowance budget. Some adjustments and reapportionment from unit to unit are made at this point and adjustment for past year’s performance, not going lower than the dynamic calculation result. A brief example of this process is available in our explanation video above.
Take this highest three-year average and multiply by the allowable NOx emission rate in lb/MMBtu for the applicable conditions (optimized start of the art controls, or optimized SCR – the emission rates of SNCR systems are not used in determining budgets as the EPA states is known they are not as effective as an SCR system), and this should produce the allowed total mass emissions for ozone season.
This is effectively the process performed using the available data the EPA shares to determine the yearly ozone season NOx budgets shown in the table provided by the EPA for the 2023 – 2029 ozone seasons (recreated below, the 2023 totals are not prorated to accommodate later entry of the 3 new states). As time goes on, recent data for the same time ranges (previous 5 years) will be collected and used for each year’s dynamic budget calculation. This is intended to capture unit retirements, retrofits, and other grid changes that weren’t planned for originally. At the unit level, there are a number of reapportionments after the contribution percentage is used to distribute the state budget, but the total for each state remains the same after reapportionment. These budgets include all units within a state’s border’s – even those in Indian country not subject to a state’s CAA implementation planning authority. Of these totals, 5% – 9% are set-aside for expected new units.
|2023 State Budget
|2024 State Budget
|2025 State Budget
|2026 State Budget*
|2027 State Budget*
|2028 State Budget*
|2029 State Budget*
* Budgets for these years are subject to results of dynamic budgeting calculation, and will be the higher of the dynamic budget and these preset levels.
The EPA has built into the budgeting method the ability for recalibration which they refer to as “Dynamic Budgeting”, which is essentially the ability for the budgets to adapt to changes in the generation profile which were not anticipated (e.g., early unit retirements or retrofits) as well as to variations in heat input to satisfy electrical grid demands (e.g. grid stability requires that certain units operate at high generation levels to satisfy power demand – the EPA specifically mentions higher adoption levels of electric vehicles).
For ozone seasons 2023 – 2025, the preset budgets that have been established will be adhered to. For ozone seasons 2026 – 2029, the higher of the preset budgets and the dynamic budgeting results will be used. This is done to provide greater certainty for utility’s planning efforts without under cutting those plans (budget can only be raised from preset levels). For ozone seasons 2030 and beyond, only the dynamic budgeting approach will be used. Each subsequent year’s budget will be announced at the beginning of the ozone season of the preceding year, providing 12 months of preparation time for the next ozone season.
Margin for Units Before Penalties & Unit Backstop Limits
There is some margin afforded to individual units (pg 394). Each unit will be afforded a 50-ton threshold before increased surrender requirements apply, which the EPA believes is sufficient to address all instances where a unit’s emissions would exceed a daily backstop rate of 0.14 lb/MMBtu during periods such as start-up, shut-down or when the SCR cannot be operated.
Similarly to the bank, a variability limit of 21% is established for each state, providing further margin before certain violations are in effect. When dynamic budgeting results will be applicable (2026 and on) there is an additional adjustment made available for states who significantly increase their heat input relative to past years. If the state’s total heat input is more than 121% of the historical heat input used in determining the state’s allowance budget for that year, then the variability limit will be increased to match the proportion that the year’s heat input is greater than historical heat input. For example, if the 2025 season’s state heat input was 130% of the historical heat input used in determining the budget, the variability limit for that state will be increased from 21% to 30% for that season.
We mentioned a daily backstop rate. Beyond the budgets, large coal-fired EGUs (capacity >= 100 MW) with an SCR installed will be subject to a daily backstop NOx limit starting in 2024 for existing SCRs, or in the second control period following SCR installation. This backstop limit is a daily average of 0.14 lb/MMBtu. It will be applied in determining the number of allowances to be surrendered if the unit exceeds by more than 50 tons a daily average rate of 0.14 lb/MMBtu for the unit’s actual heat input day to day. The total amount beyond 0.14 x daily heat input + 50 tons during the ozone season will result in surrendering allowances sufficient to cover that value at a 2-to-1 rate in addition to the standard 1-to-1 (a total of 3-to-1 allowance surrender rate).
Starting with the 2030 ozone season, all operating large coal-fired units will be subject to the daily backstop limit, regardless of if they’ve completed installation of post-combustion equipment.
There is also a season-long backstop limit for units WITH post-combustion controls that are found to have contributed to their state’s exceedance of the assurance level (121% of the budget), that they must surrender at a 3-to-1 rate allowances for emissions equal to the higher of maintaining a seasonal average rate of 0.10 lb/MMBtu or 125% of the unit’s lowest pervious season average emissions rate under any CSAPR seasonal NOx trading program.
Banking of allowances is permissible within the Group 3 Trading Program. However, the “bank” will be recalibrated annually on August 1 such that the amount of allowances for the control period does not exceed 121% of the control period budget. In other words, the target bank size for each year is 121% of all allowances within the Group 3 program, so the sum of all state’s allowances. To further drive this home, the target bank size for 2024 will be 198,014 x 0.12 = 41,583. If the sum of all banked allowances across the country exceeds this amount, all accounts with banked allowances will be recalibrated. This is intended to help drive allowance trading to better redistribute allowances as well as promote overall NOx emission rate reduction.
Following an ozone season, compliance and allowance trading will not be finished until June 1 of the following year. However, the EPA will release a projection of the degree of recalibration expected by early November, just after the ozone season those allowances apply to. Shortly after June 1 finalization, on August 1, bank recalibration will occur. At that time any banked allowances over the 21% threshold will be removed from circulation.
Data on the amounts of allowances held will be publicly available on a website (pg. 462). All allowance trades, transfers, deductions, etc. must be properly recorded in the EPA’s Allowance Management System.
Carry-over of Existing CSAPR Program Allowances
To prime the bank for the 2023 ozone season, states coming from Group 2 to Group 3 will have their banked allowances recalibrated and available for the 2023 season (pg 552). On pg. 549, it’s stated that they expect 149,386 Group 2 allowances to be available, and that the conversion ratio will be 6.5-to-1 (6.5 Group2 to make 1 Group3)
Failure to timely surrender all required allowances is potentially subject to penalties under the CAA’s enforcement provisions. The CAA authorizes fines of up to $25,000 per day for emissions violations. Criminal sanctions include fines of individuals up to $250,000 and up to 15 years in prison. Corporations can be fined up to $1 million per incident for knowingly endangering people with emissions and up to $500,000 per incident for negligent emissions. There is also the potential for shutdown of the violator if the EPA does not directly regulate the source by establishing emissions limitations and a compliance schedule.
Allocation of allowances will follow the EPA’s method (proportional assignment based on historical heat input). However, each state is allowed to submit it’s own implementation plan (SIP) to describe an alternate allocation method, which the EPA must approve. If at the end of a control period those allowances that were set-aside for new units have not been allocated, they will be distributed among existing units.
What to do about this?
With this action going into effect in the very near future, what are affected EGU’s going to do? This is the point where Taber International hopes to be able to help EGU’s to ease the burden of continually monitoring performance and ensuring they remain within their allocated budgets, as well as to assist units in achieving optimal performance of their existing and soon-to-be-deployed combustion and post-combustion NOx treatment systems.
As was mentioned in the previous section, the EPA has recognized a number of mitigation strategies to improve NOx emissions, as these are what influenced their budgeting calculations and final state allowance totals. The six mitigation strategies that the EPA discusses in some detail are:
- Optimizing Existing SCRs
- Installing (and Optimizing) State-of-the-Art NOx Combustion Controls (low-NOx burners, overfire air, etc.)
- Optimizing Already Operating SNCRs or Turning on Idled Existing SNCRs
- Installing New SNCRs
- Installing New SCRs
- Generating Shifting
As can be seen, this effectively boils down to optimized combustion & post-combustion control and generation shifting. The EPA has made it very clear within this action that they not only want all units to have state-of-the-art controls (including post combustion control) but that they want these to be used consistently and as effectively as possible, which is accomplished through controls optimization.
Taber International was founded in 2006 for exactly this purpose: to provide electric generators with outstanding optimization and system-performance enhancing tools and services. Through our collaboration with Griffin Open Systems, LLC., we provide and customize a software platform to perform data analytics, process monitoring, real-time visualization, and closed-loop combustion and SCR/SNCR optimization to meet the unique needs of every individual process we work with. These systems have also been extended to explore multi-unit dispatch optimization, selecting individual units to operate at loads which most agree with the stated objective (or combination of objectives), which in this circumstance would be to minimize NOx emission rates while maintain unit and grid stability.
Real-time Monitoring and Estimation
The first step in operating within the latest Good Neighbor Plan action is going to be having a reliable, real-time understanding of where each unit and fleet of units are at relative to NOx allowances and heat input. Taber has developed a NOx Aggregator & Optimization application that is designed to track NOx emission rates and estimated heat input in real-time to show the user how each unit that they monitor is performing. This dashboard is also outfitted with the capability to interrogate the system with expected ongoing performance or unit capacity factors to predict how operating in those conditions will affect their NOx allowance and heat input figures at the close of ozone season. The system will be directly integrated into the available NOx allowance market data, so that in one location the current market price and available allowance will be easily viewable, as well as the value of expected unused allowances for the unit and/or fleet.
This information can be used to determine an effective cost of operation at various load ranges which can be provided directly to dispatching authorities to consider while dispatching units to ensure the units remain within budget and producing at the maximum possible capacity.
The dashboard system can be fully customized, such that the graphics and their mode of display match the styles and objectives of each individual customer and user. Notifications of certain events, daily summaries, or other analysis results can also be configured to be automatically distributed using whatever methods is most applicable (email, Microsoft Teams, etc.)
To take that a step further, the system can recommend the optimal dispatching levels for a fleet of generators based on input operating costs and projected NOx allowance costs and effects throughout ozone season, effectively accomplishing optimal generation shifting, and providing the results to dispatch to act upon.
Finally, Taber’s years of experience in combustion optimization have proven that this platform and approach, combining artificial intelligence and expert unit knowledge to develop a highly customized and adaptive combustion optimization system is decidedly effective in getting the most out of existing state-of-the-art equipment and minimizing NOx emission rates, while continuing to meet other operational objectives. Our average install realizes a 20% – 30% NOx reduction, with some of our best installs seeing sustained reduction in excess of 50%.
This same process of fine-tuned, high-resolution optimization is directly applicable to SCR and SNCR optimization, individually adjusting urea and ammonia flow rates to zones within the equipment that best matches with flue gas stratification to optimally reduce NOx while decreasing slip and usage rates, further reducing operating costs. When a post-combustion optimization system is coupled with a combustion optimization system, the two can be integrated to ensure the combustion portion of the process runs optimally with low mechanical NOx emissions, while realizing optimal post-combustion treatment temperatures to maximize the efficiency of these systems further enhancing NOx minimization achievements.
There are many implications of the latest Good Neighbor Plan action, and it will be important for EGUs to consider these as they prepare for this upcoming ozone season as well as seasons to come. Taking the needed steps now to be ready is sure to make the process easier and maintain thermal generators as important components of the electric grid and as a critical part of electric generating company’s portfolios.
We hope that this article has been informative and helpful to you. Please contact us using the form below to explore how we can assist you through system monitoring, optimization, or within any other aspect of your process and workflow!