Why NERC thinks the EPA Clean Power Plan could threaten grid reliability
There are 'physical limitations' on how fast states can build new infrastructure, analysts say
For many observers, the saga over the EPA's proposed Clean Power Plan is getting more and more like the first human trip to Mars: Nobody’s gone yet, but half the critics think it’s not worth going and the other half think they know a better way to get there.
15 states and a coalition of fossil fuel companies are suing the EPA because they believe the Obama administration's centerpiece carbon regulation, which aims to cut U.S. greenhouse gas emissions 30% from 2005 levels by 2030, is an illegal overreach of EPA's authority. Dozens of additional energy stakeholders, from states to utilities and researchers, have submitted comments critical of the plan's proposed timeline, which would require states to begin to significantly cut carbon emissions by 2020.
Now, the North American Electricity Reliability Corporation (NERC), the federal non-profit charged with overseeing the reliable delivery of power, has released a report saying the CCP’s proposed timeline may indeed be problematic, and will likely need to be altered to allow enough time to build the necessary generation and transmission to comply with the regulations.
The large-scale deployment and use of renewables and natural gas generation represents a significant change in the power flow, both in direction and magnitude, for the bulk power system (BPS) and could be “a significant planning and operational challenge," concludes NERC’s report, "Potential Reliability Impacts of EPA’s Proposed Clean Power Plan, Phase 1."
"Sufficient time and coordination is needed to determine region specific solutions,” the report, which was released last week, reads.
“The generation mix is undergoing a transformative change and the Clean Power Plan is expected to accelerate it,” said Tom Burgess, vice president of reliability assessment and performance analysis at NERC, in a press call. “The goal of this assessment was to identify threats to reliability and make recommendations to ease the transition and reduce risks.”
“You will need a lot of work and a lot of construction to get to the CO2 emission cuts the EPA is proposing, and that is going to take a lot of time and coordination,” added NERC Director of Reliability Assessment John Moura. “This study paves the way for that discussion.”
Methods, findings and recommendations
NERC's November 2014 Initial Reliability Review examined the CPP’s approach and took a high-level look at potential reliability risks. Now, this new new three-part reliability assessment goes deeper, analyzing five potential market scenarios, estimating transmission and generation infrastructure adequacy, and reviewing available studies on the CPP.
The five scenarios begin with a business-as-usual base case. In the second scenario, compliance plans are limited to individual states. A third scenario incorporates regional compliance implementations. A fourth projects the future to 2030 where there is no CPP and natural gas prices are low. In the last scenario, there is a CPP but no regional compliance and natural gas prices remain low.
The generation and transmission adequacy analyses assess how much new generation, electric transmission, and natural gas pipeline capacity would be needed to meet EPA’s 2030 goals and how long it would likely take to build them.
The review of studies examines how NERC reliability authorities, including local and regional transmission planners and operators, have evaluated potential impacts of the CPP. It identifies cumulative region-wide or interconnection-wide impacts.
The report’s four key findings:
- The CPP will accelerate the U.S. BPS resource mix transition to natural gas and renewables and “transform grid-level reliability services, diversity, and flexibility,”
- The BPS industry will need more time to plan for this transition and build the infrastructure to support it than is currently allowed by the proposed regulations,
- Coal-sourced generation will diminish and serve mainly seasonal peaking purposes, significantly altering its economic viability,
- Much more transmission and natural gas pipeline infrastructure will be needed to support new generation needs.
Some states, such as California and those in the Northeast’s Regional Greenhouse Gas Initiative, won’t need extra time to meet the initial 2020 compliance target, but others will, Moura said.
“We are not prescribing a specific date, but we do think 2020 is too soon for many states," he said. "Only when the states start developing their plans and coordinating with other states will they know what they need to do and how long it will take. That might not be until 2023 or 2025. They need to take advantage of EPA’s flexibility to work those things out.”
The accelerated changes in the generation mix indicate “a significant reliability concern,” Moura said. ”Around 2019 to 2020, there is a cliff where many existing coal units reduce their capacity factors from around 70% to 80% down to around 30%.” That is produced by a 7.8 GW increase in wind, a 4.7 GW increase in solar, and a 45.9 GW increase in natural gas by 2020.
"But resources can only deliver their power if the transmission is there,” Moura said.
The report assesses power flow cases using western interconnection and eastern interconnection models with already-existing transmission upgrades and focuses only on the transmission backbone of 200 kV and above lines, Moura said. The transmission adequacy modeling only called for new lines where “reactive power health” was threatened.
“When we retire coal generation and replace it with wind generation, there are not the same kinds of essential services,” Moura said. “Reactive power and system voltages will need to be propped up to maintain system reliability.”
Some 7,000 miles of new lines will be necessary and about 1,000 miles of that will need to be 500 kV lines, NERC concludes. “And only a couple of entities have experience building that, so it is another layer of uncertainty in building the transmission backbone needed to support the new resource mix.”
It is likely to take, from inception to powering up, an average of 64 months to build a new combined cycle gas plant, and 37 months to build a new, under-100 MW solar power plant, the report concludes. A new urban 20 mile, 115 kV transmission line comes in at an estimated 64 months, a rural 230 kV line at 87 months, and a 500 kV line with new rights of way at 15 years.
Those estimated construction schedules, taken from NERC observation of past projects, likely mean that the CPP's timeframe — which would have states make significant emissions cuts in anticipation of the beginning of the compliance period in 2020 — likely needs to be extended. There are "physical limitations" on how fast the states can build new energy infrastructure that need to be taken into account, the analysts said.
The report makes five recommendations as we near the final plan's release this summer:
- NERC’s reliability assessments should continue
- They should incorporate state and regional planning groups’ assessments
- Deployment of infrastructure may take more time than the CPP allots
- The CPP should have a formal “reliability assurance mechanism”
- Reliability authorities should be consulted as states and regions develop CPP compliance plans
Reactions to the NERC study
National Rural Electric Cooperative Association (NRECA) CEO Jo Ann Emerson, whose organization members serve 12% of U.S. electricity customers, endorsed the questions NERC raised.
“This study from the regulatory authority charged with ensuring reliability bolsters arguments made by electric co-ops and others that the EPA’s interim deadlines are, quite simply, not workable,” she wrote in a statement.
EPA projects co-ops will have to shutter at least 21% of their coal generation fleet by 2025, but the NERC report found replacement generation and transmission will be complete no sooner than 2031, she explained.
“[W]e wish EPA had not given reliability such short shrift,” she wrote.
Malcolm Woolf, vice president of policy at the renewables advocacy group Advanced Energy Economy, disagreed.
“The NERC assessment tells us very little about how to move forward other than to slow down,” he wrote in a statement on the report. The four flaws in NERC’s report, Woolf said, are:
- It overlooks many of the operational tools that grid operators have developed, and are continuing to develop, to manage a grid that NERC recognizes is already changing in line with CPP requirements,
- It excludes technologies like demand response and puts artificial constraints on technologies like renewables and energy efficiency,
- It imposes annual emission caps in place of EPA’s interim targets but states have the flexibility to spread emissions reductions over the 10-year compliance period
- It imposes assumptions before the rule is finalized.
“With all the options available to them for compliance with EPA’s final rule, states, working with their utilities, will be able to modernize their electric power systems,” Woolf wrote.
John Moore, Senior Attorney at the Natural Resources Defense Council, largely agreed with Woolf.
The NERC report makes “faulty assumptions” and comes with “significant caveats,” he wrote in a blog post on the report. But it still concludes “reliability can be maintained — in other words, no blackouts.”
One bad assumption, according to Moore, is that energy efficiency will grow 0.5% per year while the BPS is presently adding efficiency at 1.5% per year. Another bad assumption is that solar will grow between 12 GW and 20 GW by 2030 whereas industry data projects a 20 GW expansion by the end of 2016.
NERC also assumes 2 GW of new wind per year through 2030, but the U.S. Department of Energy projects at least 4 GW per year through 2020 and 3 GW per year from 2020 to 2030. And a recent industry forecast reported a “wind boom” that could add 18 GW by the end of 2016.
The NERC report also assumes only 11 GW to 12 GW of new power plant capacity annually, but the U.S. grid has added 30 GW of new capacity yearly since 2000, Moore wrote. And the report's over-reliance on natural gas necessitates unnecessarily big assumptions about pipeline needs.
With the Clean Power Plan's flexible design and the many tools available to grid operators to manage reliability issues, there is no reason states need a compliance waiver or delay, Moore wrote.
“The final Clean Power Plan will grant states and utilities sufficient time and the flexibility they need to design their own, individual plans.”
Renewables and reliability
A big open question in the quest for a reliable grid under the Clean Power Plan remains if and how utilities will adapt to cope with increased renewable and distributed energy on their grids, especially as large baseload generation plants that usually provide backup for variable generation retire.
The BPS industry has learned much about how to use renewables, Burgess acknowledged. NERC recently worked with the California Independent System Operator (CAISO) to better understand how higher renewables penetrations can reliably be accommodated, he said.
A forthcoming NERC report will detail how a system in which large synchronous generators are no longer providing “essential reliability services” like ramping capability, voltage follow through, and frequency response can be reliably maintained.
The CAISO’s flexible ramping product makes an ancillary services market practical, Moura explained. On a system with shorter dispatch intervals, more accurate capacity allocation, more accurate forecasting, improved communications and control available to grid operators, more capable inverters for wind and solar, and cost effective storage, Moura said, it is possible that “variability doesn’t affect reliability.”
At this point, there remain more open questions than answers on both renewables use and general system reliability under the Clean Power Plan, the NERC analysts acknowledged, because the regulation is not yet finalized. As with the planned human trip to Mars, we likely won't be able to grasp the intricacies and challenges of the landmark regulation until its launched in mid-summer.