By Michael O’Boyle, Sonia Aggarwal, Eric Gimon, and the experts of America’s Power Plan
In recent statements, the North American Electric Reliability Corporation (NERC), the Southwest Power Pool (SPP), the Midwest Independent System Operator (MISO), and the Electric Reliability Council of Texas (ERCOT) raised concerns about electric reliability as states implement the EPA’s Clean Power Plan (CPP). At the same time, recent studies by PJM Interconnection (PJM), the state of Minnesota (in collaboration with MISO), and California’s utilities have found that integrating higher-than-required shares of renewables (30-50%) is possible without affecting reliability. What’s causing these well-established reliability authorities to reach such different conclusions, and is more analysis needed?
The NERC Reliability Report
First, it helps to understand the reasons for concern. Essentially, Clean Power Plan reliability questions hinge on whether there will be adequate resources available to meet demand, and whether there will be adequate flexibility and reliability services with higher shares of variable renewables. With regard to resource adequacy, NERC is concerned with potential shortfalls that could arise from rapid coal retirement mixed with slow approval and construction times for transmission, pipelines, coal to gas fuel conversion, and new generation. Historically, planning and construction of new plants and transmission have required five to ten years. With state implementation plans due in 2016–2018, and compliance periods beginning in 2020, NERC emphasizes that the timeline for replacing retired coal-fired generation with new infrastructure and generation is shorter than historical trends indicate is necessary.
NERC also suggests that EPA may have overestimated the potential from fuel switching, heat rate improvement, and energy efficiency to cut carbon emissions by 2020, which would mean that states may need to use more natural gas and variable renewables to comply. NERC concludes that because renewables demand more flexibility and provide fewer essential reliability services than fossil-fueled generators, the combined strain of losing these generators and adding variable generation could leave a deficit in grid reliability services.
These are legitimate concerns, and these bodies are playing an important role in assessing future reliability given current policy and future trajectories. But experience in several parts of the country—as well as the detailed analyses conducted by PJM, Minnesota, and California—suggest that reliability concerns can be mitigated via institutional leadership, careful planning, and intelligent grid operations, particularly at EPA’s projected levels of renewable penetration. Proactive states and grid regions can begin by gathering information and stakeholder input for planning studies now, with the intention to conduct more detailed reliability analyses as soon as the rule is finalized next year.
Pathways Exist to Address Reliability Concerns
To help meet new infrastructure demands, Planning for and Investing in Wires and Finding a Home for Renewable Energy and Transmission provide specific recommendations for how to maximize the value of existing infrastructure while expediting the process for new generation and transmission. Effective approaches require engaging stakeholders early, enhancing coordination among regulatory bodies, employing smart strategies to avoid the risk of environmental and cultural resource conflicts, and taking better advantage of existing infrastructure to reduce costs.
As for the growing need for flexibility and other reliability services, several recent studies suggest the grid can remain reliable with higher shares of variable renewables than would be required under the Clean Power Plan.
For example, a February PJM report, which looks at whether flexibility constraints exist at 30% renewables in 2026, found that plenty of flexibility exists across PJM Interconnection in a system-wide, hourly model supported by hundreds of sub-hourly market operation simulations using actual PJM ramping capability and examining transmission adequacy. Additionally, the analysis found that—in aggregate—renewable energy availability matched well with times of higher demand. Similarly, an October report commissioned by the Minnesota Department of Commerce (in coordination with MISO) finds that 40% wind and solar in Minnesota (15% renewable energy for MISO) in 2030 would successfully support system operation “for all hours of the year with no unserved load, no reserve violations, and minimal curtailment of renewable energy.” The Minnesota study further examined voltage regulation and system stability, finding no problems. A similar January study commissioned by California’s utilities concluded that there would be no flexibility constraints with 50% renewables in 2030.
While these studies establish that flexibility is not a bottleneck for reliability at high shares of variable generation, the availability of all essential reliability services in the case of major contingency events is much more difficult to ascertain. There are at least four options that grid operators can call upon in the coming years to add flexibility and reliability services: expanding balancing areas, demand response, storage (on a steady price decline), and well-designed power markets that enable all resources to compete to deliver flexibility and reliability services.
In addition to these four sources, renewable generation can also provide reliability services like frequency response and system inertia. For example, GE’s DOE-sponsored study, presented at the Western Interstate Energy Board’s System Flexibility Forum, found that advanced control capabilities on solar and wind generators can provide additional frequency response to supplement traditional sources in the Western Interconnection. Active power controls are already being used on wind turbines to provide reliability services in Quebec and Texas. Much work needs to be done to facilitate product development, operationalization, and compensation for these services, but plenty of low- or zero-carbon sources have demonstrated the ability to provide the essential reliability services at low cost and in time to support CPP compliance.
NERC’s view (still mostly based on concerns, not hard analysis of specific future situations) is that institutional inertia, planning processes, and long periods required for infrastructure build-out may be too slow to replace retiring reliability resources, but these time concerns do not restrict demand response, storage, or (depending on need for new siting) renewable energy. ERCOT’s analysis reasonably points out that market reforms (additional compensation for more precisely defined ramping and regulation services), more demand response, and fast-ramping capacity could support the transition to low-carbon resources without relying on time-intensive planning processes.
Renewable Integration is Moving Quickly on the Ground
Renewable integration on the ground supports this conclusion: if renewable integration continues on pace, EPA’s assumptions for renewable generation are very likely to be achieved and surpassed ahead of schedule. Renewable deployment is moving more quickly than many analyses indicate, and the lights are still on.
In particular, NERC cast doubts on EPA’s assumption that the 214 terawatt-hours (TWh) of renewables on-line in 2012 can grow to 281 TWh in 2020, yet the actual 2013 non-hydro renewable generation number (253 TWh according to the latest Energy Information Administration figures) indicates that we were almost 60% of the way to the 2020 goal only one year past the baseline. Current monthly 2014 generation data from EIA show non-hydro renewables likely to meet EPA’s 2020 goal by 2015, having attained an annual average growth of 14% each year over the past five years. Assuming a slower growth rate of 10% per year, non-hydro renewable generation is on course to surpass the 2030 goal by 2021. Even states with no renewables requirements, like Georgia, are seeing new supply being contracted from wind and solar providers on a cost basis alone.
NERC makes some important points about reliability hinging on all the details of where specific plants are retired, efficiency achieved, infrastructure projects completed, and new capacity added. Additional renewable energy will require more integration and transmission, but existing planning processes, under the aegis of FERC Order 1000 and existing NERC rules, should already be gearing up for this—and need not wait on CPP State Implementation Plans (SIPs). It will be important for EPA, state regulators, and all stakeholders to look at the detailed analyses that reliability engineers eventually produce after SIPs are submitted, and to work early and decisively to plan and build transmission, pipelines, and other assets necessary to make the CPP a success. Equally important, though, will be attention to the lessons offered by real-world deployment rates and costs, commercially available flexibility and reliability services, and important institutional reforms that can deliver on both environmental and reliability goals, while delivering net benefits to American families.
On December 10, FERC announced that they will hold four technical conferences to delve into implementation and reliability concerns tied to U.S. EPA’s proposal to cut carbon emissions beginning on February 19.
Thank you to Greg Brinkman, Jim Caldwell, Gene Hinkle, and Brian Parsons for their input on this piece. The authors are responsible for its final content.