by Roy Harvey and Kent Wittenburg on behalf of the FTG Technical Committee
The Future Grid Reliability Study was undertaken by ISO-NE to plan for a transformed grid in which the generation mix changes radically from a large dependence on fossil fuels to one that meets the goals of decarbonization of the NE states and is supplied in large part by renewables. The Future Grid Reliability Study Phase 1 (FGRS Phase 1) seeks to identify operational and reliability challenges and was released in stages starting in July 2022. Phase 2, expected in 2023, will be an economic study to determine whether the existing market structures will be sufficient to provide the resources needed to meet the challenges identified in Phase 1. Both are part of the ISO-NE Future Grid Initiative, and materials are available on the ISO-NE website.
Groups such as Fix the Grid that are advocating for a rapid and complete transition of the grid away from fossil fuels need to look closely at studies like these since the need for reliability, a fundamental mission of any grid authority such as ISO-NE, could be used to slow walk the green transition. It is much easier to manage reliability in a grid composed of dispatchable fossil fuel generation than in one that is made up of distributed, variable generation like wind and solar until long-term storage technologies mature. And yet we must acknowledge that the challenges of providing reliability with a decarbonized grid are real.
Study Summary
The scenario of primary interest (Scenario 3–”deep decarbonization”) analyzed in the study consisted of modeling a 2040 NE grid that meets the goal for the New England states to reach an emissions reduction of 80% from 1990 levels as well as meeting the requirements of a doubling of electricity demand by that date. Electricity demand is projected to double because of the transition to electric heat and electric transportation. The complexities of such a grid include meeting demand with variable and distributed generation (wind and solar) and managing battery storage. The most important question posed by this study is whether a deep carbonization scenario could meet reliability criteria and, if so, how. The requirements for reliability, which come from federal and regional regulatory bodies, include that outages (aside from weather-related) caused by deficient resources shall happen on average for no more than 2.4 hours per year. Resource adequacy tools were used in the study to model the NE grid system on a minute-by-minute basis under variable weather conditions to determine if generation could meet demand at all times.
Here were the most significant results:
- The base case models of Scenario 3 satisfied emissions reduction and increasing electricity demand expectations without considering variability due to weather.
Scenario 3 in this study modeled a generation mix that met the emission reduction goals while satisfying a doubling of demand through “nameplate” capacity. (“Nameplate” is an indication of ideal generation capacity, but renewables are affected by weather conditions.) The scenario included all planned generator retirements and also the retirement of all remaining coal, oil, and refuse-burning plants. Natural gas generation continued but amounted to roughly half of the average 2005-2019 generation. The amount of renewables grew significantly to a total of 45.6 GW and included 17 GWs of offshore wind, 28 GW of solar, and .6 GW of battery storage–the amount was 2.5X the amount of renewables assumed in the base scenario 0, which was a continuation of current growth trends and conservative demand increase forecasts. It included two-way flows through interconnections outside NE to “bank” energy. Aside from the pending 1200-MW Hydro Quebec tie-line, the model included an additional tie-in from NY and Quebec. - Alternatives were then explored to meet reliability requirements in the face of expected weather extremes and time-of-day limitations (on solar, in particular).
The study’s authors explored alternatives to try and satisfy the reliability requirements while at the same time minimizing cost and curtailment conditions, in which renewables are generating more energy than can be consumed or stored for future use. Most significantly:
- Scenario 3_P7 explored scaling renewables sufficient to meet reliability goals. It required doubling the amount of wind and solar over the base Scenario 3 case to 90 GW. There was also a relatively small amount of battery storage added. Analysis suggested that a large amount of generation curtailment would be expected at times when the renewables’ energy was not needed and storage was at capacity. Also, such a large increase in renewables would put an undue strain on the transmission system and require huge areas for wind and solar farms.
- The authors then examined the possibility of replacing part of this projected need for an increase of renewables with dispatchable (on-demand) power. They came to the conclusion that reliability could be satisfied in a scenario where 19% of the otherwise required renewables were replaced by the equivalent of 3% of dispatchable resources. Today’s dispatchable resources are almost all fossil-fuel based, but possible future alternatives include green hydrogen, small modular nuclear units, and co-located storage and renewables. This conclusion was considered to be one of the most significant findings of the study. However, note that the role of storage was largely ignored.
- The study’s authors acknowledged that the current set of modeling tools is inadequate to evaluate options for a grid of the future.
In the authors’ own words: “The FGRS Phase 1 is a turning point study for our region. Many existing long-term assumptions were called into question as part of this analysis, and results show that the methods by which the ISO and region at large evaluate future grids require an overhaul. The hypothetical future resources and demand mixes assumed by the FGRS are very different from today’s system and cannot be fully evaluated with our current tools or assumptions. The FGRS identifies and quantifies many reliability and operation challenges, transmission problems, and ancillary services gaps. Additionally, this analysis identifies areas where gaps of a future grid cannot as of yet be identified or quantified in sufficient detail to suggest a potential path forward.”
Observations from the Fix-the-Grid-Technical Group
The FTG-Technical group believes that studies such as this are important. However, they are only as good as the tools with which they are modeled and the validity of the assumptions that they make.
- A grid of the future will be heavily dependent on managing demand as well as large amounts of storage.
It is misleading at best and potentially harmful to come to conclusions about a pathway for a future grid largely composed of renewables if one cannot model demand response and large amounts of different types of storage. The FGRS authors acknowledge these limitations. We look forward to next-generation modeling tools for ISO-NE planners and caution that policy makers should not draw premature conclusions from this study. - Other studies have come to different conclusions about the ability of 100% renewable systems to be reliable.
- A study from Stanford, which has been examined by the FTG-Technical group, has come to the conclusion that a reliable grid throughout the US is possible with 100% renewables. One of the significant differences between the Stanford study and the FRGS study is the use of green hydrogen for long-term storage in the Stanford study.
- A recent white paper from a California energy company showed through modeling that providing all-renewable energy 99 percent of the time could meet hourly demand in its region. It concluded that the cost of doing so added minimal to no cost compared to matching demand with renewable power supply on an annual basis. Although there are clear regional differences between NE and California, it is still puzzling that the FGRS study concluded that a doubling of renewable capacity over nameplate capacity was needed to meet reliability.
- A study from Stanford, which has been examined by the FTG-Technical group, has come to the conclusion that a reliable grid throughout the US is possible with 100% renewables. One of the significant differences between the Stanford study and the FRGS study is the use of green hydrogen for long-term storage in the Stanford study.
- We need to think outside the box and question the assumptions that large increases in electricity demand are inevitable and that current standards for reliability are appropriate for a future of weather extremes.
The US has one of the largest amounts of energy use per capita in the world. The US uses five times the energy per capita as Uruguay, which has succeeded in building close to a 100% renewable grid (see NYT article). It is time that our politicians and regulating authorities talk about reducing overall demand as well as shifting demand away from peak times. The tendency so far is simply to plan for meeting increased demand with increased generation. As for reliability, does it make sense to emphasize the objective of building a grid to achieve less than 2.4 hours of anticipated downtime annually when a much bigger problem is extreme weather events that are happening at an ever-more rapid pace that bring parts of the grid down for extended periods? - We are facing a full-blown climate emergency and our grid-planning authorities should act accordingly.
The “deep decarbonization” scenario for 2040 that became the focus of this study still emits 13.2 million tons of CO2 in New England annually. Phase 2 of the study should address the latest state policies, and Massachusetts, for example, includes a 2050 limit for electricity generation that is a 93% emissions reduction rather than an 80% reduction from 1990 levels. We will continue to monitor these studies.