How much “backup power” is needed for solar and wind?

This blog was written by John D. Wilson, former Deputy Director for Regulatory Policy at the Southern Alliance for Clean Energy.

Guest Blog | May 28, 2015 | Energy Policy, Solar, Wind

Answer: None.

Not if the utility is planning correctly!

Ok, that was a smart-aleck answer. But I’ve got a point: When people talk about a “backup,” they tend to think of a one-for-one replacement. I remember when my kids were in diapers, and I had to carry four diapers for a long day out, just in case. That’s three “backups” actually!

But for a large utility, solar and wind power do not need a special “backup” generator. Utilities can plan their system to anticipate what additional generation they will need to ensure reliability, and it is very rare that they need a one-for-one generation backup for solar and wind.

Usually, when I’m at utility planning meetings, the concerns I hear are about the lack of solar during “Polar Vortex” conditions or the lack of wind during hot summer afternoons. Fortunately, there are excellent wind and solar resources available to most Southeastern utilities. Accessing them in combination is a smart way to ensure that renewable energy doesn’t need “backup generation.”

Rather than building power plants, utilities need to build better power plans. In those plans, there are four key ways that utilities can include solar and wind power while enhancing reliability. When planning for future needs, utilities need to:

  • Rate solar and wind properly for peak demands;
  • Recognize the contribution of above-average output from solar and wind to reducing power shortage risks;
  • Proactively identify needs for upgrades to power lines and other transmission equipment; and
  • Recognize the contribution of solar and wind to reducing system wear-and tear.

In my blog last week about reliability and EPA’s clean power plan, utilities already have the planning and operational tools they need to ensure reliability. None of these activities are “new” to utilities. Upgrading these efforts is essential for a future that is almost certain to include high percentages of renewable energy on utility systems.

Rating solar and wind properly for peak demands

Wind, and especially solar, can dependently produce energy when demand is high.  These renewable resources can be relied upon for a substantial amount of electricity, despite the fact that they can’t be relied upon to generate 100% of their maximum output in every peak hour. During an average peak hour, wind and solar will generate roughly 55% of maximum potential output (as illustrated below). Although that generation isn’t “guaranteed”, it is unlikely that generation will be close to zero. In fact, in our 15-year dataset, generation never falls below 15% for the top 5% of load hours. Occasionally there will be hours when wind and solar fall to near-zero levels, but during those hours TVA’s system demand is well below its forecast peak.

This graph illustrates the amount of power that wind (sourced from western Oklahoma) and solar (sourced within TVA) could contribute to meeting peak loads. As the graph illustrates, when TVA’s load grows, renewable energy generation tends to be higher and tends to be less variable.

Returning to the question posed in the blog title – how much “backup generation” would TVA need for its wind and solar?

If TVA added 4,000 MW of wind and solar to its portfolio, with a capacity rating of 55%, then it would be only be counting on generation of about 2,200 MW. But – what happens if output is only 35%? Because TVA is only “counting on” 55%, the shortfall is only 20% (or 80 MW) and not 65%. This 20% need for “backup generation” represents about 2% of TVA’s system, an amount that TVA’s system already provides in the form of a reserve margin. Therefore, with a properly designed reserve margin and a proper estimate of the dependable capacity rating for renewable energy, TVA’s planning process would have all the “backup generation” it needs.

These findings aren’t unique to TVA. We’ve also studied the Southern Company and Duke Energy (Carolinas) systems, and we’ve started on an analysis of the Florida utility systems. Many of our findings are discussed in technical detail in Southeast Renewables and Reliability, a report prepared in support of our comments to EPA on its proposed Clean Power Plan rule.

Average output from solar and wind will help reduce power shortage risks

What’s more, by giving renewable energy a capacity rating of 55%, TVA is actually reducing power shortage risks during many hours – contrary to what many people think about variable energy resources!

Resource planning isn’t about a fixed threshold – it’s about reducing the risk that there will be exceptionally high demand and inadequate generation. The nice, neat tables that count up the utility’s generation and expected customer demand for power in the coming year don’t actually reflect the reality that will occur on any particular day. There’s a lot of variability in the utility system already and so-called variable generation resources, like solar and wind, are just another source of variability.

Again, if TVA added 4,000 MW of wind and solar to its portfolio, with a capacity rating of 55%, then it would be only be counting on generation of about 2,200 MW. During half of peak hours, generation would be more than 2,200 MW – providing additional reliability during those hours and essentially providing “backup power” for conventional generation when it occasionally fails. As shown in our Southeast Renewables and Reliability report, we found far more hours in which reliability was enhanced by solar and wind generation than hours in which the system had a higher risk of reliability problems.

Supporting future solar and wind with upgrades to power lines and other transmission equipment

Adding 4,000 MW of solar and wind will require some changes to utility transmission systems. Recently we hired Mitsubishi Electric Power Products, Inc. (MEPPI) to study a very high renewable energy generation scenario on the TVA transmission system. In the study, we replaced 4,800 MW of coal generation with roughly 4,100 MW of solar and 3,500 MW of wind (sourced from Oklahoma via HVDC transmission). MEPPI evaluated a range of high and low generation scenarios during TVA’s summer and winter peak hours.

MEPPI’s analysis showed that with relatively modest improvements, TVA’s transmission system could be updated to support these new patterns of electricity flows. MEPPI’s rough cost estimates for these improvements was around $89/kW. That’s less than 10% of the cost of a new natural gas combined cycle plant. This estimate may be on the high side, as it doesn’t include any optimization efforts in terms of ideal siting or coordinated planning with other transmission system upgrade needs.

Practically, MEPPI had two key observations that would be relevant to any utility’s plan. First, MEPPI noted that the “number of lines and transformers that are required to be upgraded will require a significant amount of coordination in regards to scheduling outages. The timeframe that would be required to make these system upgrades could be fairly significant.” Second, that “reliability can be assured through methodical implementation of standard mitigation measures utilized by the utility industry at a cost that is relatively modest compared to other investments being made by TVA.” The MEPPI study therefore shows that strong advance planning is needed to achieve high levels (e.g., 7,600 MW) of renewable energy deployment on a large utility system, but that the actual work that needs to be done is “standard.”

Solar and wind can reduce system wear-and tear

Annual system regulating capability (SRC), expressed as a percentage of peak load, measures the flexibility of the five planning strategies. TVA considers flexibility – the ability of the system to respond to load swings – as a key consideration for long-range resource planning, but has not established a minimum or optimum flexibility score for the TVA system. Even though improvements are lower for Strategy E, all five strategies have better SRC scores in 2024 and 2033.

Another “reliability” issue that’s been raised is the concern that the variability will result in more wear-and-tear on conventional generation plants, which must be adjusted to compensate for both changes in solar and wind generation as well as overall load changes. We studied this issue as well in Southeast Renewables and Reliability and our analysis showed that utility operations actually become less challenging (with more gradual hourly changes in power generation)  for the first 4,000 MW or so of renewable generation. After that point, power operations could become more challenging but with the changes occurring gradually.

These greater “challenges” may actually be quite easy to meet. Last week, my blog discussed TVA’s analysis of several strategies’ impact on the flexibility of its generation fleet. As shown at right, even without considering the technical capacity of wind to quickly and easily “ramp down,” TVA found that system flexibility will rise in all the strategies it is considering, even its “Maximize Renewables” strategy. So even if renewable energy requires more active hourly adjustments to power plant output, it would appear that TVA (and  likely other utilities) are already planning to build fleets with these capabilities.

In summary, putting renewable energy in utility plans is both practical and will provide for even more reliable electric service. With the proper planning practices, utilities can include renewable energy without a need for “backup generation” for wind and solar power resources. The utility system, properly planned, provides all the “backup generation” that is needed. Its a little counter-intuitive to many people, but that’s why we ran the numbers …

Guest Blog
My Profile