Some 292,000 megawatt-hours of wind power were curtailed by the Midwest transmission system operator in 2009; in 2010, that figure jumped to about 824,000 megawatt-hours. At a rule-of-thumb 7 cents per kilowatt-hour, that makes the rough cost for one system’s 2010 generation capacity losses $37,240,000 more (at $57,680,000) than losses in 2009 ($20,440,000).
Even in energy generation, those are big numbers -- but the rate at which they have increased is even more intimidating to a wind industry struggling to keep costs competitive.
“Curtailment,” Gary Moland, Director of Market Analysis with GL Garrard Hassan, explained, “is reducing generation at a facility below what it could be capable of producing.”
“For most generation facilities,” Moland said, “if you stop putting power out, you also stop consuming the fuel so the consequences of being curtailed are sort of offset.” If, however, “you are not able to take advantage of the wind blowing, that wind is gone.” Moland went on, saying this practice “has a bigger impact on the economics of the plant.”
“Capacity factor measures the power output from a generator as a percentage of its maximum capability. With most forms of generation, you would operate at whatever capacity factor is economic,” Moland said. “With wind, it is dictated by however much wind is blowing.”
The best wind sites have about a 40 percent capacity factor, meaning they produce an average of 40 megawatts per 100 megawatts of nameplate capacity over “all the hours of a year.”
“If you’re losing -- curtailing -- two percent of that generation,” Moland said, “it really has a big impact.” Capacity factor is reduced to 38 percent, perhaps five percent of a wind farm’s production. “All the cost of a wind plant is upfront,” Moland explained. “All the financial models looking at rate of return on investment assume a certain level of wind availability.”
In Texas, where transmission congestion is a significant problem because wind development has been big and fast, “It’s not unheard of for wind to be curtailed between ten and twenty percent,” Moland said. “You’re moving from 40 percent capacity factor down to 32. That’s twenty percent of your output.” It is a serious hit. “Your revenue stream is less and your return on investment is lower.” At a time when wind is locked in an urgent fight with natural gas for market share, curtailment could be the make-or-break issue.
Where wind is curtailed, it can ruin a plant’s profitability. Where it is anticipated by forward-looking models such as the ones Moland uses to perform congestion studies at GL Garrard Hassan and those used by transmission planners, it can make a viable development non-viable.
Far and away the primary cause of curtailment is limited transmission, Moland said. There has in recent years been large-scale rapid development in the wind-rich Midwestern regions of the U.S. from the Canadian border to Texas, “where you get that 40 percent wind availability.” Transmission development has not kept pace.
“You can go from starting construction to delivering power at a wind site in twelve months, or even less, whereas building a transmission line to service a new site,” Moland said, “usually takes three to five years.”
Initial development usually does not burden existing lines, he explained, “but once you get over a certain level, transmission lines become congested during high wind hours and the only solution to not overloading those lines is to reduce the amount of electricity coming out of the wind turbines.”
Moland mentioned other factors that could idle turbines, including (1) an operator’s requirement for instantly available spinning reserves, (2) scheduled or unscheduled turbine maintenance or (3) extreme weather conditions. However, none of these come near to being as significant as the nationwide impact on wind’s economics due to inadequate and congested transmission.
“Lenders who finance plant development expect a certain return,” Moland said. If “that plant under-performs, it impacts both the return economics and their appetite to finance additional plants,” Moland said.
Most lenders now require congestion curtailment studies, which assimilate a wide range of curtailment risk factors. “Along with how good the wind is and the potential markets to sell the wind, transmission is now a piece of the analysis,” Moland said. “If transmission limitations are identified, it’s going to make it an extra hurdle in getting the site financed.”
One additional -- and particularly crucial -- way that curtailment impacts wind economics is through utility power purchase agreements (PPAs).
Utilities motivated to add wind to their portfolios, either because of state mandates or because they anticipate that such assets will accrue in value, now may find it necessary either to require developers to accept a certain level of unreimbursed curtailment or to commit to paying even if production is curtailed. Often, it is only by insulating developers from curtailment losses that utilities can meet planned or mandated goals.
“The utility has some mechanisms in the market to hedge congestion risk,” Moland said. But what the utility cannot hedge may, of course, be passed along to the ratepayer.
The best solution for eliminating curtailment, Moland speculated, would be cost-effective energy storage. But the three best candidates (pumped hydro, compressed air and battery storage) presently have challenges that limit their cost-effectiveness.
New transmission is the ultimate answer. It requires an infrastructure investment but, Moland pointed out, studies such as the 2008 Joint Coordinated System Plan from eastern U.S. system operators suggest that spending to build transmission saves costs, such as those of curtailment, in the long run.
COMMENTARY: I always thought that the utility company transmission lines could handle the additional electricity load and voltage generated from wind turbines, but apparently this was a wrong assumption. You would think that wind turbine developer's would play closer attention to transmission line issues that could result in curtailment of power generated. You can't control the amount of wind. That's up to nature, but you should be able to determine how much curtailment is likely to occur and metigate your risk. If curtailment is a regular issue, then the wind turbine developer may have overbuilt, and has more wind turbines than it needed in the first place. No wonder ROIs are down for wind farms.
Utility company transmission lines can only transmit a certain amount of electricity. It should be simple math (okay this is an over-simplication on my part). Either the utility transmission lines can take the additional load from wind farms, or they can't. It's like determining how much energy a plant will consume and making sure that the utility company can furnish that amount of electricity and has the necessary infrastructure in place to deliver the electricity to the plant. If the utility companies cannot take the additional load from wind, then they need to increase transmission line infrastructure. This is easier said than done, of course, but somebody is to blame.
Curtailment is certainly a complicated issue to understand and model, but an article dated September 3, 2010 appearing in The Pterra Blog explains wind farm curtailment issues that can result using graphs depicting curtailment under three different situations:
Transmission capacity versus wind capacity by load level
In the above graph designed for one wind farm, curtailment ccurs within a certain range when wind farm's capacity exceeds the utility companies transmission capacity based on a given percentage of peak load.
Transmission capacity with competing wind farm versus wind capacity by load level
The above graph is designed for more than one wind farm. Curtailment involving several wind farms results in a higher level of curtailment over a longer range of peak loads. Wind farm developers need to take into account the capacity output of competing wind farms and the utility companies transmission line capacity under different peak loads.
Transmission capacity with transmission line outages due to maintenance versus wind capacity by load level
The above graph depics curtailment levels that are likely to occur with the utility company transmission lines are out due to maintenance. Under these conditions, wind farm capacity curtailment can be severe because there is no where to transmit the electrical output. I would assume that if you modeled for multiple wind farms using the same utility company transmission lines, curtailment levels will be sky high, a bad situation if there ever was.
The above graphs show wind farm output curtailment under different conditions and utility company line transmission peak loads, but it does not provide insights into how to actually forecast curtailments.
A white paper dated December 2009 by the National Renewable Energy Laboratory titled, "Central Wind Power Forecasting Programs in North America by Regional Transmission Organizations and Electric Utilities", provides a list of forecasting programs utilized by various North American utility companies including their forecasting vendors and models they use.
The NREL also provides a broad resources for forecasting and modeling. There's a rather lengthy list of papers and research data about wind curtailment HERE. One NREL paper in particular dated April 2010 titled "Status of Centralized Wind Power Forecasting In America", could be a use if you can understand it. Boy, I think I am over my head.
It sure as hell looks like forecasting wind farm curtailment issues can be very complicated, at least from the utility side. You would think that the utility companies would share this information witgh the wind farms.
Courtesy of an article dated March 24, 2011 appearing in GreenTechMedia