This content originally appeared on AWEA’s Into the Wind Blog.
By: Michael Goggin
The cost of reliably integrating large conventional power plants onto the power system in Texas is more than 17 times larger than the cost of reliably integrating wind energy, based on new AWEA analysis of data from the state’s independent power grid operator.
This analysis rebuts one of the most widely-held misconceptions about how wind energy is reliably integrated onto the power system. While it is true that wind energy’s variability does slightly increase the need for the balancing reserves that grid operators use to keep supply and demand in balance, all forms of energy impose integration costs on the power system. (pages 11-16).
In fact, Texas grid operator data show that the integration costs for conventional power plants are far larger than the integration costs for wind generation, even though Texas has more wind energy than any other state and one of the highest levels of wind generation for a U.S. grid operator. Because changes in wind output occur gradually over many hours and can be predicted, while failures at conventional power plants occur instantly and without warning, more reserves and more expensive reserves are required to reliably integrate conventional power plants. For example, the Texas grid operator ERCOT holds 2800 MW of fast-acting reserves 24/7/365 to keep the lights on in case one of the state’s large fossil or nuclear power plants experiences an unexpected failure, as all power plants do from time to time.
The following table compares the reserve costs for wind versus other sources of variability on the ERCOT grid.
|Factor||Total annual cost (million $)||% of total reserve cost||Cost per electric bill|
|Conventional power plant failures||$239.690||67%||76 cents|
|Conventional and demand deviations||$103.359||29%||33 cents|
The table above is directly calculated from the following ERCOT data. The first three rows in the following table list ERCOT data on the incremental amount of reserves it holds to accommodate various sources of variability, while the fourth row lists the average cost of those reserves in 2013. The last three rows use this data to calculate the total reserve cost for each source of variability.
|Regulation down||Regulation up||Responsive reserves||Non-spinning reserves|
|Contingency reserves for conventional power plant failures (MW)||2,800|
|Incremental reserves for wind (MW)||14||42||328|
|Electricity demand variability and deviations at conventional power plants (MW)||476||508||1,474|
|Cost of reserve ($/MW)||$4.89||$8.57||$9.77||$3.47|
|Annual reserve cost for conventional power plant failures (million $)||$239.690|
|Annual reserve cost for conventional||$0.585||$3.159||$9.996|
|Annual reserve cost for electricity demand variability and supply deviations at conventional power plants (million $)||$20.372||$38.126||$44.860|
As the table shows, the cost of additional reserves to accommodate wind accounts for about 4.3 cents out of a typical Texas household’s $128 monthly electric bill, or 1/30,000th of a typical electric bill. In contrast, the $240 million cost of reserves to accommodate conventional power plant failures works out to about 76 cents per monthly electric bill. In other words, the total cost of contingency reserves for conventional power plant failures is more than 17 times larger than the cost of all wind-related reserves.
On a per MWh of energy produced basis, wind’s reserve cost is still about half as large as conventional power plants’ reserve costs. Wind’s reserve cost is about $0.37/MWh of wind when allocated across the wind MWh generated in ERCOT last year, which equates to roughly 1% of the typical cost of wholesale electricity. In contrast, the cost of contingency reserves was $.65/MWh when allocated across all MWh generated in ERCOT last year, and even higher if only allocated to generation from the larger conventional power plants that cause the need for contingency reserves.
Wind’s contribution to reserve needs and cost is small because many changes in wind output are canceled out by opposite changes in electricity demand, resulting in only a small incremental increase in total reserve needs. In addition, because changes in wind output occur gradually, incremental wind variability is mostly accommodated using low-cost non-spin reserves, which cost about 1/3 as much as the expensive fast-acting reserves used to accommodate conventional power plant failures.
Attacks on wind’s integration costs fall flat
Anti-wind groups often attempt to criticize wind for what they claim are its sizeable integration costs, often relying on inaccurate or obsolete estimates of the reserve needs caused by wind and almost always ignoring the far larger integration costs imposed by competing forms of energy. Ironically, these attacks often come from groups that receive funding from those competing energy sources that impose far larger integration costs.
As an example that occurred today, Tom Tanton, who is affiliated with fossil fuel industry-funded groups, gave Congressional testimony attacking wind energy for its “hidden costs,” referring primarily to integration costs. Yet he was completely silent about the far larger integration costs associated with other energy sources. Similarly, Jonathan Lesser, who works as a consultant for competing energy sources, last year gave Congressional testimony attacking wind energy for its integration costs. When asked in a direct written question from a Congressman about the integration cost of conventional power plants, he attempted to dodge the question by wrongly interpreting “integration costs” to mean transmission costs, and only answering that question. This dodge stretches credibility as Mr. Lesser had used the term “integration” and “integration cost” more than 35 times in his own testimony to refer to the balancing or operating reserves needed for wind.
Adding insult to injury, in some parts of the country wind plants are charged for reserves that are used to integrate wind, while the cost for the contingency reserves used to accommodate abrupt failures of conventional plants is never paid by those power plants but is rather rolled into your monthly electric bill.
Some have recently attempted to use integration costs to attack wind energy in Texas. For those concerned about the impact of integration costs on ratepayers, this focus clearly misses the mark. Focusing on wind integration costs that account for around 1/30,000th of a typical electric bill, particularly while ignoring the 17-20 times larger integration costs imposed by large conventional power plants, makes no sense if one is truly concerned about protecting consumers.
Weighing in on the Texas discussion, Forbes contributor Loren Steffy gets nearly all of the facts wrong in attempting to support the argument that wind should be singled out for paying these and other costs in Texas, while competing forms of energy would get a pass. He argues that “wind companies should be treated the same as other generators when it comes to accessing the grid,” which is exactly the wind industry’s position, while in contrast others are arguing that wind should be singled out for paying its transmission and integration costs. He also falsely writes that “[Consumers] pay the cost of maintaining the system in their bills, while paying part of their taxes to subsidize the construction of the additional lines to the wind farms.” To be clear, no taxpayer money is or has been used to pay for transmission lines in Texas, for wind or any other energy source.
Thankfully, the attempt in Texas to single out wind for paying transmission and integration costs while all other energy sources do not pay those costs has fallen flat. The San Antonio Express-News recently weighed in in opposition to the proposal to discriminate against wind with an editorial headlined “Attack on Wind Energy is Hot Air.”
As part of our analysis, we confirmed that these results hold for other periods of time and other ways of analyzing the data. For the last 365 days ending on July 20, 2014, the cost of conventional power plant contingency reserves was more than 20 times larger than the cost of wind-related reserves, even higher than the 17 times larger figure for 2013 calculated above. We also took the monthly average reserve cost and multiplied it by the equivalent calendar month’s average reserve need (i.e. the October 2011 reserve need was multiplied by the October 2013 reserve pricing), instead of using yearly averages for both, and found it increased the calculated cost of wind-related reserves by about 2%. Because ERCOT’s reserve level analysis was conducted several years ago and to our knowledge has not been updated, the reserve need data are not synchronized with the more recent reserve pricing data, though there is no reason to believe the lack of synchronization itself would significantly bias the result in either direction because reserve pricing is driven far more by energy market pricing and electricity demand and supply than by reserve demand.
A synchronized analysis was conducted using 2011 reserve pricing data, and the total wind reserve cost was found to be slightly higher at $.50/MWh, likely reflecting the fact that natural gas prices and electricity prices, and therefore reserve prices, were somewhat higher in 2011 than in 2013. ERCOT’s data on wind’s incremental reserve needs were calculated when the grid operator had about 9,800 MW of wind on its power system, while today there are slightly more than 11,000 MW of wind on ERCOT’s system, so the need for wind-related reserves is likely to be marginally higher today.