This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.

Post

Storage Not Needed to Accommodate Higher Levels of Wind Energy

Michael Goggin's picture
Electric Industry Analyst American Wind Energy Association

Michael Goggin joined AWEA in February 2008. Michael works to promote transmission investment and advance changes in transmission rules and operations to better accommodate wind energy in the...

  • Member since 2009
  • 45 items added with 9,277 views
  • May 8, 2015
  • 2334 views

Recently, energy storage has been in the news. Contrary to a common misconception, very high levels of wind energy can be reliably integrated without energy storage. Energy storage is typically more expensive than grid operating reforms, which can provide the same flexibility services that storage provides.

Large amounts of wind energy are already being reliably integrated.

The U.S. has been able to install enough wind power to power the equivalent of 18 million average American homes without adding any large-scale energy storage. Similarly, European countries like Denmark, Spain, Ireland, and Germany have successfully integrated very large amounts of wind energy without having to install new energy storage resources. In the U.S., numerous peer-reviewed studies have concluded that wind energy can provide 30 percent or more of our electricity without any need for energy storage.

The key to doing so lies in using the sources of flexibility that are already present on the electric grid. There is large variability already present on the electric grid due to changes in electricity demand and supply as consumers turn appliances on and off and power plants unexpectedly go out of service. Grid operators constantly accommodate this variability by increasing and decreasing the output of power plants and turning to other sources of flexibility. By controlling the output of power plants, water held behind hydroelectric dams or natural gas left in a pipeline is used as energy storage. Because this flexibility has already been built into the system, it is almost always much cheaper to use this flexibility than to build new sources of flexibility like energy storage facilities.

The high cost of energy storage relative to other sources of flexibility, including those on the existing power system, is the chief reason why it is not more widely used today. As shown in the National Renewable Energy Laboratory chart below, improved grid operations are the low-hanging fruit for making the power system more flexible. These reforms more than pay for themselves by allowing more efficient power system operations, and are more than sufficient to accommodate even very high levels of wind energy.

energystorage1.jpg

Most storage technologies have limited ability to provide the services needed at very high penetrations of wind energy.

Wind energy’s changes are gradual and increasingly predictable, making it cheaper to provide flexibility using slow-acting reserves. Fast-acting reserves, such as flywheels and advanced batteries, can be cost-effective for accommodating variability that occurs on the second-to-second time frame (as shown in the chart below from ITM Power), but changes in wind tend to occur over time periods of 30 minutes or more. That means these technologies provide little to no value for wind integration. Pumped hydroelectric storage, with its ability to store large amounts of energy for long durations, is the only energy storage technology that is currently available that comes close to providing the type of service that wind energy would need at very high penetrations.

http://www.aweablog.org/wp-content/uploads/2015/04/energystorage.jpg

In some cases having certain types of energy storage on the grid can modestly reduce the cost of integrating wind. However, in other cases, energy storage has been found to actually provide negative value for the integration of wind energy, even if the energy storage was provided at no cost. Regardless, given the low cost of using existing flexibility to integrate wind energy, and grid operating reforms that enable far greater use of existing flexibility at negative cost, energy storage technologies should not be viewed as an essential tool for the integration of renewable energy.

There is no need for individual power plants to provide constant output.

Some people incorrectly assume that wind output must be “firmed,” i.e. have its variability leveled out, by storage or another resource to make it valuable to electric utilities or system operators. In reality, many changes in wind output actually cancel out opposite changes in electricity demand or supply, as the electricity supply and demand is constantly in flux. Therefore, storage should not be used as a dedicated resource for a single generator or load, as attempting to “firm” a source of variability that was already being canceled out can actually add to the total variability on the electric grid. Regardless, a wind plant is seldom the optimal location for deploying energy storage.

If storage is used, it should be seen as a system resource. The only form of energy storage that is currently operational on a large scale in the U.S. is pumped hydroelectric storage, with a little over 20 GW of installed capacity. Much of this storage was built to provide flexibility to help accommodate the significant increase in nuclear generation that occurred during the 1960’s, 70’s, and 80’s. Just as it is typically not economic for wind plants to increase their output in response to grid demands, all U.S. nuclear plants and many coal plants tend to provide little to no flexibility. This shows that all inflexible generators benefit when other sources of flexibility, including energy storage, can relieve them of having to accommodate changes in electricity supply and demand. In fact, studies in the Netherlands and Ireland found that coal plants were the primary beneficiaries of energy storage. Energy storage allowed coal power plants to run more at night, with this low-cost energy being stored and used to displace more expensive natural gas generation during the day, interestingly causing a net increase in electric sector carbon dioxide emissions. In the U.S., data from the Department of Energy show that pumped hydro storage use declined drastically in 2012 when abnormally low gas prices created an incentive for coal plants to begin cycling their output, reducing the need for storage to provide this flexibility.

Storage’s usefulness in niche applications does not mean that it is needed to increase wind use.

In certain rare situations, it could make sense to site energy storage near a wind plant. If a constraint on the transmission grid prevents a wind plant or group of wind plants from selling their full output on a consistent basis, it could be economical to store electricity that would otherwise have been curtailed. However, this type of application is a short-term fix; building out the transmission grid is typically the more optimal long-term solution to a transmission constraint.

In addition, it is important to keep in mind that while energy storage can be an economically attractive option in certain niche applications, such as small island power systems, this does not indicate that energy storage is an economic option on large mainland power systems. Small island power systems, due to geography and fuel mix, often lack low-cost sources of flexibility such as an ability to exchange power with neighboring grid operators. In contrast, mainland U.S. power systems can far more cost-effectively manage variability from all sources by using transmission to exchange power with a neighboring power system.

Continuing advances in energy storage technology can make it more economically competitive as a source of grid flexibility, and improving the performance and reducing the cost of battery storage remains critical for enabling greater electrification of the transportation sector. There is significant potential for the batteries of plug-in vehicles to be used as energy storage for the grid, particularly by simply altering the rate of charging of these batteries rather than discharging and recharging the batteries. While the potential of such technologies is exciting, it is important to remember that resources like wind energy can already be cost-effectively and reliably integrated with the electric grid without energy storage.

This blog is based on Chapter 11 of the AWEA white paper “Wind energy helps build a more reliable and balanced electricity portfolio.”

Michael Goggin's picture
Thank Michael for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member
Discussions
Spell checking: Press the CTRL or COMMAND key then click on the underlined misspelled word.
Clayton Handleman's picture
Clayton Handleman on May 8, 2015

Michael,

Thank you for the wealth of resources provided in your post.  In the two posts linked to below I looked at the relatively untapped potential of high CF wind in the great plains.  I.e. the best sites have little transmission access so most great plains wind is coming in at about 40% CF. However, there are regions with substantial amounts at 50% and higher.  In my first pass on your post above, it appears that it is looking at the potential for high penetration as things currently are.  Are you aware of sources that look at the added potential for penetration if the high CF sites are accessed?

Best sites have limited build out due to lack of transmission access.

Using 100 meter towers, there is nearly 2 terawatts of potential wind power at 50% and higher CF.  That is land based and does not take into account off shore potential.

Nathan Wilson's picture
Nathan Wilson on May 9, 2015

very high levels of wind energy can be reliably integrated without energy storage

This obviously depends on your definition of “very high“.  The 2012 report from Lawrence Berkeley (LBNL-5445E) which provides the reference case for the 2014 report you’ve linked gives numbers that will be good news for the wind industry, as they allow a wind penetration which is10x higher than the current US average. The results are bad news for anyone hoping for a near complete phase-out of fossil fuel based on wind and solar.  Their analysis shows the marginal value of wind power drops by around 40% as wind penetration grows from 0 to 40%.  Solar PV fared even worse, losing 40% of its value at only a 10% penetration (and falling steeply below the value of wind).

The 2014 report (lbnl-6590e) studied ways to mitigate the plummeting value of wind and solar, without much success.  Storage boosted the value of solar for penetration between 10 & 30%, but the increase in value was only by about 3 ¢/kWh (about 10-20% of the cost of the new Tesla battery).

For wind power at 40% penetration, adding storage only helped by about 0.4 ¢/kWh, well below the most optimistic projects for future storage costs.  The best option they found gave 1 ¢/kWh in savings with better geographic diversity in the wind farm placements, but the reference scenario already spread the wind farms across California, Washington State, Oregon, Wyoming, Idaho, Arizona, and New Mexico, so it may be that they are accepting less economical sites to boost market value.

We are often told that wind and solar are somewhat complimentary, but the 2014 report showed that there is still some overlap, as adding 10% PV to the 40% wind scenario pushed down the value of the wind power by another 0.5 ¢/kWh.

With fossil fuel lock-in implied by wind and solar running out of steam at around 40 and 10% penetration respectively, even with geographic diversity spanning almost the entire western US and California’s excellent solar resource and flexible hydro, it is striking to note that France gets around 75% of its electricity from nuclear power, and by combining nuclear with hydro, France, Sweden, and Switzerland have achieved nearly complete elimination of fossil fuel from their electricity generation.  The nuclear option offers additional elimination of fossil fuel use with reliable night-time charging of EVs, by powering district heating networks for cheap domestic heat and hot water, and by supplying low cost industrial process heat. 

Clayton Handleman's picture
Clayton Handleman on May 9, 2015

See Reply to Nathan Wilson Post below.

 

Clayton Handleman's picture
Clayton Handleman on May 9, 2015

Nathan,

The author’s premise that there can be considerable expansion of wind power without adding storage is not impacted by whether or not there is Fossil Fuel Lock-In (FFLI).  However FFLI is an important concern and should be one of the factors considered in the development of a 21st century grid.  I point out that the LBL study is for California and not a good national proxy.  Using it in an attempt to prove the inevitiability of FFLI is not sound.  While it does draw on a large geographic area, only a small fraction of that is viable for wind power.  AWEA is clear that they support higher levels of transmission to decorrelate sources thus reducing the intermittency.  Under that scenario, vast, high Capacity Factor, decorrelated wind sources are tied together, reducing intermittency.  This also will diminish the need for FF backup.  A quick glance at a US wind resouce map shows that that the LBL study only nibbles at the edges of the best wind resources in the country which are in the great plains states and off the coast on the Atlantic seaboard. 

 

 

Graphic Above:  Blue line shows the LBL study area.  Note that power goes as the cube of wind speed.  So, for example 7 m/s wind has about half the power of 9 m/s wind.

In the Great Plains, using energy as the criteria, the LBL study only accesses about 15% of the available wind energy.  Based upon their map and the NREL wind resource map, if one makes reasonable assumptions about Montana that drops to about 7%.  Further reducing its relative value, it only includes a very small fraction of the highest Capacity Factor wind sites (those in Wyoming).  The vast swaths of 50% CF wind in TX, KS, NE, ND and SD are not included.  Clicking on states in this map offers easy access to data on the CF of wind in those states.  High CF, while not a full proxy for intermittency, is a good indicator.

The mostly decorrelated, off-shore, Atlantic wind is not considered either. 

Taking it a step further, even if we don’t have a supergrid I remain at a loss as to how FFLI is inevitable.  Molten salt reactors can be designed to be complementary to renewables.  Forsberg of MIT advocates a variety of nuclear based storage scenarios in which heat is stored and utilized to smooth out renewables and load intermittency.

While FFLI should be a consideration in selecting the generation mix, presenting it as an inevitble outcome of using renewables seems to me to be off the mark and unconstructive.

Nathan Wilson's picture
Nathan Wilson on May 10, 2015

Clayton, the NREL wind resource plot you’ve provided gives an important reminder that the heavily populated US west and east coasts are not near the best wind resources.  I’ve seen back-of-the-envelope calculations showing that HVDC transmission could allow central plains wind to power the coasts, but the cost of transmission roughly offsets any cost savings from the superior resources.  

Transmission from central to the west has the additional large barrier that the AC grids are not synchronized.  Obviously using HVDC solves part of the problem, but it raises the initial investment even higher.  Normally grids are built incrementally, with each new link providing some benefits, while still being backed-up by alternate routing for contingencies.  There is no meaningful existing transmission linking these two grids, so any new HVDC lines could not be heavily loaded until there was two or three running in parallel.  At around 6 GWatts per 800 kV HVDC line, this is like trying to find investors to fund 4 new nuclear reactors all at once, in a new market, and locating those reactors (and the associated jobs) hundreds of miles from customers, in an industry which is dependent on stable government policy support, for a project with a build time that exceeds the election cycle length.

Remember, the problem with wind power (even with aggregation over larger areas) is not lack of resources (the US has plenty) or low capacity factor, it is negative load correlation (daily and seasonal), and variations spanning multiple days.

It is certainly possible to use nuclear power to back-up variable renewables, particularly the high temperature LFTR and TRISO fueled FHR, which will likely allow thermal energy storage at a cost in the same range as pumped-hydro.  However the 2014 LBNL report   (lbnl-6590e) found that while storage can help solar, wind doesn’t play well with storage (this is part of the reason for my advocacy of syn-fuel production as a dispatchable load, while not currently competitive with fossil fuel, it is a much more potent solution for balancing supply and demand than conventional storage, especially for seasonally storable fuels like ammonia).  The MIT-UC_Berkeley presentation you’ve linked shows a nuclear plant which has both storage, and natural gas power boost; it’s largely the fast-ramping fossil fuel addition that facilitates wind complementation.

While throwing money at wind power’s fossil-fuel-lock-in problems will mitigate them somewhat, I’ve seen no evidence of public willingness to write blank checks for sustainable energy (in fact I’ve read your complaints about nuclear’s cost, even though it is cheaper than off-shore wind, solar thermal, and most PV; and the 80 year plants lives mean that nukes always pay back any needed subsidies with future cheap power).

We must also be aware of our ethical obligations to developing nations, regarding export of ideas and technology.  We know that nuclear plants eliminate fossil fuel power plants on a Watt-for-Watt basis; LBNL (LBNL-5445E see figure 4 on p. 45)  shows once again that even very high penetration of wind and PV only replaces a tiny amount of the required firm generation, so rather than letting them leap-frog fossil fuel technology, variable renewables require them to almost fully deploy it, in addition to the renewable supplement with its transmission & storage.  For nations which lack the enormous size of the US, wind aggregation also means a grid that is dependent on neighboring countries, which may not be economically stable or even friendly.

Engineer- Poet's picture
Engineer- Poet on May 10, 2015

If you have either molten-salt or fast-spectrum reactors (both of which bypass the xenon-poisoning issue which restricts power variations), you don’t have much use for the renewables.

Engineer- Poet's picture
Engineer- Poet on May 10, 2015

I keep seeing this claim in various forms.  I think it’s become a Green shibboleth:

many changes in wind output actually cancel out opposite changes in electricity demand or supply, as the electricity supply and demand is constantly in flux.

This is a half-truth at best, and the implied meaning would only be significant if there was a strong positive correlation between the output of wind generation and grid demand.  I understand that in many areas there is a negative correlation (e.g. the night-peaking wind curve is suggested to be good for charging EVs, not for meeting current demand).  In short, wind does not ameliorate the variability in net electric demand, it exacerbates it.  (The “duck belly curve” shows that the same is true of solar PV.)

This heightened variability is one element of the integration cost of wind.  That cost can be transferred or hidden, but it cannot be eliminated.  Argonne National Laboratory found that even the carbon-emissions benefits of wind power shrank by roughly half at the 35% penetration level.  That just projects fuel consumption, not other costs such as maintenance budgets increased by greater thermal cycling.

However, in other cases, energy storage has been found to actually provide negative value for the integration of wind energy, even if the energy storage was provided at no cost.

This is so counter-intuitive it requires a check of the claim, especially because it contradicts the earlier claim about energy storage being provided by natural-gas pipelines and hydro dams.  Here’s what the actual document says:

At the same time, the assumed low cost of storage capacity reduces the capacity value of wind. Since storage is now the option with the lowest investment cost, it becomes the new capacity resource. Fewer hours with scarcity prices are required to cover the fixed cost of investment in storage compared to the number of hours required to cover the cost of a CCGT. This in turn lowers the capacity value of wind, since wind now generates less power during periods with scarcity prices.

A reduction in capacity value and negative value are completely different things.  Also, the words “no cost” do not appear in the document, and “zero” is not used in the context of storage cost.

This would be a rather shocking lack of honesty in anything except an advertisement.  Since this is a summary of material from the AWEA, it can be considered a promotion… but it really should have been explicitly marked as such at the outset.

Clayton Handleman's picture
Clayton Handleman on May 10, 2015

Isn’t that what Clean Line Energy is in the process of doing?

“There is no meaningful existing transmission linking these two grids, so any new HVDC lines could not be heavily loaded until there was two or three running in parallel.”  – 

This does not make sense to me.  Do you have a source that clarifies this point?  Clean Line is developing multiple HVDC lines mostly around 3.5 GW.  There is nothing in their plans about redundant parallel DC lines.

Regarding multiple grids HVDC allows connection of multiple grids.  It appears you are suggesting that that adds to the cost.  I think that is incorrect.  The down conversion has nothing to do with the phasing of the up conversion.  So whether you are connecting back to the same grid in a different place or a different grid, the down conversion will be the same and phasing will not have anything to do with the sourcing grid.

“(in fact I’ve read your complaints about nuclear’s cost, even though it is cheaper than off-shore wind, solar thermal, and most PV; and the 80 year plants lives mean that nukes always pay back any needed subsidies with future cheap power)”

My comments on this are simple, if it is as good as advocates say it is then it should be funded by investors just like wind farms are.  Why should the rate payers have “cheap” nuclear power jammed down their throats.  The other concern I have with nuclear is waste disposal.  The public was promised that this would be solved, it is not.  It remains an open ended back end cost that is being pushed onto someone else.  It is being externalized. 

Next gen nuclear looks interesting – I have not satisfied myself that there are satisfactory non-proliferation safeguards but based upon my current understanding I don’t criticize it.   

“Remember, the problem with wind power (even with aggregation over larger areas) is not lack of resources (the US has plenty) or low capacity factor, it is negative load correlation (daily and seasonal), and variations spanning multiple days.”

Yes if EVs go bust and don’t happen then I agree with you.  However trends appear to be moving rapidly toward high EV penetration.  That will shift load to night time – That will substantially improve the correlation between wind power and load. 

Whatever happens it will be at least 10 years before we see high penetration of wind or nuclear.  In that timeframe do you think that EVs will stagnate or do you think their will be significant growth in that industry? 

 

Nathan Wilson's picture
Nathan Wilson on May 10, 2015

The US electrical system is divided into three separate grids, Western, Eastern, and ERCOT (Texas), all roughly 60Hz, but not matching exactly, and not exchanging any power between them.  Here is a whitepaper grid expansion study done in 2007 for the DOE’s 20% Wind report.

On page 8 it shows a possible plan for adding a 765 kVAC backbone for moving wind power around.  Notice that it suggests tying the Western and Eastern grids together in 4 places with AC-DC-AC links, and tying the Eastern and ERCOT grids together in 3 places.  The new 765 kV grid would over-lay the old grid, which provides some redundancy as the new grid segments are added.  Note that the Cleanline is completely within the Eastern interconnect so there are already some 345 kV lines in the area.  Also, some of the new transmission shown for western Oklahoma, Texas, and Kansas has already been built.

Another good resource for explaining how transmission is being deployed to support wind power is the JCSP report, done in 2008 by the transmission system operators on the Eastern interconnect for the 20% Wind study.  See especially appendix 5 on p. 183, which explains HVDC lines and contingency capability.

Regarding the alleged “problem” of nuclear waste, please read thru the report from the President’s BRC on America’s nuclear future, which focuses on nuclear waste disposal plans.  The authors of this report clearly did not think we have any safety, health, or environmental problems with nuclear waste; their main concern was how to win political support for a repository (like choosing a route for a new freeway).  If those smart people are not worried about nuclear waste, why should we worry?  Our fear of radiation is guided by propaganda from the fossil fuel industry, and is not in our own best interest.  Nuclear waste has still never harmed anyone.  Also, waste disposal is not a huge externalized expense (like air pollution); it only amounts to about 0.1 ¢/kWh, and is normally fully funded by electricity sales.

Your implication that investors will flock to the good energy sources is laughable.  Investors will happily fund only the dirtiest planet killing energy sources if that makes them the most profits.  In the US, most wind farms are funded by investors only after the utilities have taken all of the market risk by signing fixed-price power purchase agreements; the only risk the investors take is that construction cost might be different than it was for the last few dozen projects (i.e. low risk).  Power purchase agreements are also being used to fund nuclear plant in other countries such as the UK’s Hinkeley Point C.  Expect the construction risk of nuclear projects to mostly go away as soon as the industry has some recent experience with new builds.

Unfortunately, nuclear projects carry a large risk of political interference, due to the Shoreham fiasco (in which an investor funded nuclear plant was built, then never allowed to operate).  I don’t think it’s fair to ask investors to accept that risk, so it must be born by the public.  That is a burden that anti-nuclearism has rammed down society’s throat; the way to get rid of this risk is not to try to shift it to someone else, but to face our fear and accept that anti-nuclearism is bad science and causes serious harm.  

Nathan Wilson's picture
Nathan Wilson on May 10, 2015

No, I’m saying HVDC can’t be done.  Within an already connected region, HVDC can be used to boost capacity.  I’m just saying that there is an extra barrier to connecting the Western and Eastern grids together, because they aren’t connected yet, and any attempt to do so must be very large scale.  So it wont’ happen gradually, only as a result of a large push with big money.

While nuclear cost over-runs are frustrating, they don’t change the underlying appeal: nuclear still produces reliable clean energy, with easy scalability to a very low carbon grid, with no depence on people changing their behaviors or technology breakthroughs, at a fleet average cost (averaging over the life of the plant) which is competitive with any existing alternative fleet.  Costant attacks on this or that attribute of nuclear power by green groups has only resulted in prolonged fossil fuel use (which is much worse than a cost over-run).

Nathan Wilson's picture
Nathan Wilson on May 10, 2015

Georgetown just decided to build 100% wind and solar by 2017…”

Wind and solar only produce “variable power”, and essentially all electricity users only use “power on demand”. So Georgetown absolutely did not decide to switch to 100% wind and solar.  They may have decided to “offset 100% of their electricity consumption” with wind and solar, but they are still mostly dependent on fossil fuel to make the “power on demand” that their residents use.

Of course the grid can make “power on demand” from a mix of wind, solar, and fossil fuels, but as LBNL-5445E shows, wind and solar rapidly become less economical as the combined pentration approaches that of fossil fuels.

Amory Lovins (who’s RMI is funded in part with fossil fuel money) is constantly making predictions that don’t come true, using his own analysis that are designed to support his ideology.  Whenever the big studies get done by NREL or LBNL, the results always show costs rising rapidly as renewable penetration increases.  It is not fashionable today for the national labs to study the economics of large scale nuclear power, but real-world experience shows that phasing-out nuclear power always makes power costs go up:  eg Japan, Vermont, California.

Clayton Handleman's picture
Clayton Handleman on May 10, 2015

“with no depence on people changing their behaviors”

This is a point on which you and I differ significantly.  Some people would enthusiastically change their behavior particularly if they got a two fer – cleaner power and lower cost. 

Now that the means (real time pricing and TOU metering) are available to monetize electricity, I think people and businesses should pay the cost based upon supply and demand.  People make arguments about this being hard on the poor.  Poor or not I like the idea of people being able to pay for what we are using instead of averaging in the peak rates.  If rich people want to drive up rates leaving their AC on why should the poor subsidize that.  Or for that matter, why should anyone be forced to subsidize it?  

 

“I’m just saying that there is an extra barrier to connecting the Western and Eastern grids together, because they aren’t connected yet, and any attempt to do so must be very large scale.”

Tres Amigas, if built, will be less than the cost of a nuclear power plant for the East-West connection and they are assuming the market risk.  They are currently having difficulty raising the funding and I look forward to researching it further.  However in answer to your statment made without support, they provide a ballpark of $500M for the first East-West Node.  While “very large scale” is somewhat subjective I think you have set the bar at 1 nuclear plant being reasonable and this is of that scale. 

The point is that it is far more granular than your handwaiving is acknowledging.  This can be built incrementally. 

Nathan Wilson's picture
Nathan Wilson on May 11, 2015

Oops, I meant to say, “No, I’m not saying HVDC can’t be done.”

Bruce McFarling's picture
Bruce McFarling on May 11, 2015

No, I’m saying HVDC can’t be done.  Within an already connected region, HVDC can be used to boost capacity. I’m just saying that there is an extra barrier to connecting the Western and Eastern grids together, because they aren’t connected yet, and any attempt to do so must be very large scale.”

But that is confusing one application of AC-DC converters with the technological capabilities of HV-DC converters. HVDC could be USED to connect the Western and Eastern grids, but that does not mean that connecting the Western and Eastern grids is a pre-requisite to having HVDC run between them. If there was a commercial case for a 1GW cross-haul line running from the Pacific Northwest to Chicago, tying into the BPA, MISO and PJM systems, the fact that one is in the Western grid and two in the Eastern grid would not be a problem … the AC-DC/DC-AC conversion isolates the AC characteristics on each side of the conversion, and it does so whether the conversion happens at a single “bridge” or thousands of miles apart.

The economic difference with the HVDC cross-haul transmission vs an AC-DC/DC-AC bridge is that the savings per mile over that distance over HVAC transmission over that distance more than covers the cost of the AC-DC / DC-AC conversion (both economically and in terms of total power losses in operation), so the AC voltage/frequency isolation happens “for free” in selecting the least cost means of moving power over that distance.

Clayton Handleman's picture
Clayton Handleman on May 11, 2015

You usually can go back and edit in a correction. 

Bruce McFarling's picture
Bruce McFarling on May 11, 2015

The high cost of energy storage relative to other sources of flexibility, including those on the existing power system, is the chief reason why it is not more widely used today. As shown in the National Renewable Energy Laboratory chart below, improved grid operations are the low-hanging fruit for making the power system more flexible. These reforms more than pay for themselves by allowing more efficient power system operations, and are more than sufficient to accommodate even very high levels of wind energy.”

While the chart is only illustrative (“Relative costs are illustrative, as actual costs are system dependent”), it does make the point that PHS is often competitive with NGCC and cheaper than coal ramping, despite the fact that an absence of broad carbon prices in line with the external costs of NGCC and coal means that when the “economic cost” of NGCC and coal ramping refers to internal cost, it is substantially less than the actual economic cost.

However, in other cases, energy storage has been found to actually provide negative value for the integration of wind energy, even if the energy storage was provided at no cost. “

As noted by Engineer-Poet, that is not what the source says. Energy storage can provide A negative value for integration of wind-energy, in reducing the value of capacity addition, but the capacity benefit is only a secondary revenue source for wind, which remains primarily an energy resource even for the higher quality wind resource regions that offer some capacity benefits. And when the the flexibility benefits provided by energy storage to wind as an energy resource, and the carbon costs of NG and coal ramping as a source of flexibility are internalized, the most economic energy storage options in that table provide a net benefit to the integration of wind.

One problem with a “Lego building block” approach to analysis of the economics of PHS in the integration of wind is that it all-too-often involves the AVERAGE configuration of installed PHS which are optimized for a particular type of use. However, hydro resources (both reservoir and PHS) have substantial capacity factor design flexibility, and can trade off between lower cost per kWh stored and lower cost per kW capacity. Each of the flexibility challenges of wind power cited in the NREL work linked to (p. 2):

“Wind and solar generation can create the need for more flexibility. The figure illustrates how wind generation can lead to steeper ramps, deeper turn downs, and shorter peaks in system operations.

1. Ramps – the rate of increase or decrease in dispatchable generation to follow changes in demand. Ramps can be steep if wind generation is decreasing at the same time that demand rises.

2. Turn-downs – operation of dispatchable generators at low levels. High wind output during periods of low demand creates a need for generators that can turn down output to low levels but remain available to rise again quickly.

3. Shorter peaks – periods where generation is supplied at a higher level. Peaks are shorter in duration, resulting in fewer operating hours for conventional plants, affecting cost recovery and longterm security of supply”

… are factors where rather than undermining the economics of PHS, they SHIFT the economics of PHS. For a given kWh storage capacity, a longer peak, smaller ramps, and smaller turn downs rewards a smaller capacity, with PHS primarily used to time-shift power from a nightly off-peak into a daily peak period.

From bottom to top, shorter peaks rewards a larger capacity, which can sell into a larger share of that shorter peak. PHS has much lower turn-down costs than NCGG and coal ramping as a source of flexibility. And steeper ramps increases the opportunities to be paid for grid stability services BOTH when storing and when generating.

Of course, for all of those integration benefits, reservoir hydro is a cheaper option, so a cost-optimizing path would first entail building out our untapped small-reservoir-hydro capacity for use as flexibility resources, and adding PHS when the flexibility available from reservoir hydro is on track to being fully committed. 

 

Willem Post's picture
Willem Post on May 15, 2015

Mike,

You are right, storage is not required to integrate variable energy, such as wind energy, to the grid, BUT, as has been PROVEN in Ireland with 15-minute Eirgrid operating data and by Wheatley, using even more accurate date for each generator on the grid, at about 17% wind energy on the grid, the OTHER gas-fired generators are operated so inefficienctly (more Btu/kWh, more CO2/kWh), that that extra fuel and CO2 offset about 50% of what wind energy was supposed to save, as claimed by the AWEA, et al. 

What you may not yet know, the interesting development in the past few months is that Eirgrid has, after many years of obfuscation, PUBLICLY admitted that this is indeed the case, and informed the EU in Brussels of that fact.

http://judithcurry.com/2015/04/27/wind-turbines-co2-savings-and-abatemen...

Government officials and wind energy promoters, such as the EWEA, BWEA, etc., usually claim one MWh of “clean” wind energy offsets one MWh of “dirty” fossil fuel energy, which is true regarding energy, but not regarding CO2 emissions, because of the inefficient operation of the other generators on the grid due to wind energy.

Below is summary of wind energy CO2 emission reduction effectiveness versus annual wind energy percent, for various grids:

1.0 at 0% wind energy on any grid.

0.97 (my assumption) at 1.0%, New England grid.

0.70 (calculated by Dr. LePair) at 3.36%, the Netherlands grid; based on at least 10 years of actual fuel and production data.

0.706 (calculated by Dr. Udo) at 12.6%, Ireland grid; based on deficient EirGrid data.

0.526 (calculated by Wheatley) at 17%, Ireland grid; based on SEMO data of individual generators, including increased start/stop CO2 emissions, and increased capacity and hours of spinning plant CO2 emissions, i.e., better than Eirgrid data.

http://theenergycollective.com/willem-post/64492/wind-energy-reduces-co2...

http://www.clepair.net/IerlandUdo.html

http://docs.wind-watch.org/BENTEK-How-Less-Became-More.pdf

http://www.clepair.net/windSchiphol.html

http://www.clepair.net/Udo-okt-e.html

http://www.clepair.net/Udo-curtail201205.html

http://www.clepair.net/statlineanalyse201208.html

http://docs.wind-watch.org/Wheatley-Ireland-CO2.pdf

Wind energy CO2 reduction effectiveness of Irish Grid = (CO2 intensity, metric ton/MWh, with wind)/(CO2 intensity with no wind).

Ireland = (0.279, 17% wind)/(0.53, no wind) = 0.526, based on SEMO data.

If 17% wind energy, promoters typically claim a 17% reduction in CO2, i.e., 83% is still left over.

If 17% wind energy, actual performance data of the Irish grid shows, 0.526 x 17% is reduced = 8.94%, id est, 91.06% is still left over.

http://theenergycollective.com/willem-post/89476/wind-energy-co2-emissio...

Nathan Wilson's picture
Nathan Wilson on May 12, 2015

“…a 1GW cross-haul line running from the Pacific Northwest to Chicago…”

I was trying to argue that you can’t economically build a 1 GW line on that route.  Due to economies of scale of transmission lines, you’ll probably find that a 5 GW lines is the minimum for economy and power efficiency, and 5GW of power is too much to subject to the single-point-failure risk of a single line, so you’ll really end-up building two or three 5 GW lines in parallel.  Once the first set are in, subsequent expansion can be one line at a time, and can be in larger line sizes too, while maintaining redundancy.

The paperwork for the Tres Amigas superswitch says each of the three terminals will carry  5 GW at 765 kV.  Internally, it’s made from 0.75 GW modules, and the internal wiring is underground, so maybe they will claim to be fault tolerant.  But I suspect that each of the three connection will be fed from two separate transmission lines for redundancy.  Note that Tres Amigas is not planning to build the connection from their facility into the main grids, rather they’ll let the grid operators do that.

Mark Heslep's picture
Mark Heslep on May 14, 2015

As the NREL wind resource map indicates, the midwest enjoys the best onshore wind resources in the US. The power region MISO is large and covers much of the midwest,  including the majority land area of 10 states and Canadian Manitoba.  MISO should then be illustrative of wind performance.

In 2012, MISO reported installed wind capacity of 11.8 GW. MISO archives indicate the following for wind generation in 2012, for non-contiguous time periods:

  • 287 hours < 500 MW (4%)
  • 17 hrs < 100 MW (0.8%)
  • 1 hr < 10 MW

Also:

  • Throughout the year, instananeous power often swings some through some 2/3 of installed capacity over ~30 hour periods, cycling through periods of several days. 
  • The period from July 29 through Aug 4th averaged 14% of wind capacity, with repeated half-day periods less than 5%.  Inspection of other MISO years also show low generation for the period of July through September.

As the midwest contains the dominant US wind resource, and the wind outage is seasonal,  it is unclear how a dramatic expansion of the transmission network (at $7 million/GW-mile) can significantly alter the low-side percentages experience by MISO.  As it is,  MISO wind capacity requires 96% conventional power (thermal and hydro) backup. 

 

Paul O's picture
Paul O on May 20, 2015

You should read this before continuing : http://www.txses.org/solar/content/solar-photovoltaic-end-life

Nathan Wilson's picture
Nathan Wilson on May 20, 2015

Correction, “used nuclear fuel has never hurt anyone”.  

I noticed that all of your supposed example of “nuclear waste” hurting people are refering not to spent nuclear fuel in its raw form, but instead to concentrated highly radioactive isotopes, which were used not as waste, but valuable and useful materials, which caused harm only in the most extreme circumstances (e.g. the polonium-210 used in the Litvinenko case as an injectable poison, the caesium-137 in the Goiânia case was used as a glowing body paint and sandwich topping, the Tokaimura accident in which a dangerously large amount of uranium, 7 times the licensed amount, enriched to four times the normal level for LWRs, was added to tank, went critical, killing two workers, and scaring many others).

These are not supportive of your claims of “incredibly dangerous” compared to common industrial materials.  Although I will admit that spent nuclear fuel is probably a similar hazard to “incredibly dangerous” household substances like bleach, drain cleaner, and propane.

I do agree with your assessment that nuclear waste is much less dangerous that fossil fuel.

 

Bruce McFarling's picture
Bruce McFarling on May 23, 2015

Wind sources are incorrigibly random walk streams. Adding them together does NOT reduce their randomness.”

It depends on what you mean by “reduce their randomness”. If you mean the magnitude of expected fluctuations of a given frequency, yes, adding them together reduces their “randomness”.

After all, you may model windpower in your mind is an incorrigible random walks, but in the real world, weather systems across North America tend to move west to east, so day by day fluctuations from distinct wind resources that lie east and west of each other tend to have negative correlation.

And at the micro level, much of the variability of windpower available at a given windplant on a five minute at a time frequency is offset by negative correlation at that frequency with other windplants in the same windfarm, because they represent gusts within a windmass moving at a current average rate of speed.

That is, after subtracting the predictable component of available windpower at the windplant level, there is a random component. But a substantial part of the randomness is where in the wind column the strongest wind will occur, not whether there will be an area with stronger wind in the wind column, and so the output of the windfarm is less variable than the total variability of the output of each windplant in the windfarm.


Bruce McFarling's picture
Bruce McFarling on May 23, 2015

I was trying to argue that you can’t economically build a 1 GW line on that route.  Due to economies of scale of transmission lines, you’ll probably find that a 5 GW lines is the minimum for economy and power efficiency, and 5GW of power is too much to subject to the single-point-failure risk of a single line, so you’ll really end-up building two or three 5 GW lines in parallel.”

The rail corridor there would have a power demand on the order of 1GW, so that is the scale that can be financed by the electricification of the rail corridor … additional capacity would have to be financed on some other basis. However, the power efficiency of the AC-DC conversion does not have much scale economy to it at the 1GW+ level. The main commercial economies of scale are in corridor acquisition costs, and since the transmission corridor can fit within the transportation ROW for a compacted 1GW VSC bipole system, the corridor acquisition costs are not substantial.

As far as single-point-failure risks, the 1GW transmission would be balanced symmetric bipole, so in the event of a line break it could be operated for 0.5GW monopole transmission corridor with ground return until the line break is repaired.

 

Bruce McFarling's picture
Bruce McFarling on May 23, 2015

The difference between the use of pumped storage to store wind energy versus nuclear energy is that …”

… since no practical all-renewable energy scenario would be based on 100% wind, analysis of the needs of PHS in order to serve a 100% wind energy scenario are exercises in over-stating the storage cost for wind energy within a 100% renewable scenario.

Clayton Handleman's picture
Clayton Handleman on Jun 5, 2016

Solar is summer peaking and coincident with peak load as well. Some argue that solar drops off in the early evening when AC is still on. However if time of use metering is utilized power at that time will command a premium. The result is that solar arrays will be built facing westerly. They will produce less total energy but more at the time it is most needed. There also are a variety of load shift and conservation opportunities that can work well with wind. Dishwashers with timers are available and will be more available if it is requirement for energy star rating. LEDs reduce demand for electricity in early evening both because less energy us used for lighting AND because less energy is required from the AC since it does not need to pump as much heat out of the building.

Coastal wind – both Texas coastal and Eastern off shore are largely decorrelated with great plains wind. While off shore wind is more costly, if we have TOU metering and put a price on carbon its economics start to look much more compelling.

Clayton Handleman's picture
Clayton Handleman on Jun 5, 2016

Hmmm, no links. This site has a good deal of information and referenced sources that clearly debunk your claim.

Bob Meinetz's picture
Bob Meinetz on Jun 5, 2016

Hmmm, a WordPress renewables site with 12 links – to itself? This site clearly debunks your claim.

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »