To Store or Not to Store: That is the Question
20 July 2011
Perspective by The Townsend Company
|
|
| Figure 1. Typical daily grid demand. Click to enlarge. |
“Why do you need energy storage on the grid, when the grid is full of power plants that generate all the energy that is needed?”
It would seem like a good question to ask and applies the same logic has been used for the past fifty years in determining grid strategy. However, when peak demand is not required the inefficiencies of idling coal, nuclear and gas powered power plants has become both very uneconomic and non-responsive to changes in demand for today’s market. This has led to a change in thinking of how energy generation is optimized and a new strategy has developed in which flexible energy generating sources are deployed to meet peak requirements while maintaining a steady state base load from low cost less-flexible solutions.
With the world’s growing population and the political, environmental and economic pressures in place to reduce the reliance on fossil fuels, alternative energy has become this flexible energy means. Supplementing baseload coal-, nuclear- and gas-powered power pants in a grid strategy for tomorrow. This is demonstrated in figure 1. Baseload energy is generated (coal, gas, nuclear) at a steady state and stored when not used, supplemented by flexible energy devices (hydro, solar, wind) for shoulder peak demand and stored energy for priority peak demand.
Due to the inherent capital costs and characteristics of the leading alternative energy mediums: Hydro, Wind and Solar, Solar deployment has started to emerge as the leading candidate of choice. Hydro is limited by geography and the building of dams is very time and capital intensive, Wind generates most of its energy at night, during off peak demand and is drawing more visual pollution criticism. In contrast, solar provides energy during peak periods, is mostly unobtrusive when installed and is becoming more economical with technology and scale.
|
|
| Figure 2. Solar field annual output. Click to enlarge. |
The largest disadvantage with both wind and solar is, however, the issue of what happens when it is not sunny or the wind is not blowing. For example, in June, California experienced an 1800MW drop in electricity production in one hour due to wind fluctuations. The variable output from a solar field is shown in Fig 2. The need for energy storage should start to seem obvious.
Whenever energy storage is mentioned the majority of us think batteries, however from a United States grid perspective, pumped hydro is the incumbent technology, followed by compressed air (CAES) and then the relativity new grid storage concepts of flywheels and electrochemical batteries. This order of preference is also representative of the capital cost, high to low, and the flexibility of deployment low to high—which is the fundamental driver for the growth of large format batteries on the grid. Although the largest potential benefit for storage on the grid is peak load shifting, other benefits are also realized in the form of: load firming, voltage support, frequency regulation, ramp management, community storage and security of the grid.
The United States is, however, faced with a dilemma: the Federally structured markets of generation, transmission and distribution do not allow for the recognition and compensation of storage placed on the grid. This leads to its current limited use in frequency regulation (power plant hybridization) and pilot projects, and requires either bold market steps from independent power producers or a change in the Federal market structure for the majority of the market to embrace its deployment.
With the implementation of the renewable energy portfolio standards, the penetration of wind and solar energy generation will only increase along with the variability it places on the grid. Although not an issue today, it is expected that once these renewables reach 20-30% of energy generated on the grid, the grid will not be able to guarantee a continuous supply of power across all its spectrums of demand without storage—placing the United States in its current stand-off situation between the Federal structure of the markets, renewable implementation goals and the power producers that generate energy for the grid.
Other countries are not faced with this Federal dilemma; as such the United States is lagging behind the rest of the world in developing a flexible grid structure, only having the capability today of storing 2.2% of its energy compared with the average of 4-5% for other developed countries.
For example, the State of Abu Dhabi is implementing a 1.35GW storage system with the objective of peak load shifting. Mexico is adding a 1GW energy warehouse to its grid to improve its reliability. India is deploying an energy storage plan to address the 1.2% loss in its GDP due to unreliable power. China, the largest growth market for solar and wind, has made energy storage a fundamental part of its five year renewable energy plan. Even in the mature German market, the use of energy storage is regarded as strategic enabler to the decommissioning of its nuclear footprint, in favor of increased use of renewable energy means.
This changing grid architecture represents an incredible opportunity in the market, not only from the direct sales opportunities associated with renewable energy and energy storage, but also the systems that make them more efficient—systems that incorporate the electric fleet onto it; the market for secondary-life batteries; and the potential to incorporate vehicle fleets into grid storage solutions.
About The Townsend Company. The Townsend Company l.l.c. is a consultation and business development practice formed to help companies wanting to enter, grow and become more profitable in the alternative energy market.
I don't understand Figure 2 "Solar field annual output".
How can you have solar power at 1am ?
It looks more like a wind power graph.
It is realistic to do peak shifting on a daily model as shown in Fig 1.
It is realistic to do load shaping to reduce the effect of renewables ramping up or down quickly.
It is not realistic to store energy from day to day, i.e. to store (say) wind energy across a winter inversion (When there is no wind and it is very cold).
Hence, you have to keep more or less all your fossil plant, even if you only use it a few times a year.
Which is expensive and a bit galling.
Posted by: mahonj | 20 July 2011 at 10:18 AM
They should look at the audi e-gas and hydrogen grid energy storage project that have been presented here a month ago approx.
Posted by: A D | 20 July 2011 at 11:09 AM
Right with you mahonj on the solar chart - does not compute.
If you head to the CAISO site and look at their public data - one will find that the published Solar production data is actually VERY consistent on a day to day basis. 6 of the last 7 days have peaked right around 450 kW with the 7th peaking at 416 kW. Total daily production is 5 GWh +- 10%.
I'm sure that this is a result of the data primarily including solar thermal plants out in the desert - but that shows the value of siting these types of plants out there.
Wind has been less consistent, but appears to be fairly predictable.
Posted by: Dave R | 20 July 2011 at 11:11 AM
"Hydro is limited by geography and the building of dams is very time and capital intensive, Wind generates most of its energy at night, during off peak demand and is drawing more visual pollution criticism.
Interesting statements favoring solar.
Posted by: Reel$$ | 20 July 2011 at 11:27 AM
The most flexible, short lead-time power source is hydro. They can bring hundreds of megawatts of capacity online in seconds. Combined cycle natural gas is also pretty flexible for peaking.
I think we have to distinguish between power conditioning (which could be done with large banks of capacitors) for fluctuations in wind and solar from second to second, versus average production of variable renewable sources.
I expect good average predictability across large aggregates of renwable power (with good weather reports and runrate data, they can predict reasonably accurately what solar and wind will do on average for a given day). Natural gas can be used to tune the average output across the daily demand curve.
In winter doldrums, yes, we'd expect more base power...but it only takes days to bring a coal plant online, not months. If we add renewable capacity, but don't subtract coal, we still have the base load for down seasons.
Posted by: HealthyBreeze | 20 July 2011 at 12:14 PM
It's also possible efficiency measures (lighting, heating, cooling, appliances) could be targetted at peak applications to reduce the impacts of population growth.
Posted by: HealthyBreeze | 20 July 2011 at 12:17 PM
You should not use Solar or Wind energy sources for peak demands because they may NOT be producing when needed. Those two should be used for main or primary loads in order to use 100% of their production capability.
Hydro is ideal for peak loads because it can be made available quickly and can be turned off when not required and accumulate water (store energy) in the existing water reservoirs.
NG/SG power plants can also play that role but demand more time to be turned ON to fully production. Those plants are good to meet well known regular peak demands.
Posted by: HarveyD | 20 July 2011 at 12:38 PM
“Why do you need energy storage on the grid, when the grid is full of power plants that generate all the energy that is needed?”
Because of the Enrons of the world. Enron was able to yield its influence to manufacture power outages & manipulate energy prices for their benefit. Even in a highly regulated market, there are still issues with the grid from time to time.
Posted by: ejj | 20 July 2011 at 07:21 PM
MARCH 29, 2011
"California lawmakers passed a bill Tuesday that would require a third of the state's power to come from renewable energy sources by 2020" - WSJ
It now stands at about 20%, this will encourage more.
Posted by: SJC | 20 July 2011 at 07:36 PM
Hey folks, I never hear any comments regarding the energy DELIVERY system - the power grid. I have read that as much as 50% of the generated power is lost before it gets to the load. I don't think a smart grid can help this loss much, but decent storage systems could minimize some of the losses by distributing the 'sources' more effectively.
Seems like reducing this loss is easier than building new power plants!
Does anyone know where these losses occur? I see 'high tension' lines, switch gear, transformers with cooling fins and fans, etc. all of which is 'infrastructure'.
Those fins and fans tell me we loose power to heat in the switch yards all over the place.
Posted by: william g irwin | 21 July 2011 at 04:27 AM
Energy loses in transmission are about 7%, so don't sweat it.
Posted by: mahonj | 21 July 2011 at 05:17 AM
There is talk of a Smart(er) grid so that power plants are used optimally. There is talk of grid stabilization and power to help use renewable energy that is not 24/7. We have the ideas and some are proven, it will remain to be seen if we have the investment.
Posted by: SJC | 21 July 2011 at 06:26 AM
@ HarveyD
Actually, peak heating and cooling demands should be highly correlated with sunny days. Certainly, record hot or cold days are invariably sunny. There is also a correlation between wind and a building's heat loss or penetration.
Heat and cold can also be readily stored for near term usage as steam/hot water and ice/cold water, respectively.
Posted by: NorthernPiker | 21 July 2011 at 08:03 AM
Solar and wind power are not renewable. Please stop calling them that.
Posted by: mph | 21 July 2011 at 08:47 AM
mph...where have you been in the last 24 hours?
Posted by: HarveyD | 21 July 2011 at 10:58 AM
NP....cold nights are not very sunny.
Posted by: HarveyD | 21 July 2011 at 10:58 AM
william irwin makes a point about overall energy losses in the grid. While transmission losses hover around 7-10%, the overall system losses are closer to 50% or more depending on the generating system.
http://ehsmanager.blogspot.com/2010/04/60-of-energy-lost-in-power-grid-maybe.html
All of these issues fade dramatically with the introduction of distributed energy produced by CHP units in the home. Most homes will operate fully and efficiently with a 2.5-5kWh CHP system. Combined with local micro-grids, these systems eliminate transmission towers, transformers, miles of wire, convertors, stabilizers, insulators, inverters, conditioning etc, etc. etc.
Why keep patching up a distribution technology that is a century old and no longer practical? We have MUCH better ways of generating local energy and using it. Remember, think global, act local?
Posted by: Reel$$ | 21 July 2011 at 12:52 PM
HD....night-time is not a peak load period for the grid.
Posted by: NorthernPiker | 21 July 2011 at 02:19 PM
Oh, my. Reel trolls again, folding conversion losses in powerplants into line losses in the grid. Either that, or he's just clueless.
Posted by: Engineer-Poet | 21 July 2011 at 07:32 PM
Thanks Reel$$. That means the distribution system is not the big culprit as was implied. I agree that distribution efficiency can be improved with peak offloading at the local level.
This may seem over simplistic to some, but why couldn't a solar and/or wind farm pump water up hill into places like Lake Mead and stand totally separate from the grid? Then the peak matching and power conversions all go away. Use the juice when its available!
Posted by: william g irwin | 22 July 2011 at 04:20 AM
EP
Quit being such a pain, you are not helping.
Posted by: SJC | 22 July 2011 at 08:03 AM
HarveyD...so you think that the sun is renewing its energy when you can't see it?
Posted by: mph | 22 July 2011 at 08:06 AM
@will
There's no reason why they couldn't, in fact that is what they do in many places.
Posted by: ai_vin | 22 July 2011 at 08:09 AM
mph....the sun is as close to sustainable energy source as you can get. It's been at it for billions/years and will probably be there a few more billion years. What more to you want.
Posted by: HarveyD | 22 July 2011 at 08:29 AM
N-P...one our our highest power demand peak period is from sunset (about 16h) to 19h in winter time because of street/parking-security lights, and because 80%+ use electric heating and e-appliances. On arrival from schools and work places, people start the electric home heating system, e-cooking, e-washing, e-dryers, e-hot water heaters etc and create a major peak demand Mondays to Fridays. It would not be the right time to charge the EVs.
However, demand goes down after 20h and even more so after 22h. That would be the right time to start charging EVs.
Posted by: HarveyD | 22 July 2011 at 08:40 AM