Solar Versus Wind Power: Which Has The Most Stable Power Output?
by John Laumer, Philadelphia on 03. 6.08
Imagine yourself working for a major electric power utility or distributor, perhaps in the strategic planning department. Government is introducing mandates for renewable energy portfolios: a 5 to 20% required green power component is a certainty within the next few years. Both solar and wind power are technically viable choices within the service area. Which of the two technologies should the company first support or most invest in?
Restating the question, which source, wind v.s. solar, has the most even capacity, will be the least disruptive to customers, and require the least backup investment to achieve synergy with the existing business?
In part, this question can be distilled down to "Capacity Factor." Two Carnegie Mellon Electricity Industry Center researchers sampled a few solar and wind power facilities over period of two years to document the capacity factor differences. Look at what they found:
"The power spectral density of the output of utility-scale wind farms and solar photovoltaic (PV) arrays is examined to provide information on the character of fluctuations in real power output; the power spectrum constrains the character of fill-in power. Both one second and one hour samples from several wind farms and ten second and one minute resolution data from four solar PV arrays are analyzed. The measured output power for wind follows a Kolmogorov spectrum over more than four orders of magnitude, from 30 seconds to 2.6 days. That for PV is significantly flatter; thus fluctuations at short time scales are larger relative to those at long time scales for PV than for wind. While wind’s capacity factor varies from 32% at the sites examined to 40% at excellent sites, the capacity factor for a 4.6 MW PV array in Arizona is determined to be19% over two years."
A money quote from the full paper:
Due to the larger magnitude of solar PV power output fluctuations relative to those of wind at time scales shorter than approximately 31⁄2 hours, the costs of large scale solar PV integration are likely to be larger than those of wind.
Note:login with advance approval required to download full paper from CMEIC.
Citation & Image credit:Carnegie Mellon Electricity Industry Center Working Paper CEIC-08-04; The Spectrum of Power from Utility-Scale Wind Farms and Solar Photovoltaic Arrays, by Jay Apt and Aimee Curtright - "Fig. 3. Real power output (kW) sampled with one minute resolution for a 4.6 MW solar PV array in northeastern Arizona for one week."
this is also why XSNX is banking on amorphous silicon PV, since it provides much more stable output, even if polycrystaline PV has peak output that is a bit higher.
http://www.xsunx.com/
Lex
just put a storage mechanism on it (one of those giant rotating underground cylinders, compressed air, heat storage etc.). It may not work out to provide storage capacity like a battery... but it would like smooth things out considerably.
Best to diversify and hedge one with the other.
Am I missing something here? It's bad that Arizona gets a lot of sun in the middle of the day, because you then get a lot of electricity from it?
Since when did it become optimal to produce all of our energy from one source? How about considering the implementation of various storage media? How about imagining a large scale electrical grid (like we have) as a storage device?
Yes, if you size your power generating system according to the poorest performing moment you can model, they you will have to install an awful lot of it to meet your peak needs. Since we size mall parking lots for the day after Thanksgiving....
But a robust power supply system is generally more than one facility and one type of power generation equipment. Conventional power plants often perform poorly when running at part load, too. That's why they're brought on and off line as the demand warrants.
Also didn't notice any recognition that power demand generally varies with time of day, workdays versus weekends, etc. No sun in the middle of the night? Not such a big problem, as most folks are asleep and not using that much power.
Does the study suggest that all heating needs should be provided by electricity too? Oy!
When considering a utility investment in renewable energy several factors need to be considered - abundance, intensity and reliability of the renewable resource, expense of the investment per net unit of energy generated, maintenance and repair costs, and time out of service, the existing power supply network and composition of the installed base, conservation efforts and load shifting potential...
Solving this complex problem for a single variable is a fool's errand.
=== author's response follows ===
Not foolish when capital is limited under the cloud of a carbon cap fiscal risk.
Otherwise I agree.
English please
John + jon,
I did not get the impression that this paper was trying to determine the relative merits of wind and solar, or which technology would be prioritized in a utility's capital investment decision-- at least from the excerpt quoted.
Instead, they are addressing the technical issue of how these two intermittent generation resources behave from moment to moment-- they are describing the character of each respective technology's "intermittent-ness". (Should be a word!)
The concluding quote above could be paraphrased as, "solar PV tends to have wider swings in power output than wind units in shorter time periods." The implication is that the backup or balancing generators on the grid will have to react more quickly when backing up solar PV than when backing up wind. (Balancing units are necessary because total power supplied to the grid must be maintained at a certain level-- demand-- at a given moment, even if e.g. a unit goes down.)
So in English translation, a grid operator has to pay a lot more attention to a solar PV array on a moment-to-moment basis than to a wind array of similar size, and has to react more sharply-- because a solar PV unit will tend to go up or down more quickly than a wind unit.
This is just a technical grid stability issue, and has nothing to do with overall supply, demand, which technology is "better" for a given region, etc. They're just saying that integrating PV to the grid is not the same as integrating wind.
BTW I am a graduate of Carnegie Mellon and studied at the EIC. Go team!
=== author's response follows ===
yes...but paying more attention means more effort and that costs money.
>
Right now wind is cheaper.
John writes: yes...but paying more attention means more effort and that costs money.
John, fine-grained grid stability issues are managed automatically as a general rule. Every generator's output will fluctuate to a degree, and of course demand will as well.
I think you are overstating the cost effect of solar PV's greater variability. Grid managers are constantly moving output up and down to rebalance load moment-to-moment; this study is simply describing the way in which solar PV output will behave.
Any cost difference arising from the differing backup power requirements of solar PV and wind will be swamped by the differences in capital cost or site appropriateness.
Each option is cheaper than a trillion dollar war for oil.Decentralized power is always the best due to line loss and loss at the generating station.And only solar or wind can do this.
The output fluctuations on small time-scales of solar energy systems are caused by passing clouds and are easily mitigated: just build several installations at some distance from each other: the fluctuations will average out. Since PV is the extremly suitable for distributed generation, this should not be a problem.
By the way:
The capacity factor is the actual annual yield divided by the yield the system would have had if it produced at maximum capacity all the time. This does not necessarily say something about intermittency or reliability: A power plant that works constantly at 50% of its capacity is not-intermittent, but will have a capacity factor of only 50%. A powerplant that works at full power, but goes off-line for 5 minutes at some random timepoint in every hour has a capacity factor of 92%, but is much less reliable
Maybe the word you are looking for is capacity credit: the amount of conventional base load power that can be left out due to the alternative power producer.
@jon - the article was not talking about replacing 100% of power generation with PV / Wind. The article states that the utliity is having to consider implementing "a 5 to 20% required green power component" as part of it's energy portfolio. In a 100% replacement of just wind and solar, yes you would need to combine the two. In a 5 to 20% replacement, one or the other is adequate.
With that being said, friend in the business appears to have a good grasp on the concept presented above - the need to examine the moment by moment fluctuations in output. No need to repeat the rest of what he said about backup generation and peak demand, etc.
While I think it is cool that "wind follows a Kolmogorov spectrum", I am unsure that I understood how the conclusions were made.
Also, it sure would be helpful to have similar data for solar thermoelectric.
I agree with Chris...
Capacity Credit is the bottom line here.
To what extent does Wind or Solar efectively displace Conventional Power Stations.
At least in the case of Wind Power It's not a direct 1-1 relationship, although Wind Power people will like us to think so. This will be a rare ocurrance.
Who is asking the key questions?
Not many will dare question the greeness of Wind Farms nor Solar Panels, but it is not about an agenda to thrash one technology or the other is about qualifying the TRUE benefits and then making decisions based ON REALITY.
Energy savings far outperform any of the above.
Should we invest in Wind and Solar then? Yes but its development must be smart and based on reality.
This gets blurred with so much money running around....
Consider using Solar Power on sunny days and Wind Power on windy days. Often the sun does not shine very much on windy days here. Use both.
All of the Solar Power systems I have seen drawings for use batteries for power storage
What are the costs of a war for oil, again?!?
We are nothing but brutal cads if we only count dollars: no?