Wednesday, January 20, 2016

Renewables at 90% only IF grid-level storage comes first

Renewables can't be a 90% unless we have grid-level storage

Renewable installations are growing fast, but we can't have huge percentages of wind and solar on the grid unless we first have grid-level storage.  With the exception of pumped storage, grid-level storage does not exist at this time.  (No, I don't count an occasional 2 MW project as "the answer.")

We can't grow wind and solar to a higher percentage on the grid than their capacity factor implies, unless we have storage. Moving to 50% wind on the grid is not possible, without utility level storage.  

The explanation follows, based on the New England grid.

A year on the New England grid (2008)
Shows the necessity of being able to dispatch electricity
Note rise in gas usage while nuclear plants refuel
Click to enlarge
Five million down and five million up (natural gas takes the lead)

In a guest post on January 7, Michael Twomey of Entergy used grid operator (ISO-NE) data to show that between 2014 (when Vermont Yankee was running) and 2015 (when Vermont Yankee was closed), nuclear kilowatt-hours decreased by about five million MWh and gas-fired generators increased their output by almost exactly the same amount.  Natural gas went from 46,200,000 MWh to nearly 51,900,000 MWh.  Nuclear kWh nuclear went down by almost the same amount of MWh.
Click to enlarge table

Wind and solar growing fast: Jeff Schmidt's comment

Looking at the table above, you can see that wind went from 1,892,000 MWh to 2,135,00 MWh, growing by approximately 243,00 MWh, and solar grew from 327,500 to 436,200 MWh, approximately 108,700 MWh.

This rapid rate of growth (though still only adding up to 2.4% of the power on the grid), prompted Jeff Schmidt to write this comment on the Twomey article:

"This article seems to dismiss the growth of wind and solar. While I am pro-nuclear, and think that nuclear needs to play a vital role in our future energy mix, I think the author of the article is neglecting something important - growth of wind and solar.

It's true that they are still small. But, if you look at the year-over-year growth rate, as shown by the statistics provided by the ISO and called out by Mr. Twomey, we see that Solar grew 33% in a year, and Wind grew 41%. Of course, one can't predict future growth rates based on one year, but IF wind and solar can keep up strong growth like that, they could conceivably become a very large proportion of the New England energy mix inside of 10 years.

It's true that it's likely an overly optimistic and simplistic projection, but just for the sake of argument, if they can keep up that growth rate, then 9 years from now, Wind could produce about 50% of the energy, and solar about 5%. If you projected it to 10 years instead of 9, that would account for more than 100% of current grid generation.

However, at the same time, it's very likely that at some point, Wind and Solar's growth must slow. Still, it's a valid point to concede that Wind and Solar, while currently small in absolute terms, are actually growing at a pretty fast rate."

Why renewables can't keep growing--unless we have storage

I wrote the following response to Schmidt.  I oversimplified, but I am also worried that "we can't grow wind and solar" arguments are often based on cost, or on complex technical issues that are hard to explain.  So, here's my oversimplification.  Basically correct, but oversimplified.

Basically, we can't grow wind and solar to a higher percentage on the grid than their capacity factor implies, unless we have storage. Moving to 50% wind on the grid is not possible, without utility level storage.

I wrote about Vermont's plans to be 90% renewables in today's blog post. Of course, renewable growth from 1 to 3 to 5% is possible and looks great. However, it simply does not scale. Let's oversimplify a little, though not a lot.

Most of Vermont is one weather pattern, with some exceptions. Hot, dry and sunny...all over Vermont. Windy at night...all over Vermont. Cold and windless....all over Vermont. Now, obviously, the mountains are different from the river valleys and so forth, but the statement "weather is the same all over Vermont" is far closer to true than its opposite would be.

Okay. We cannot turn wind on and off. Let's say that wind has a 30% capacity factor. For wind to grow to 30% of the electricity supply overall, that means when wind is on the grid (the wind is blowing in Vermont)...the grid has to be 100% wind. Without this high percentage when wind is available, wind is not going to be able to be 30% of the electricity, overall. So we have to build a lot of wind to get wind to 30% of the electricity supply, and we have to turn everything else off if the wind is blowing.

Well, what if we build more wind? If we do that, when the wind is blowing....what then? We have to curtail some of the wind, because the grid can't take more than 100% of wind. So, without grid level storage, wind reaches a VERY hard stop at 30%.

Well, it is windier in the mountains, and the southern part of the state gets less wind and so forth and this is an oversimplification. And the grid requires more power in the day, and less in the night (when the wind usually blows). So it is quite complicated in reality. But the basics remain.

IF you can turn things on and off, you don't reach this sort of hard stop. 100% of the electricity from natural gas...this could work. No "hard stop" involved. 100% from nuclear...well, current nuclear doesn't follow load well, but there is no "hard stop" involved, where you have more nuclear than you can use on the grid. You don't need grid level storage for nuclear, just plants that follow load a little better. And so forth.

This is why I am so cynical about the Vermont energy plan. The plan is kind of "We don't just hope for miracles, we expect them."

Jeff Schmidt has two guest posts at this blog:

The Nuclear Safety Paradox, which describes how experience (such as building new nuclear plants) increases safety.

Flawed Analogies, which describes the analogies nuclear opponents made in a debate against nuclear energy.


The illustration showing the need for dispatchable power
From Sustainability presentation by David Lamont
Vermont Department of Public Service
October 18, 2010
Presentation is no longer on the web, but I had saved it to my computer.


Anonymous said...

Your analysis seems to assume that the wind is either blowing, or not; that a given wind machine is either running at full capacity, or not. Is that really how the turbines work? If the wind is blowing at "half speed" can't the turbine make "half power"? In that case, building say, twice as many turbines would let you produce twice as much power on the "wind at half" days.

Meredith Angwin said...

Excellent question!

Indeed, the turbines make less power on less windy days. So if you are willing to build more turbines to fill out when the "original" turbines are at half speed, you can add a little more wind to the grid. The new turbines will get very little utilization, however, because the 30% capacity factor for the "original' turbines already includes the times the turbines are going slowly, as well as the times they are going full out. So the extra added turbines would be waiting around for only a limited number of wind conditions. They would have very low capacity factors.

I oversimplified in the post, and I said that frequently in the post. In general, you would not build the extra turbines that would have such limited usefulness.

Meredith Angwin said...

Basically, you can't turn the wind on or off. So backing up wind with more wind would lower the capacity factor of all of the wind. You can think of the extra turbines as sometimes useful, but often "curtailed." If the wind is blowing strongly enough to be a high percentage of the grid...the new turbines cannot run In contrast, we don't "curtail" a gas turbine--we can turn it on when it is helpful, and turn it off when it is not needed. We can use gas when we need it, not just when it is available, like wind.

The grid can't take more than 100% of what it needs (unless it has utility-level storage), so when the wind blows well, the new turbines won't be participating. Or if they do participate, the main turbines will have to back off. The financial viability of a wind turbine depends partially on its capacity factor, which is why they put them on ridge lines. So backing up wind with wind is not practical. It adds SOME more wind to the grid but causes ALL the wind to have lower capacity factors. As I said, the new turbines would be waiting around only for particular weather conditions, in order to be useful.

Engineer-Poet said...

If you look at time plots of wind generation in Texas, the edges of the curves are very steep.  It really does look like the wind is essentially "on or off", where the "on" value is maybe 80% of the peak of the current episode.

Joris van Dorp said...

Great article and topic.

I'd like to weigh in.

Another way of looking at this problem is to look at the effect of large amounts of wind (or solar) on the price of electricity. This price will plummet whenever wind or solar is oversupplying the market. Prices can then even dip below zero, in which case consumers actually get paid to take the electricity. (this already happens in Germany during periods of high wind/solar, and/or in the weekend when demand is low)

While low prices may seem nice, they are not, because it means that electricity producers all lose money, meaning they will go bankrupt sooner or later. Also, no new capacity will be built as long as the broken market conditions persists. This is the situation now in Europe in and around Germany. All producers are loosing money due to the broken electricity market, so all are clamoring for more subsidies. It's not just the wind/solar suppliers anymore, but everybody, including fossil fuel generation.

This is what makes claims of some RE advocates that "wind and solar cause electricity prices to decline" so misleading.

Put another way: there is a difference between the *price* of electricity (set by the law of supply and demand) and the *cost* of electricity (set by the fixed and variable cost of all the installations and workers generating and distributing the electricity). Adding large amounts of solar or wind pushes down the *price* of electricity, but increases the *cost* of electricity. This situation is unsustainable. Either subsidies keep increasing to cover the increasing difference between price and cost of electricity (German market model), or electricity suppliers go bankrupt and black-outs occur (won't happen, politicians won't allow it for fear of getting lynched by angry mobs).

Meredith Angwin said...

Thank you all for your comments.

Engineer-poet: your note on the "spikiness" of wind energy was very helpful. I was tying myself in knots trying to say that we could only balance wind-with-wind under quite limited circumstances. You provided a factual record that supported my claim. Thank you.

Joris: thank you for the European perspective. This phenomenon exists in America also. The "price" of electricity supposedly goes down, but the "cost" to the consumer somehow goes up.

For example, wind can sell into the grid at a very low price or below zero, because wind also receives Production Tax Credits (PTCs) and wind sells Renewable Energy Credits (RECs) for several cents per kWh. So the grid price is low, but the wind turbines are collecting REC payments.

But where are those REC payments coming from? They are coming from utilities who are required by law to buy RECs. Those utilities put the cost of the REC into the bill to the consumer (of course) but meanwhile, on the grid, electricity producers are facing low "prices" for their power. So the price of power goes down (on the grid) and the cost of power goes up (to the consumer)>