How Important is Baseload Generation Capacity to U.S. Power Grids' Reliability?

John, you deserve a medal for undertaking a complex evaluation of the relative merits of baseload power vs. a power-equivalent hodgepodge.

To engineers, of course, the idea of making a complex dynamic more complicated than it needs to be – that of producing a steady supply of reliable electricity to millions across a grid – is antithetical to everything they’ve learned in school. That it’s a loser in both expense and thermodynamic efficiency is obvious.

Raising the question of why some want it to be complicated, a question best answered by federal regulators FERC and the SEC – not science.

I guess I favor a high tech grid over high tech generation.

I agree “you deserve a medal” to confront this topic. Let’s call it the Puerto Rico epidemic; advocating borrowing money for solar and wind energy while sitting on the beach is great politics. But when a hurricane hits and you can’t flush the toilet or eat a decent meal you go looking for a TV camera to demand life support.

Build your foundation solid, then play with the decorations. Infrastructure is, well, infrastructure. Fracking didn’t produce the oil boom claimed, but it did produce more gas than expected, and baseload electric power generation is rapidly getting more efficient.

We can also celebrate lower baseload electric consumption. The next infrastructure in Puerto Rico will require far less wired power to turn the lights back on.

Efforts like yours are great teaching tools. Much to learn, if we try.

John,

Your expose, using the US grid as an example, is a major contribution to sanity.
Wind and solar lack synchronous rotational inertia.
Traditional generators provide the synchronous rotational inertia that serves to stabilize the grid.
There are pro-RE folks who claim rotational inertia is not needed, because it can be artificially provided.
NG plants provide synchronous rotational inertia, plus they provide the peaking, filling-in and balancing, 24/7/365, i.e., whenever variable, intermittent wind and solar are insufficient to meet demand.

I look forward to another expose using the German grid as an example.

In 2016, German gross electricity generation was 648.4 TWh, of which 460.1 TWh was from conventional generators and 188.3 TWh was from renewables, i.e., about 188.3/648.4 = 29% of gross electricity generation was from renewable sources, such as wind, solar, hydro, bio, etc.

Of the 188.3 TWh, about 77.4 TWh was from wind, about 38.2 TWh from solar, for a total of 115.6 TWh. About 21 TWh was from hydro and 51.7 from bio, etc. On an annual basis, wind and solar (stochastic sources) was 115.6/648.4 = 17.8% of electricity generation.

That stochastic percentage is about 3 times greater than the US. If Germany did not have the export/import safety valve, it would have to engage in major wind and solar curtailments to stabilize its grid.

wind, if making either a grid or generation “high tech” makes it more complicated than it needs to be, I guess you favor a system which is more costly, less efficient, and less reliable.

Antoine de Saint-Exupéry, a philosopher who also happened to design airplanes, once said: “A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away.”

John,

Good article on a topic that desperately needs this sort of illumination.

A big confusion factor in discussions about baseload generation and whether it is “unnecessary” and “obsolete” is the ambiguity about just what “baseload generation” actually means.

Activists in the “Pure RE” camp use “baseload is unnecessary” as an attack meme against nuclear power. Nuclear is thought of as the quintessential baseload generation, so arguments against baseload become, by association, arguments against nuclear. The essential feature of “baseload”, by their definition, is inflexibility. It can’t readily be throttled to follow load, and startup and shutdown are drawn out processes that take hours to complete.

You use a different interpretation of “baseload”. As you say in the article, you take it as essentially synonymous with “dispatchable”. (I have to say “essentially”, because peaking units are dispatchable, though nobody considers them “baseload”.) You are addressing the importance of “fully dispatchable” generation for a reliable grid, not baseload generation per se (under other potential interpretations). That’s quite valid, but where you say “dispatchable / baseload”, RE activists would say “flexible / non-baseload”.

Historically, I believe “baseload generation” has been defined as much by economics as by technology. Baseload generation is generation capacity normally kept online at rated output to meet baseload demand. It’s not that it can’t be used otherwise, but that it doesn’t make economic sense. Units designated for baseload service have traditionally been the more efficient generation resources, with high capital and low fuel costs. Their marginal cost for power generated is very low.

The motivation for RE stakeholders wanting to see baseload generation reduced is simple economic interest. If there is less baseload capacity operating, there will be fewer times when high VRP availability drives wholesale electricity prices to near zero. There will be less need for curtailment, so VRP resources will sell more power at higher average prices. Of course VRP stakeholders would want that. But the reduction in baseload generation means that more energy will need to be supplied by peakers or by flexible generation units operating at reduced CFs. What’s good for VRP stakeholders isn’t necessarily good for ratepayers and the system as a whole.

Roger, your comment raises the question of what defines both baseload power and baseload demand.

Historically, Edison and other electricity generators didn’t have the luxury of maintaining stable line current with generation from different plants dynamically. When load increased lights dimmed, another boiler was brought online, and lights became bright again. That presented reliability/safety issues, but also economic ones: it didn’t take long to figure out boiling more water, then boiling less, etc. wasted coal. So a minimum number of boilers were kept online constantly, and demand variability was matched by more agile resources like natural gas.

With the inclusion of renewables it becomes necessary to divide the challenge of maintaining stable line current into two distinct functions: 1) load following, and 2) supply balancing. When wind and solar obstinately refuse to oblige consumers by ramping up during early-evening peak consumption, load must be met by burning natural gas. But when the wind dies in Altamont Pass or a cloud front rolls over Topaz Solar Farm during hot afternoons, supply must be balanced with natural gas to maintain grid stability.

The need to balance supply presents grid engineers with two new and distinct challenges: unpredictability and temporal variability. Unpredictability must be addressed by maintaining spinning reserves to balance unpredictable changes in the weather. But the temporal variability of renewable sources means the rate-of-change (derivative) of supply from renewables is faster – generation must be replaced or removed more abruptly than that necessary to meet the smooth changes in demand presented by millions of consumers spread over a grid.

In the mad rush to shut down nuclear plants and profit from sales of natural gas, utilities are more than eager to portray nuclear as flat-line, baseload energy which is out of touch with a renewables-rich grid. Conspicuously omitted is the fact carbon-free nuclear plants online in 2017 are entirely capable of load-following the gradual, predictable changes of customer demand. Take away the variable supply of renewables, and the problem goes away.

Trivia which might be of interest: Kelly Johnson, famous aeronautical engineer and head of Lockheed’s legendary Skunk Works in the 1950s/1960s, is credited with coming up with the acronym KISS, for “Keep it Simple Stupid”.

Early in his career, Johnson was the lead designer of Lockheed’s P-38 fighter – the plane in which Saint-Exupéry was killed when he crashed into the Atlantic in 1944 under mysterious circumstances.

Excellent power summary. Germany’s ‘failure’ to operate reliably and independently even at grossly increased cost should be fair warning to any green dreamers of intermittant energy waste of public funds.

John, I hope you will engage in an equally illuminating discussion of Chinese power grid experience, especially as they move forward into the 21st century with liquid salt nuclear plant designs propelled forward by US technology from the mid last century.
Thank you again!

John I do think it was a good article.

However, you never brought storage into the equation, which is rapidly becoming the transformation tool of the grid as costs continue to plummet. It has the potential to change the grid from load following and “best guess” for demand to accurate supply/demand. In doing so, it does replace “baseload” no matter how you define it.

In fact to get the 1MW, 90 day supply you can actually do it with aluminium air batteries. Which have almost no overhead costs and are instant on. You can get 6-8 kw/kg and I forgot which japanese company it was but they were saying the cost was around $1.1/kg so roughly 300M or .15/kwh, which is expensive, but you can recover the aluminium, which is the majority of the cost and it can sit idle for 10+ years which covers the complete disaster 90-day supply scenario. IE the nuclear plant meltdown, or the coal plant blew up, hurricane, NG pipeline issue, sky rocketing prices, etc.

I will say, we aren’t at the point where we have enough storage to make a difference for the entire grid, but I don’t really anticipate it taking more then 10 years before it is extremely commonplace if only for the simple reason it makes the grid easier to operate and cost effective.

Renewables aren’t going away. Grid operators already know they work, and they are cheaper. The coal industry just wants a bail out after refusing to make any improvements for ten years. The money is truly better spent elsewhere because you simply don’t make the grid more resilient.

John, the U.S. has never been a world leader at addressing carbon emissions, and there’s no evidence that would change by becoming party to the Paris Agreement. A country whose citizens generate per capita emissions equivalent to five times the world average certainly has no right to claim that distinction.

If it isn’t already, China, with less than half the U.S.’s per capita carbon emissions and 20GW of new nuclear under construction, is poised to become a world leader. But time will tell – signatures on a non-binding international agreement, like the plastic trophies handed out to all players at Little League baseball games, are meaningless tokens which place a higher value on participation than results. As such, they harm more than help.

the U.S. has never been a world leader at addressing carbon emissions, and there’s no evidence that would change by becoming party to the Paris Agreement.

Actually the US did they -best- thing they could to help emissions. They developed and scaled out a whole system other countries could replicate. The US has electric cars, utility scale battery storage, HFCs. basically you name it, we tried it. What it did was lowered the cost of alternative technologies that banks use to hand out loans which hadn’t been updated since the 70s.

You don’t have to be #1 to be a leader. simply by demonstrating cost effective alternatives and viable business models to make it work for the rest of the world, while we continue our internal bickering which will go on for another 20 years, because we already have the cheapest energy in the world. The US basically led everyone to the solutions that worked best for the other 180 countries.

The EU fell drastically short because they all picked what technology they were going to sell the rest of the world and make money on. It left huge gaps in the implementation and damn near killed off renewable energy for another 30 years.

It really didn’t cost the US any money. At the end of the day, the investments by all the countries have reduced the price of oil, which saved the US more money then it spent on the whole program and far more emissions then the US could have done by itself.

At the end of the day, the Paris agreement was a terrible deal for the US, but it didn’t have to be good. we just had to get countries especially developing nations like India and China to the table and agree to do something. However, the US has already massively reduced emissions and is on track to achieve the stated goals.

Which is not much challenge any more on financial side….

This actually works more to promote nuclear than than the unreliables.  The round-trip efficiency of aluminum-air is very low (20% max), but production is key.  The electrolysis cells where aluminum is made must be kept in the molten state.  if you have excess power every night you can keep electrolysis cells hot much more easily than if you have extended slack periods.  A big, fat dump load takes away the whole issue of “inflexible” baseload generation as a weakness and makes it into a strength.

John, though U.S. residential electricity sales are up 4%, and commercial up 7% since 2005, industrial sales are down by 9% due to manufacturing moving offshore. So before giving fracking credit for reduced emissions, we should take into account about half of U.S. reductions aren’t reductions at all but emissions outsourced to Southeast Asia.

Moreover, since 2005 the U.S. has tripled LNG exports. How can a leader at extracting and selling fossil fuel burned elsewhere be considered a leader in reducing emissions?

This article follows exactly the lino of argumentation a it was presented by the german utilities in the late 1980’a and early 1990’s, culimninating in ths well known advertisment in all mayor newspapers at that times:

We all know the reality today.

Something that just struck me:  even at 20% round-trip efficiency, Al-air batteries remain attractive if you can get electricity for 2¢/kWh (interruptible rate?) and fixed costs are low.  This goes especially for EVs, as the whole issue of range anxiety goes away if all you have to do is start with fresh aluminum plates.

How do the generators make money in this regime?  Arbitrage and using interruptible loads as spinning reserve.  If you have a large chunk of demand which can respond in half a cycle (meaning 1/6 cycle in practice as the 3 phases are 120 degrees apart and thyristors cut off at the zero crossings), you get amazingly fine control over the supply/demand balance.

What’s the business case for the driver?  The EV owners would probably pay 20¢/kWh for metal-air power, but run most of their mileage as PHEVs using their buffer batteries.  By moving most of the energy through an 80% efficient pathway rather than a 20% efficient one, the waste of electricity and consequent costs are greatly reduced.

Al-air batteries are considerably older than Li-ion.  There was already a demonstration vehicle written up in one of the glossy magazines in the 70’s (Popular Science?) if memory serves; Li-ion was only invented in 1980.  We know how to do this, but the economics haven’t been there.  Maybe they will be soon.

The one detail I haven’t seen anyone deal with is GHG emissions from the aluminum smelting.  IIUC, one of the byproducts besides CO2 is CF4, perfluoromethane.  This is an extremely powerful and stable greenhouse gas.  It would be necessary to capture and destroy CF4 and perhaps recycle the CO2 using something like the molten lithium carbonate electrolysis trick.

John,
You missed it. Al-O2 batteries are one time use batteries. In order to get the 90 day supply to participate in the program, you can have a warehouse full of al-o2 batteries as your 90 day supply. phienergy is claiming their last 10 years without degradation.

They don’t -replace- other forms of storage which are used for day to day types of regulation, like hydro, lion, etc. What they replace is a whole power plant that is on standby in case of some massive failure which is why you need the 90 days supply. (another reasons for a 90 day supply for the FF industry is price spikes, but a solar/wind system is a fixed cost.)

it is a warehouse available at a flip of a switch vs a coal generator unit along with 100 people, a huge stack of coal and a bunch of equipment that needs to be maintained.

They do NOT replace other day to day storage used for regulation. So you may need a stack of Lion batteries, pumped storage or something else to handle that that are more expensive initially, but lower cost for the lifetime because they are rechargeable.

I did not want to enter this discussion because it is not worth it. But the arguments which lead to that advertisemet were exactly tose “baselode” arguments which you peresent here.
And it’s still the german grid which helps out the neighbours with it’s flexibility, be it France or poland, when either their nuclear fleet is too small, or too much switched off, of if the coal power fleet in poland is insufficent.
But your imaginations about german grid makes you ignore realities.
And about coal: coal power stations are closing down one by one. The last two closing blocks two weeks ago.

The idea behind the paris agreement was to get countries to planning on a low carbon future.

If you try to make it legally binding and add a timeline, you would NEVER get anyone to sign. There are just too many variables like region, economies, existing infrastructure, etc that come into play. You would also be messing with a country’s autonomy which isn’t taken lightly because you aren’t showing any respect to other world leaders.

However, if you get everyone just to agree to plan how to implement RE and deprecate their coal plants, etc. They can do it when it becomes an economically viable option on their own terms. Since you are using economically viable, it kind of starts a keeping up with the Jones’ or status symbol type of race that everyone can participate in. The backlash is kept to a minimum.

We have seen backlash in the US, mostly because a tiny group of people have a significant vested interest in the old technology. If you can keep that vested interest from establishing itself in developing countries, over the the long term far fewer fights will occur because they never had the old technology to begin with.

Even if there is backlash like what we see in the US, the rest of the world can pick up the slack to keep driving the technology. It also changes the dynamics of trying to solve the problem. If you get everyone to cut back 20%, it is the cumulative effect of the US or China going 100% which neither country can do in the next decade because of money and logistics. whereas 20% globally is actually possible without tipping over anyone’s economy.

Well 17 year after start and 33 years before target year it is obvious that some baseload capable capacity is still exisiting. But the nubers of hours a thermal power plant is expected to run is dropping fast, so the existing plants and new capacity are built accordingly. Classical baseload plants with high system and low fuel costs are not a option any more.
The coal plants do not shut down super fast, but they close one by one, several GW per year. Looks like Hard coal output will drop at least 10% this year, maybe 20%. Lignite might have a smaller drop in output too. With the closures scheduled for next year, lignite will reduce output for sure by about 5%, and hard coal maybe more.