The Energy Collective Group

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.

9,757 Subscribers

Post

France Can't Meet Its Own Power Demand

photo EPA

Photo: EPA

France was heavily dependent on power imports from Germany during the first cold spell of this winter, despite the fact that most of the country’s nuclear reactors are back online, writes Craig Morris from the Energy Transition blog. As the US is now also investigating 17 nuclear reactors with parts from reactor producer Areva, just rescued by the French state, it shows the perilous state the French power sector is in. Courtesy Energy Transition.

In the fall of 2016, 20 of France’s 58 reactors were offline, largely for inspections. As of mid-January, most of these reactors were back online, however. The fleet’s total generation capacity is 63 GW, and RTE’s website shows an ouput level approaching 55 GW, which is near the maximum (power plants generally do not run far above 90% so they can still provide grid services, such as reactive power). EDF’s list of production by plant (zip) on January 11 seems to show that 7 reactors are still offline, which is not an especially high number (only 15 percent of the fleet).

The risk is that France gets a lot of its space heat from electric units, so a cold spell poses a challenge, as we saw in 2012, when Germany helped prevent a blackout in France.

Craig 1

Caption: A chart showing the level of power used for heat (red) relative to the other demand for power (blue) in France. Electric heaters can double power demand in France, as happened in 2012. (Source: RTE)

 

Europe got its first cold spell this winter in the first week of January. During that week, France was only a net exporter of power for three brief periods, each at around 4 am, when demand in France and neighboring countries is lowest. Otherwise, France was a net power importer, peaking at 8 GW (roughly 10 percent of around 8 am on Friday, January 6), as shown by the grey area at the top of the chart below. France has a total import capacity of 12,200 MW (in German) from all of its neighbors, so there is room for greater power imports, but four megawatts is not much relative to the potential for twenty gigawatts of greater demand to reach the record level during the 2012 cold spell.

Craig 2

France only has 3,007 MW of coal installed, so it’s coal fleet was also running practically full blast, along with gas, listed at 10,909 MW, just above the 9,181 MW the chart indicates above. The oil-fired capacity is much greater than the 1,100 MW generated, however, at 8,645 MW – so clearly, France has lots more generation capacity, but imports are cheaper than domestic production from oil.

Price comparison

The chart below shows the price difference between Germany and France in the first week of the year, as shown above. The gap is most striking on Wednesday for peak power, with France paying 7.2 cents and Germany 3.8 for a kilowatt-hour.

Craig 3

The chart below shows the power trading situation for that week’s situation with all countries bordering France. Belgium and Germany are unfortunately now lumped together (they were reported separately until last year), thereby making it impossible to say how much came from which country. But clearly, France is reliant on its neighbors during cold spells – initially, because imports are cheaper than power from oil. But the French don’t have that much reserve capacity left if the reactors cannot stay online, and the French nuclear watchdog ASN says it wants to investigate further.

Craig 4

The US is now also reviewing the safety of 17 reactors of 99 in operation in the country with parts from Areva, the manufacturer of the defective equipment made at the Le Creusot facility. Areva is to receive a “capital injection” to the tune of 4.5 billion euros from the French state; the EU approved the deal as compliant with state aid guidelines on Tuesday.

by

Craig Morris (@PPchef) is the lead author of Global Energy Transition. He is co-author of Energy Democracy, the first history of Germany’s Energiewende, and is currently Senior Fellow at the IASS. This article was first published on the Energy Transition blog of the Heinrich Böll foundation and is republished here with permission.

Original Post

Content Discussion

Jarmo Mikkonen's picture
Jarmo Mikkonen on January 24, 2017

The chart below shows the price difference between Germany and France in the first week of the year, as shown above. The gap is most striking on Wednesday for peak power, with France paying 7.2 cents and Germany 3.8 for a kilowatt-hour.

These are of course wholesale prices that vary daily. Consumers in Germany and France pay different rates.

In Germany, that is 30 eurocents/kWh, on average. In France, 17 eurocents/kWh. (1st quarter 2016).

Bob Meinetz's picture
Bob Meinetz on January 24, 2017

Craig, reliable nuclear generation not only makes French electricity cleaner – it makes the country far more self-sufficient than California. We now import more than 30% of our electricity, and we’re scheduled to get even more from out-of-state coal and natural gas plants if Diablo Canyon Power Plant is shut down in 2024-2025.

The signficant difference is the French use electricity for heating; we burn natural gas, adding millions of tonnes of unnecessary CO2 emissions.

The most environmentally-responsible answer for both is to build more nuclear plants instead of shutting them down.

Jesper Antonsson's picture
Jesper Antonsson on January 24, 2017

Taxes are kindof arbitrary, but the EEG fee of 7 cents is a real cost. So Germans, every hour of the year, pays as much _extra_ for its 30% RE content as France pays for peak power at the worst season of the year when its nuclear fleet has some rare troubles. On top of that, Germany has to pay the wholesale price of power.

This anti-nuclear guy, Craig, seems a bit desperate…

Roger Arnold's picture
Roger Arnold on January 24, 2017

A problem with infrastructure in general is that its life cycle tends to play out in long waves. After completion of an initial buildup phase, the infrastructure serves quietly for some time. People come to take it for granted. The expertise that went into building it erodes, along with the infrastructure itself. Eventually, mounting problems can’t be ignored, and we have a crisis.

France’s nuclear power infrastructure was built at a remarkably low cost in a remarkably short period of time — at least by the standards we’ve become accustomed to for new NPPs today. For 35 years, it has served served France well. I have no way of knowing whether the carbon segregation problem in steel castings which led to the recent shutdowns for safety inspection of 20 French reactors is really a serious safety issue, or whether — as some charge — it’s more a politically motivated move by Hollande’s government to placate his anti-nuclear Green Party allies.

What’s certain is that France’s nuclear power fleet is getting “long in the tooth”. Officials may have been complacent about its reliability. They evidently had not felt the need to do anything to boost reserve capacity in the event of systemic problems of the sort that led to the shutdowns. They haven’t been pushing common-sense programs that would have reduced their exposure

France should have been encouraging people to move away from electrical resistance heating for their homes and businesses, in favor of energy-efficient heat pumps. That would have reduced power consumption in general; more particularly, it would have moderated blips in power demand from winter cold snaps.

That said, and despite the recent fuss, France remains a paragon of clean energy and low carbon emissions compared to “green” Germany.

Nathan Wilson's picture
Nathan Wilson on January 24, 2017

That graph of French electricity demand versus day of the year is a great illustration of why clean baseload energy with a lesser amount of hydro is a great solution for them*. If they could get cheap and reliable solar from south Africa, then that would be a good supplement; but as it stands, domestic solar is exactly the wrong solution (due seasonal load anti-correlation).

In fact, the best solution to reduce French fossil fuel consumption while boosting energy security is to replace their widespread electric heat with a hot-water based district heat system. In this way, much of the annual heat supply can be provided by waste heat from nuclear plants. But district heating is very flexible, so peak season top-up can be provided by any source (electric heat pumps or resistance heaters, combined-heat-and-power peaking units, simple combustion boilers).

In a non-fossil energy system, peaking heat is naturally a stored energy application. District heat systems make it easy to supplement the normal heat source (e.g. waste heat or heat pumps) with a stored peaking source such as hydrogen or ammonia combustion.

Of course electrical peaking units can also be powered by hydrogen or ammonia, but free trade rules would force such systems to compete with cheap+dirty Germany coal power. And obviously such electrical peakers would generate copious waste heat which could be utilized by district heat customers.

The obsession which European green groups have in replacing or finding faults with nuclear power have distracted from the importance of replacing combustion of fossil fuels (German coal in particular seems to be escaping proper scrutiny).

* In much of the southern US, we have peak demand during summer due to air conditioning, so solar is a suitable addition to a predominantly nuclear grid mix.

Mark Heslep's picture
Mark Heslep on January 24, 2017

France is reliant on its neighbors during cold spells – initially, because imports are cheaper than power from oil

Per RTE, France has some 25 GW of combustion capacity, which is run infrequently, and some 25 GW of hydro. With the French nuclear fleet running at 55 GW, the word “reliant” does not apply, even in a cold winter. If highly subsidized intermittent power is exported by others, why not import it?

Unlike intermitent power, nuclear shutdowns can be scheduled, as they were in this case. If a demand shortage was actually a problem, and it was not, the reactors could have been taken offline one or two at time for inspection, over a longer period.

Mark Heslep's picture
Mark Heslep on January 24, 2017

remarkably low cost in a remarkably short period of time — at least by the standards we’ve become accustomed to for new NPPs today.

In the West. China today is building nuclear at a similar rate to France in the 20th century.

France’s nuclear power fleet is getting “long in the tooth”

Average age 32 yrs. US average 35 yrs, with 70 yr lifetimes under consideration. Hoover Dam is 86 yo.

They haven’t been pushing common-sense programs that would have reduced their exposure

With 50 GW of combustion and hydro capacity, plus 12 GW interconnect, what exposure?

Roger Arnold's picture
Roger Arnold on January 25, 2017

With 50 GW of combustion and hydro capacity, plus 12 GW interconnect, what exposure?

Didn’t they have to recommission mothballed coal plants before the shutdowns and then draw heavily on the interconnect to meet demand during the cold snap? When the norm has long been to be a net power exporter on the interconnect, wouldn’t you say there’s exposure in depending on neighboring countries to be able to turn around and let you import power?

Admittedly, I have a pretty weak case to say that France was negligent in not pursuing measures to improve heating efficiency and give themselves a larger margin of reserve. I’m just grousing about our human tendency to let things slip by until there’s a crisis. (Not that I don’t function that way myself. Why let that spoil a good chance to complain?)

Mark Heslep's picture
Mark Heslep on January 25, 2017

I’m just grousing about our human tendency to let things slip by until there’s a crisis

I agree about that human tendency in all of us, that manifests to varying degrees. Relative to other countries, I see France as ridiculously less exposed to energy shocks than its neighbors. Mothballing a coal plant (if they did, I dunno), so that it can be used in a rare event is prudent planning in my view, not risky behavior.

Darius Bentvels's picture
Darius Bentvels on January 25, 2017

French scientists at govt institute ADEME studied the best options for electricity in 2050.
They concluded that 80% renewable was the cheapest option, while 40% renewable and 100% renewable cost somewhat more (both being at same price level).

So they are now moving into same direction as Germany!

Nathan Wilson's picture
Nathan Wilson on January 27, 2017

As is typical, the ADEME-2016 renewable study makes various choices that serve to hide the cost increase of renewables compared to nuclear. This study, for example, assumes all nuclear has a cost of €80/MWh (US $85), as though nuclear plants only last as long as a 30 year mortgage (80 years is more realistic, and refurbishing old nukes is a very cost effective way to make clean electricity). Of course the valuable existing investment in nuclear plants is ignored.

The ADEME study also assumes France will have strong interconnections to adjacent countries, and those countries will have lots of flexible generation (including 20% fossil fuel) to help balance the French grid.

Similarly, power-to-fuel-to-power is used in the 100% renewable case for seasonal energy storage, with a cost increase of €138/MWh (US $147), and 33% efficiency. The gas grid is assumed to provide free seasonal storage for produced fuel. Carbon-neutral CO2 is assumed to be available for methanation, so no CO2 tax is involved. Fossil fuel is priced at €100/MWh for CCGT.

6 hour storage is assumed to be available for an increase of €58/MWh (US $62). 32 hour storage is €106/MWh (US $113).

Some other observations:
– going from 40% to 80% renewables, the fossil fuel use is allowed to double.
– solar PV is not favored at >=80% renewables, with the preferred wind-to-PV ratio around 4, in MWh.
– all domestic hot water and half of BEV charging is dispatched to mostly follow PV output (7.5 GW peak) with some late night use (1.7 GW).
– replacing nuclear at €80/MWh with PV and wind at €60/MWh or €65/MWh did not produce corresponding cost reductions (due to added storage? curtailment?)

“Some technologies with a flat production profile, such as geothermal electric production, or which are dispatchable, such as wood-fired co-generation, see their extra cost compensated by the benefits they provide to the overall system operations.”
This should also apply to nuclear.

The study acknowledges the very high cost of the last MWh of electricity as renewable penetration grows, (they cite €183/MWh when going from 95% to 100% renewables).

In summary, the study did not change my opinion that for a given level of difficulty, a renewable-rich grid will be 2-3x as dirty as one rich in nuclear power.

http://www.ademe.fr/en/a-100-renewable-electricity-mix-analyses-and-optimisations

Darius Bentvels's picture
Darius Bentvels on January 27, 2017

In France consumers pay very little tax on the electricity they buy.

In Germany (as well as in NL though less) consumers pay a lot of taxes to different authorities. The costs of the Energiewende is taxed as the Energiewende surcharge which is 7cnt minus the 2cnt decrease in whole sale prices due to the Energiewende.
So the real Energiewende costs is ~5cnt/KWh.

E.g.
In NL we have a general energy tax which applies for electricity. The idea:
– Govt. need the money, so has to tax anyway;
– Higher rates stimulate more economic use, so is well for the climate.
Hence it is better to tax electricity, car fuel, etc more and keep e.g. the V.A.T. lower.

Still VAT is here much higher than in USA, but that is because govt pays here far more for education, social welfare, medical, etc. Which pays off as shown by e.g. the 2years longer av. life expectancy in NL than in USA, more happiness here according to UN, etc.

Darius Bentvels's picture
Darius Bentvels on January 27, 2017

study assumes nuclear cost €80/MWh. As nuclear plants only last 30 years (80 years is more realistic)??
As shown with Hinkley C where most costs are known due to EU subsidy rules, new nuclear cost now:
£100/MWh + investment, decommission, waste, liabiliy limitation subsidies; worth ~£40/MWh. So total £140=€160/MWh.
Which is twice the amount the ADEME study assumes for the mix of new and old nuclear in 2050.

On average nuclear plants close at an age of ~40years. So 80years is an extreme optimistic unfounded estimation. Especially since whole sale prices are widely expected to decrease further in the far future, due to a.o. much cheaper wind, solar, storage, etc. (<2cnt/KWh)

a renewable-rich grid will be 2-3x as dirty as one rich in nuclear power.??
It’s very expensive to increase nuclear share substantial above base load level which is 50%-60%, also shown by the facts that:
– only France reached more nuclear (thanks to flexible hydro, etc);
– several countries reached 50-60% nuclear share but then stopped;
– France is now targeting to reduce nuclear fast to that 50%;

So for low emissions renewable have to deliver major share (as ADEME also concluded).

Furthermore
Due to its high costs nuclear is nowadays 2-10 times more dirty than renewable wind & solar. So replacing nuclear with renewable implies cleaning the grid.

Darius Bentvels's picture
Darius Bentvels on January 27, 2017

You’re right. German authorities even ordered a temporal stop on the closure of power plants in the south-west in order to help France
and of course themselves as then they were sure the whole grid incl interconnection would stay stable.

Nathan Wilson's picture
Nathan Wilson on January 29, 2017

It’s very expensive to increase nuclear share substantial above base load level which is 50%-60%, …

Baseload is actually much higher than that. I looked at the 365 day load data from the Southwest Power Pool (which operates the transmission system for the central US, from North Dakota to west Texas, source):
-Assuming 1% curtailment, baseload accounted for 83% of demand.
– Using 5% curtailment, baseload was 90% of demand.

Thermal power plants built 50% years ago had slow output ramping and rather high values for minimum output level. But this simply is not true anymore.

The fossil methane powered plants which the US is shifting towards have fast ramping gas turbines which produce all or half of their output. The (dirty) coal fired plants that Germany has deployed were designed to ramp quickly to work alongside intermittent power sources. NuScale has shown that their new SMR nuclear plant will ramp fast enough to follow total load or even harder, netload alongside Idaho’s Horse Butte wind farm. Hawaii has found that their old fossil fuel power plants can ramp quickly enough to support solar, when used with 1 hour batteries.

Darius Bentvels's picture
Darius Bentvels on January 29, 2017

Looking at day load data, implies that you don’t see the 15minuntes & hourly load changes (day/night/etc) which constitute major part of the load variation.

The graph at sheet 9 of this presentation demonstrates nuclear’s inability to load follow.