Heat wave: decrease in nuclear production

[phil59]

Very much agree with you!

[sicetaitsimple]

Sorry to have caused so much noise based on a very clear expression in my head but obviously misspoken and misunderstood.

So, the production losses linked to events of low flows or high temperatures of the rivers have been roughly constant since 2000, at a value between 1,2TWh to 1,5TWh/year, i.e. a little more than a month of production of one of the largest 1450MW units.
No geometric progression, it is a relatively constant value over twenty years.

[Remundo]

For my part, I believe that the huge maintenance gaps in the nuclear power plant (and its inevitable aging) should have a much greater impact than this 0,3%.

On the other hand, power stations on rivers can punctually drop much more.

On the other hand, no worries about power plants by the sea.

[phil59]

So, we are already in the system!

It is becoming very urgent to react, because it is more than guesswork, but it has been realistic for 20 years.

[phil59]

For my part, I believe that the huge maintenance gaps in the nuclear power plant (and its inevitable aging) should have a much greater impact than this 0,3%.

On the other hand, power stations on rivers can punctually drop much more.

On the other hand, no worries about power plants by the sea.

I wonder what moulino51 would say....

[Remundo]

oh well on rivers/streams, the variations in flow are enormous. Easily passing from a ratio of 1 to 10, or even 100, between low water and a flood.

but on nuke power plants, it is not the water that proportionally gives the power, because the water is only there for cooling and requires a small fraction of the flow.

There are a whole bunch of standards to respect (which I don't know well) in terms of the flow taken and the temperature of the water released into the river.

If the river dries up, it can become drastic until the power plant is shut down. But in my opinion it happens very rarely and never on the whole territory.

In run-of-river power stations, there is the reserved flow, and there too, if there is no more, turbines must be stopped.

[sicetaitsimple]

For my part, I believe that the huge maintenance gaps in the nuclear power plant (and its inevitable aging) should have a much greater impact than this 0,3%.

Sure, but that just has nothing to do with it. Restrictions related to low flow rates or high temperatures only concern units in operation, not those shut down for maintenance.
For example, the two 1450MW units of Chooz are shut down until the end of the year, the flow of the Meuse which sometimes leads to production limitations at the end of the summer will only pose a problem given there will be no production....

[ABC2019]

Well, yes, but how are my initial calculations wrong?

This all dates back to the 75-80s for me....

your calculations are not wrong if we drop by 0,3% per year, it will remain well 55% after 200 years and 50% after 231 years. The only thing is nobody said it would go down 0,3% a year.

[ABC2019]

the calculations are mathematical projections that do not correspond to anything.

Nothing tells you that the losses of electrical production will gently decrease by 0,3% per year "geometrically" as the mathematicians say...

But if you like that assumption, then yes, it takes about 200 years to lose half of current production.

for fans of natural logarithms... N = ln(1/2) / ln(0,997) = 230.70 years

this is the approach of technocrats or religious maths.

because the field engineer knows that it can get complicated for many reasons...

here the problem is just that some have gone on calculations on an assumption that no one had made (starting from a misunderstanding of what sicetaitsimple had said, when he had already corrected it): no one he said that the power of the power stations would drop by 0,3% per year. So obviously the calculations do not correspond to anything, but it is not the fault of the maths, it is just that at the start the hypothesis was false.

Now if we have cases where there is really a growth or a decrease at a constant rate, obviously the math is useful. The most famous case is undoubtedly radioactive decay: it is so precise that it allows rocks to be dated several billion years, or in the case of carbon 14; organic fragments of several tens of thousands of years (but we must be careful here because the rate of carbon 14 fluctuates in the atmosphere and we must take these fluctuations into account). So of course the math is useful when applied wisely.

And even in cases where the rates of change are not constant, the approximation remains useful:

a) to have good orders of magnitude: if you say that oil consumption increases by approximately 2%, that means that it will double in approximately 70/2 = 35 years (70/x being a good approximation of your formula ln(2)/ln(1+x/100) ) . Obviously this is only approximate but it gives an order of magnitude

b) it is also used to give lower or upper limits. If you want the growth to be AT LEAST 2%, then the curve is bounded below by an exponential at + 2% per year, and you can rigorously deduce that it will have doubled BEFORE 35: this is not a precise result but it is a strict upper bound.

Don't despise math, just use it wisely, that's all.

[Remundo]

It's like vaccines...