Is Nuclear Power Simply Too ‘Brittle’?

[GliderGuider]

This is a question well worth asking. I've always maintained that the greatest risk to a complex enterprise like our globalized industrial civilization is its lack of resilience. A brittle system with many single points of failure is vulnerable to failure cascades that can impact the entire system when small but significant things go wrong. Nuclear power sure looks like one of those, as the Fukushima catastrophe begins to cascade into world's supply chains and potentially the still-shaky world economy.

Is Nuclear Power Simply Too ‘Brittle’?

To be honest, it seems to me that all this focus on “the R-word” {GG: Radiation} misses an essential point about nuclear fission technology at this moment in time. It is incredibly brittle. We have seen on several occasions now that external events or internal human error have fairly quickly cascaded to the point of rendering an extremely expensive resource a complete write-off. We really don’t need the boogeyman of radiation poisoning to see that perhaps some caution and focus on legitimate fault tolerance is in order.

For me, at this moment, there simply is no fission technology, demonstrated at an industrial scale, that is up to the standard we should demand for a system which degrades gracefully when stressed. This does not mean such technology is not possible. But please, while it is (probably) true nuclear power is not the threat that is sometimes claimed, it is also not even slightly close to the maturity we should demand.

These same arguments apply to many areas of our current infrastructure. In the push for quick deployment, and quick profits, we have cut corners in many areas (think deep sea drilling, fracking, interstate transit systems, waste handling, tar sands, the threat of to our electronic and power systems … this could go on…). Over the long term we always pay more for these short term ’solutions’.

We never seem to have the will to do it right … but we are always willing to do it over. It astonishes me what we settle for sometimes.

[wtmusic]

1. Its greatest asset is its greatest liability

Put a lot of energy in a small package, add a little carelessness, and you have a recipe for disaster.

Is the answer to accept carelessness? Rely on big, "dumb" solutions which will kill us slowly but with far greater certainty?

Commercial aviation has gotten safer in the last 50 years because we've learned from the big mistakes. In the seventies it seemed there was a major disaster every couple of years - metal fatigue, wind shear, etc. You can have the will to do it right, but still make mistakes. Part of the will is learning hard lessons and moving on.

If we accept carelessness, that we won't take responsibility for our own safety, we're doomed no matter how you slice it.

[Throckmorton]

2. Large Thermal Power Plants tend to be that way

Nuclear, Coal, Oil, Nature Gas, all tend to be susceptible to outages from single points of vulnerability. Hydro suffers less so, but only because most Hydro installations contain multiple turbine/generators, such that the failure of a single pair doesn't usually eliminate most of a dam's generating potential.

Direct combustion and combined cycle plants are often the least reliable, as the technology that is employed in these stations (Diesel Generators or Combustion Turbines) are maintenance headaches.

Essentially, when you have a machine that makes a single usable product, electricity, anything that causes that machine to fail, halts the end products creation.

I suspect that there is a lot to be said for an 'Internet Model" for future power generation and interconnection schemes, multiple smaller generating plants (maybe even household or neighborhood level facilities)interconnected via multiple redundant paths. The control and coordination required to make this new topology function will be at first difficult to realize. However, given the advances in control technology and predictive management applications currently available, should be eminently do-able. The trick will be to have enough redundancy, and diversity (which will be key when renewables become major players), to allow for single failures with little disruption to most of the end users.

[kristopher]

7. That's "distributed generation" and you're right, it is a technically superior model

http://www.oe.energy.gov/de.htm

[Zoeisright]

3. Yes.

We can't control that poison. Period.

[Tripod]

4. It is all to fragile.

[diane in sf]

5. yes, let's get distributed on a smart grid as fast as we can.

[Kolesar]

6. And nuclear power point employees skip training and inspections and lie about it

...when they sign that training and inspections have been done. A company gets a lot of power when they are a monopoly.