Zamfir wrote:Tyndmyr, the Fukushima reactors were anyzed by modern analysis methods. All reactors get reevaluated that way. Those are the methods used to evaluate the new designs. The acceptance standard is often set lower, but they don't use out of date analysis practices to find the result. The Fukushima events are an observation of the reliability of the analysis approach of new plants, even if they are not an observation of those plants themselves.
The legal acceptance limit in Japan is a large release frequency below 10-5 per , same as for most other places. That implies that the analysis will identify most contributing sequences of frequency 10-6, to get a reasonable error margin.
Turns they forgot a contribution of about 2*10-4( 5*10-4 for the occurrence of a tsunami at that location, and most of the active reactors hit caused a release )
In other words, the analysis approach made a mistake 200 times larger than was supposed. For new plants, the acceptance limit is often set at 10-6, which implies that most 10-7 sequences can be reliably identified. That's just rubbish, if 10-4 sequences can slip through the net. At this point, calculated event frequencies cannot be relied on for policy decisions. Their observed error is just much larger than their signal.
It's a single failure incident. "most reactors" treats the reactors as if they were seperate, but it was a single failure that was responsible for the releases. So, you're still extrapolating from a single event.
Granted, there were some failures there, and a shortcoming in the safety protocol was found the hard way, but you can't extrapolate that to say that the measurement was equally off for everything. Lots of locations are not vulnerable to the same shortcoming, and those that are, well, folks are unlikely to make the same mistake again.
Hell, Fukushima 2 was fine during the event, was it not? Improvements were being made even before the disaster. Not everywhere, sure, but this demonstrates that things are being identified and fixed.
Focusing entirely on the single failure does not provide a very good look at the system as a whole.
If most reactors today are still designs from decades ago, this is precisely BECAUSE of anti-nuclear activism that greatly delays and often prevents the construction of nuclear facilities
Here too, Japan provides a counterexample. Japan continued nuclear construction up to 2011, overruling antinuclear activism in the process. The Fukushima BWRs had two successor models: the ABWR,a larger design aimed at lowering operating costs, and the SBWR, the first design with a passive cooling capability. Since they were not listening to anti-nuclear activists, naturally built the ABWR.
You can't simultaneous blame the activists, and tout the safety of newer designs. It's only because of the activists that some new designs are safety-oriented. Without that pressure, most of the effort would have gone to lower the costs.
Which, given the vast delta between older designs and coal, probably would be a way more practical approach. Cheap, common nuclear, even at decades older safety standards, would beat the hell out of coal, resulting in less overall deaths.
Even if the focus was on cost, safety improvements would have still happened at a slower pace, and transitioning off coal earlier(as opposed to our current practical rate of...basically not), would be huge.