How regulators grow up

7 December 2011

A hot topic at the moment is whether, in the wake of the Fukushima Daiichi incident, there should be binding international nuclear safety standards. Russia, Japan and South Africa argue there should be; others, including the USA and France, contend that the national regulator should be supreme, according to James Glasgow, a partner of US-based law firm Pillsbury Winthrop Shaw Pittman, speaking at September's World Nuclear Association symposium in London.

To me, the most interesting aspect of this debate plays out in new nuclear countries. (It is just these sort of countries, by the way, in which Russia is keen to set up self-financed, build-own-operate nuclear power plants, as in Turkey.)

I should say that although I believe that nuclear power should be for all, I also recognise that its highly technical nature occasionally pushes even the most industrialised countries to their scientific limits.

So when a deviation from standards is detected in an early commercial reactor in a new nuclear country, who can be trusted to come up with a safe and technically-acceptable solution: an inexperienced and potentially ignorant local regulator, an inexperienced local utility in a rush to get the reactor back in service, or an experienced, but potentially biased, foreign reactor vendor?

According to UxC special consultant Yun Zhou (who also spoke at the symposium), this situation actually happened at the Daya Bay station in Guangdong, China in 1995. The Chinese regulator noticed that gravity-driven emergency control rods in one of the Framatome M310 reactor were not falling as quickly as specified (2.4 seconds rather than 2.15) in the initial fuel load. The utility and vendor came up with a proposal for a temporary solution. According to Zhou, the Chinese regulator lacked the technical research and development resources to be able to independently assess their proposal. Instead, it was forced to rely on a safety margin-based argument to temporarily relax its acceptance criteria to approve the design change until the problem could be rectified a few months later.

Now, one might argue that the foreign vendor in these circumstances has an extra duty to help the country's nuclear industry and its people, by not only carrying out the work, but also by doing it to the highest standards. Dong so shows the new kids how the job would be done back home.

It is reassuring to hear that Framatome behaved as I would have hoped, and everything was fine. It is reassuring because like engineers at Fukushima after the tsunami, lacking the best possible tools, the Chinese regulatory staff used common sense to solve the problem. To me that speaks of a good nuclear safety culture.

But it is worrying too because a regulator needs the tools to do its job properly; day-to-day nuclear safety cannot be carried out on the back of an envelope. That situation also placed a disproportionate burden on the honesty and integrity of the vendor. Not that Framatome's successor Areva-or any other vendor-is untrustworthy, but anyone who carries out a job and then certifies it faces an unavoidable conflict of interest.

It remains unclear to me how a new nuclear regulator grows up. How—and when—does the initial pattern of dependence between the regulator and the vendor reverse? Must it require a big accident like Fukushima or Three Mile Island to shake up the previously-cosy relationship?

A worldwide system of independent national regulators still leaves open the possibility of problems in new nuclear countries in their formative years, when both staff and the organizations are unfamiliar with nuclear power. To overcome this, I would argue that regulatory staff should be seconded to nuclear regulators in other countries; in the case of China, I would recommend a trip across the Taiwan Straight.

This article was first published in the November 2011 issue of Nuclear Engineering International magazine.

Author Info:

Will Dalrymple is editor of Nuclear Engineering International