Fukushima tells us to look for safety risks, not learn by trial and error

PUBLISHED : Monday, 23 April, 2012, 12:00am
UPDATED : Monday, 23 April, 2012, 12:00am

Professor Way Kuo made a cogent case for the continued use of nuclear power in the wake of the Fukushima nuclear disaster ('Beyond the fear and prejudice', March 11). He rightly urged caution in the rush to move away from nuclear energy out of fear. However, his implication that we can simply learn from our mistakes - as in Fukushima - and resolve to do better in the future does not adequately capture the severity of the problem posed by the vulnerabilities of nuclear plants. What we need is a revolution in nuclear plant reliability assessment and monitoring.

If the engineers who designed Fukushima didn't foresee what seems to us now to have been the obvious risk of earthquake and tsunami damage, with what confidence can we say that another nuclear plant doesn't have a critical safety vulnerability that would seem obvious to us following a disaster? We've been able to contain each of the three main nuclear disasters that have occurred. But we can't count on being so lucky in the future. The risks of learning by trial and error are unacceptably high. Instead, we need to get out in front of this issue and identify critical risks before they become a reality.

These safety risks were not accurately assessed in the Fukushima power plant, and thus, provided misleading expectations about the service life of the plant. Equipment does degrade over time, but this degradation must be taken into account to avoid its failure. Reliability expectations should not be lowered with ageing; standards for a nuclear power plant must remain high and equipment degradation should be monitored through the entire life of the plant.

Prognostics and health management - based on understanding of the physics of failure - offers a safe way to assess potential vulnerabilities in a critical system before it fails.

Kelly Smith, Mark Zimmerman, and Michael Pecht, Centre for Advanced Life Cycle Engineering, University of Maryland, USA



Send to a friend

To forward this article using your default email client (e.g. Outlook), click here.

Enter multiple addresses separated by commas(,)

For unlimited access to:

SCMP.com SCMP Tablet Edition SCMP Mobile Edition 10-year news archive