I've found this old scientific paper from 1988 about introduction of AI into nuclear power fields.

Were or still are there any dangers by application of such algorithm? Are nuclear power plants or human life in risk if the algorithm will fail?

Especially applications to the core, like cooling systems and other components which can be affected in negative way.

  • 2
    $\begingroup$ Your question may suffer from link rot. Please add details of the algorithm into the question. As it stands, the answer is "depends on the algorithm". $\endgroup$
    – Pimgd
    Aug 4 '16 at 12:43
  • $\begingroup$ Asking about algorithms is likely off-topic. I'm asking about general, if there were any incidents related to AI applications or what are the current real-life risks. $\endgroup$
    – kenorb
    Aug 4 '16 at 12:46
  • 1
    $\begingroup$ I agree that this very much depends on the algorithm and the implementation in question. Automation in general has been used in nuclear plants for decades, though much of it is closer to control theory than AI itself. There is a huge difference between using something like a neural network running on a server and having just a simple heuristic that switches between two or three different PID regulators. $\endgroup$ Aug 4 '16 at 13:03
  • $\begingroup$ As for the link rot, the name of the paper, the DOI and perhaps an excerpt from the abstract would be sufficient, I think. The paywall is a bigger problem. $\endgroup$ Aug 4 '16 at 13:06
  • $\begingroup$ @DisenchantedLurker There is a new branch in Control Theory Using ML Methods for solve non-linear control problems for which linear control methods are not applicable. Named Machine Learning Control (MLC) has been successfully applied to many nonlinear control problems you can check more here en.wikipedia.org/wiki/Machine_learning_control $\endgroup$ Oct 5 at 19:29

Any technology in the nuclear industry represents variance--it may be an improvement in safety or efficiency, or it may contain some unseen defect that allows a catastrophe to happen.

But the simple possibility of harm isn't enough to swing the decision one way or the other. The application of AI methods--whether to the real-time control of plant variables, or the early detection of problems, or to the design of plants and their components--seems likely to be as beneficial as in other realms.

For example, check out the publication list of a lab active in this area. Their paper I'm most familiar with is one in which they build a fault detector paired with a fault library classifier, so that the operators can be alerted not just that something is abnormal but what fault has probably occurred. This is done in such a way that standardized plants (such as, say, the French nuclear system) can share records with each other, meaning that any plant has the experience of every plant at their fingertips.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.