AI in nuclear, myth or reality?
With the fourth industrial revolution currently undergoing across the globe, AI is on everybody’s lips. Is AI in nuclear also a thing ? Would we see one day an AI piloting a nuclear reactor? This debate already started a whilst back. As seen here, the U.S. Department Of Energy published an analysis back in 1985. At the moment, the U.S. Nuclear Regulatory Commission is trying to understand where AI and ML are currently being used and what are their perspectives in terms of evolution to provide a safer and more stable electricity grid.
Understanding what is AI
Until few years ago, when mentioning AI, many people would think of some American blockbusters such as Terminator. Now, it has been largely accepted amongst our society what AI is. You start to see it every day. Autonomous cars hit the news as the next big thing in the car industry alongside with EV’s.
The fathers of this field would describe AI as “any task executed by a machine that would have originally required a human action”.
AI in nuclear
With the improvement of online monitoring systems for nuclear power plant components, and acknowledging the high volume of data to process, AI and its subfield ML – machine learning – are extremely useful to support operator’s decisions. By being able to extract all the variables and understand the connections between the different information, an AI algorithm can improve diagnostic and prognostic, leading to a higher rate of plant performance.
Up to now, the lifespan of a component is mostly estimated through a statistical average. However, with the right algorithm, it will be possible to predict more accurately when a component will fail, automatically reducing the costs of operation and maintenance.
Finally, by requiring less human inputs, AI and ML will be able to be more reactive, lowering the risks of incidents/accidents and mitigating any potential risks on the environment.
Is AI in nuclear ethical?
The big question is: Is the use of AI in nuclear ethical? In nuclear medicine, a mistake can cost someone’s life. Should the key decisions be taken by an algorithm?
The same idea applies in electricity generation. What happens if there’s a glitch in the algorithm and the plant operator does not have access to some essential information?
Innovation is necessary but must be thought carefully.
If you’re interested in innovation, why not joining us and discover our COTS innovations?