Nuclear Energy Data: AI and Virtual Reality

Rhodalene Benjamin-Addy
February 6, 2019

Submitted as coursework for PH241, Stanford University, Winter 2018

Nuclear Energy Data Importance

Fig. 1: Example of a nuclear power plant control room (Kozloduy plant in Bulgaria). (Source: Wikimedia Commons)

With the increasing popularity of artificial intelligence (AI), there is a possiblility for this arena to enter into nuclear energy technologies. AI relies on the access of large amounts of data and this is no small task. With data from over 22157 experiments, the EXFOR Database, where much of experimental nuclear reaction data is stored, is constantly trying to find ways of storing better data. [1] From figuring out more output formats to store new data to questioning the economic risks of changing the data storage formatting, in order for nuclear energy intersect with AI, there has to be proper consideration for how data is continued to be stored and expansions of the softwares that are being used.

Virtual Reality the in Nuclear Energy Field

There has already been the use of virtual reality (VR) in nuclear energy which has been helpful in providing an ergonomic evaluation of control rooms. [2] There is are many safety protocols that go into designing a nuclear power plant and by creating a simulation of the control room (Fig. 1), the place where people can control the environments and prevent accidents from occuring, there is opportunity to create the most efficient control room. Not only simulating the room, but also the actual plant has been done and this allows for constant redesign to create the safest environment for workers in these plants. [2] The continued use and improvement of VR in nuclear energy can really minimize the errors that may occur and the harness of these energies more efficiently but more importantly safely.

Artificial Intelligence Use in the Field

In pressurized water reactors (PWR), AI has been used to model and control these systems. By the use of artificial neural networks, which is essentially the way in which a computer learns over time through new data that it is being giving, there is an ability for this to be used to model and analyze these PWR systems. In one study, they simulated a PWR system using AI and then created fuzzy controllers from this system and compared them to that of the standard ones. The results showed that there was a reduced error by 32 percent when with the fuzzy controller modeled from the AI one. [3] This just shows the benefit of minimzed error and creating stronger sensitivity which could be beneficial when dealing with these high energy systems.


Clearly, there is already an incorporation of AI and VR in nuclear energy and has proved to be quite beneficial especially in the processes of design and testing and constant effort for better data collection which inherently makes these technologies easier to work with. There could, in the future, be incorporation of this in terms of training workers in these plants and maybe there might be a point where there would be a reduced amount of workers because the computers could eventually learn so well that the error rate is lower than a humans and the control room could essentially be controlled remotely. There is a strong potential for this field and its relationship with these computer technologies to go farther than one could imagine.

© Rhodalene Benjamin-Addy. The author warrants that the work is the author's own and that Stanford University provided no input other than typesetting and referencing guidelines. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.


[1] R. A. Forrest, V. Zerkin, and S. Simakov, "Developments of the EXFOR Database: Possible New Formats," Nucl. Data Sheets 120, 268 (2014).

[2] L. B. S. Gatto et al., "Virtual Simulation of a Nuclear Power Plant's Control Room as a Tool for Ergonomic Evaluation," Prog. Nucl. Energy 64, 8 (2013).

[3] M. V. de Oliveira and J. C. S. de Almeida, "Application of Artificial Intelligence Techniques in Modeling and Control of a Nuclear Power Plant Pressurizer System," Prog. Nucl. Energy 63, 71 (2013).