AI and ML approaches in nuclear engineering analysis
The launch of the first all-encompassing AI and ML benchmark for predicting the Critical Heat Flux (CHF) is a major step forward. In a boiling system, this CHF is equivalent to the point where wall heat transfer begins to drop dramatically, a phenomenon known variously as the critical boiling transition, the boiling crisis, and (depending on operational circumstances) departure from nucleate boiling (DNB) or dry out. Wall temperature increases caused by CHF in a heat transfer-controlled system, such as a nuclear reactor core, can hasten the oxidation of the walls and eventually cause the fuel rods to fail. Despite its significance as a design limit criterion for reactor safety, CHF is notoriously difficult to anticipate precisely due to the intricate nature of the local fluid flow and heat exchange dynamics.
Read: Top 15 AI Trends In 5G Technology
Future generation nuclear systems and deployment of innovations
Artificial intelligence (AI) and machine learning (ML) have recently made significant performance advancements, which has sparked extraordinary interest among nuclear engineers. Despite the advancements, the usability and widespread use of AI and ML approaches in nuclear engineering analysis are constrained by the absence of specialized benchmark exercises.
To meet the NEA’s strategic goal of “contributing to building a solid scientific and technical basis for the development of future generation nuclear systems and deployment of innovations,” the Nuclear Science Committee’s Working Party on Scientific Issues and Uncertainty Analyses established the Task Force on Artificial Intelligence and Machine Learning for Scientific Computing in Nuclear Engineering within the Expert Group on Reactor Systems Multi-Physics (EGMUP). From single physics to multi-scale and multi-physics, the Task Force will focus on building benchmark exercises that will target critical AI and ML operations across a wide range of computational domains of interest.
Where is it heading?
This level of participation is indicative of the seriousness and dedication of the international scientific community to the use of artificial intelligence and machine learning in the field of nuclear engineering. In the end, the Task Force hopes to use what it has learned from the benchmarks to lay down some ground rules for the use of AI and ML in nuclear engineering’s scientific computing.
Read Top 20 Uses of Artificial Intelligence In Cloud Computing For 2024
Current CHF models are generally based on empirical correlations created and verified for a specific application case domain. Using an extensive experimental database given by the US Nuclear Regulatory Commission (NRC), this benchmark seeks to enhance CHF modeling via the application of AI and ML techniques. The enhanced modeling can lead to a better knowledge of the safety margins and create new options for design or operational optimizations.
Read OpenAI Open-Source ASR Model Launched- Whisper 3
[To share your insights with us, please write to sghosh@martechseries.com]
Comments are closed.