Benchmark on Artificial Intelligence and Machine Learning for Scientific Computing in Nuclear Engineering. Phase 1: Critical Heat Flux Exercise Specifications

Adobe Acrobat PDF Document - on 2/8/24 at 5:09 PM

Benchmark EGMUP EGTHM …

Under the guidance of the NEA Nuclear Science Committee (NSC), the Working Party on Scientific Issues and Uncertainty Analysis of Reactor Systems (WPRS) studies the reactor physics, fuel performance, and radiation transport and shielding in present and future nuclear power systems. In 2022, the WPRS Expert Group on Reactor Systems Multi-Physics (EGMUP) mandated a new Task Force on Artificial Intelligence (AI) and Machine Learning (ML) for Scientific Computing in Nuclear Engineering to develop a benchmark that will provide guidelines and exercises to help participants to develop and evaluate the performance of their artificial intelligence and machine learning methods. The benchmark activity of this Task Force is structured into two phases:

Phase 1: Regression, Classification and Verification, Validation and Uncertainty Quantification (VVUQ); Dimensionality Reduction and Anomaly Detection.
Phase 2: Generative Deep Learning and Data Augmentation; Design Optimisation.

This document provides the specifications of the critical heat flux exercise, which is part of the Phase 1 activities.