Physics-Informed Neural Networks for the safety analysis of nuclear reactors

Physics-Informed Neural Networks for the safety analysis of nuclear reactors


This work explores the development of surrogate models for estimating the evolution of quantities of interest during nuclear reactor accident scenarios. Physics-Informed Neural Networks (PINNs) offer a promising surrogate modelling approach because they allow integrating laws of physics and domain knowledge into traditional Neural Network (NN) surrogates. 

Specifically, the proposed solution incorporates an additional term in the PINN loss function to enforce physics-based constraints in correspondence of allocation points, which are randomly sampled points whose corresponding target output is not known. As a result, accuracy of the estimation of the quantities of interest and their adherence to the laws of physics are improved. 

Applications to a synthetic case study and to the response of a nuclear microreactor system during a Loss of Heat Sink scenario confirm that the developed surrogate model based on PINN with allocation points improves the estimation accuracy with respect to other state-of-the-art methods.

Global Particle Physics Excellence Awards


#Sciencefather 
#PhysicsInformedNN
#NeuralNetworks
#NuclearSafety
#ReactorAnalysis
#AIinNuclear
#DeepLearning
#ComputationalPhysics
#SafetyEngineering

Comments

Post a Comment

Popular posts from this blog

Numerical investigation and optimization of vertical pneumatic separation of film-like particles in lithium iron phosphate battery recycling

Image
  Numerical investigation and optimization of vertical pneumatic separation of film-like particles in lithium iron phosphate battery recycling Battery recycling is crucial due to its environmental and economic benefits, with efficient separation of the crushed battery mixture forming the basis of the process. However, numerical studies on this process remain limited primarily due to the highly non-spherical shapes of the particles involved.  Spherical particles are commonly used in limited numerical investigations with compromised accuracy and reliability. In this study, the separation of crushed battery mixture by vertical pneumatic separator is simulated using the coupled method of computational fluid dynamics (CFD) and discrete element method (DEM).  The non-spherical particles in the crushed battery mixture is resembled by cylindrical film-like particles, and the Hölzer and Sommerfeld drag coefficient is applied to enhance the accuracy of the drag force calculation. A...
Read more