guest ::
| Home > Publications database > Neutron instruments simulations using machine learning techniques with KDSource |
| Home > Publications database > Neutron instruments simulations using machine learning techniques with KDSource |
| Conference Presentation (Invited) | FZJ-2022-03619 |
; ; ; ; ; ; ; ; ;
2022
Please use a persistent id in citations: http://hdl.handle.net/2128/33118
Abstract: The design of neutron instruments usually is related to the simulations of neutron beams. These simulations are normally decoupled from the neutron source since the nuclear reactions that govern the generation of neutrons in the source are independent of the specific interactions that take place in the neutron beam path. Also, neutron beams are usually transported far away from the source to reduce the background signal in the measurements and the radiation dose of the personnel. When evaluating the neutron beam under different operating conditions, it is useful to have a source that can be re-sampled. KDSource [1,2] is an open-source code that uses the adaptive multivariate Kernel Density Estimation (KDE) method to estimate the source distribution at a given point in the beam trajectory, which seeks to overcome limitations of other variance-reduction techniques. The approach presents a novel methodology to optimize source modelling, which may be especially suited for neutron beam and radiation shielding simulations. The core idea of the methodology is to use some machine learning libraries and algorithms to optimize the bandwidth selection for each source variable. With this strategy, smooth estimates of the variable distributions may be obtained from particle lists at a given point in a simulation that maintains correlations among all the variables. The code implements the proposed methodology in Python, and it consists of a module for KDE model optimization, and another for sampling (i.e. generating new particles using the previously optimized model). This work aims to present the KDSource code with some usage examples to design the neutron imaging instruments for the HBS project. [1] N.S. Schmidt, O.I. Abbate, Z.M. Prieto, J.I. Robledo, J.I. Márquez Damián, A.A. Márquez, J. Dawidowski, 2022. KDSource, a tool for the generation of Monte Carlo particle sources using kernel density estimation. Ann. Nucl. Energy 177. URL: https://doi.org/10.1016/j.anucene.2022.109309 [2] O.I. Abbate, N.S. Schmidt, Z.M. Prieto, J.I. Robledo, J.I. Márquez Damián, A.A. Márquez, J. Dawidowski, 2021. KDSource, a tool for the generation of Monte Carlo particle sources using kernel density estimation. GitHub repository. URL: https://github.com/KDSource/KDSource
|
The record appears in these collections: |
PDF
Fulltext by OpenAccess repository