Radiation Transport Calculations Around the ESS Accelerator Using Kernel Density Estimation
(2024) FYSM34 20241Department of Physics
Particle and nuclear physics
- Abstract
- In order to safely design accelerator-based research facilities, such as the European Spallation Source, shielding calculations are carried out using Monte-Carlo particle simulations. However, these calculations gain high uncertainty when large amounts of shielding are needed, due to decreasing particle statistics over distance and mass. To circumvent this issue, variance reduction methods are used to reduce errors in calculations. A possible method for this purpose is Kernel Density Estimation (KDE). In this thesis, KDE methods has been applied to shielding calculations done around the ESS linear accelerator. A model of the accelerator was constructed in which the calculations were run using a Monte Carlo particle transport code. It was... (More)
- In order to safely design accelerator-based research facilities, such as the European Spallation Source, shielding calculations are carried out using Monte-Carlo particle simulations. However, these calculations gain high uncertainty when large amounts of shielding are needed, due to decreasing particle statistics over distance and mass. To circumvent this issue, variance reduction methods are used to reduce errors in calculations. A possible method for this purpose is Kernel Density Estimation (KDE). In this thesis, KDE methods has been applied to shielding calculations done around the ESS linear accelerator. A model of the accelerator was constructed in which the calculations were run using a Monte Carlo particle transport code. It was found that using KDE can be used to reduce uncertainty in shielding calculations. However, the parameters of the bandwidth selection method, such as the property importance, needs to be worked with in order to get a good estimate of the distributions of the particle properties. (Less)
- Popular Abstract
- Radiation shielding calculations are an important aspect when building a particle accelerator, such as the one at the European Spallation Source (ESS), a research facility located in the northern part of Lund. They are applied to determine how much shielding is needed to protect workers and the public from radiation during operation. However, these types of calculations can have high uncertainty when done for large amounts of shielding, due to the number of particles decreasing over distance from the accelerator from interactions in the shielding.
To circumvent this issue, there are several different types of variance reduction techniques that can be applied to make the problem more tractable. As the name implies, they are used to... (More) - Radiation shielding calculations are an important aspect when building a particle accelerator, such as the one at the European Spallation Source (ESS), a research facility located in the northern part of Lund. They are applied to determine how much shielding is needed to protect workers and the public from radiation during operation. However, these types of calculations can have high uncertainty when done for large amounts of shielding, due to the number of particles decreasing over distance from the accelerator from interactions in the shielding.
To circumvent this issue, there are several different types of variance reduction techniques that can be applied to make the problem more tractable. As the name implies, they are used to decrease the uncertainty of calculations where needed. A possible variance reduction method for high-energy accelerator applications is Kernel Density Estimation (KDE). This is a method for estimating the probability distribution of a set of particles. Doing so allows the generation of new particles relative with said distribution. By creating a new source some distance away from the original source, the uncertainty beyond the new source term location can be decreased more efficiently.
The goal of this project was to apply KDE techniques to shielding calculations done around the ESS accelerator. A computer model of the accelerator was constructed, in which the calculations were performed using Monte Carlo particle transport code. Properties of simulated particles crossing certain regions were recorded, and then used to create new source terms using KDE. The distributions of the properties from the different sources were compared to evaluate performance. The simulated dose rates were also compared to check the quality of the results. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9165942
- author
- Sjöö, Rasmus LU
- supervisor
- organization
- course
- FYSM34 20241
- year
- 2024
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- European Spallation Source, ESS, Linear Accelerator, Kernel Density Estimation, Monte Carlo, Particle Transport, Shielding Calculations.
- language
- English
- id
- 9165942
- date added to LUP
- 2024-06-20 08:50:00
- date last changed
- 2024-06-20 08:50:00
@misc{9165942,
abstract = {{In order to safely design accelerator-based research facilities, such as the European Spallation Source, shielding calculations are carried out using Monte-Carlo particle simulations. However, these calculations gain high uncertainty when large amounts of shielding are needed, due to decreasing particle statistics over distance and mass. To circumvent this issue, variance reduction methods are used to reduce errors in calculations. A possible method for this purpose is Kernel Density Estimation (KDE). In this thesis, KDE methods has been applied to shielding calculations done around the ESS linear accelerator. A model of the accelerator was constructed in which the calculations were run using a Monte Carlo particle transport code. It was found that using KDE can be used to reduce uncertainty in shielding calculations. However, the parameters of the bandwidth selection method, such as the property importance, needs to be worked with in order to get a good estimate of the distributions of the particle properties.}},
author = {{Sjöö, Rasmus}},
language = {{eng}},
note = {{Student Paper}},
title = {{Radiation Transport Calculations Around the ESS Accelerator Using Kernel Density Estimation}},
year = {{2024}},
}