Ultra Fast Simulation Algorithmics for Industrial Applications in Muon Tomography using Generative Adversarial Networks

ABSTRACT: Muon tomography is a Non-Destructive Testing technique that consists in using cosmic muons as a probing tool, in order to generate images of objects from the information given by the interaction with the muons. The development and application of this techniques requires the production of c...

Descripción completa

Detalles Bibliográficos
Autor: López Ruiz, Rubén
Tipo de recurso: tesis de maestría
Fecha de publicación:2022
País:España
Institución:Universidad de Cantabria (UC)
Repositorio:UCrea Repositorio Abierto de la Universidad de Cantabria
Idioma:inglés
OAI Identifier:oai:repositorio.unican.es:10902/29319
Acceso en línea:https://hdl.handle.net/10902/29319
Access Level:acceso abierto
Palabra clave:Muon tomography
Cosmic muons
Generative adversarial networks
Simulation
Machine learning
Tomografía muónica
Muones cósmicos
Redes generativas adversarias
Simulación
Aprendizaje automático
Descripción
Sumario:ABSTRACT: Muon tomography is a Non-Destructive Testing technique that consists in using cosmic muons as a probing tool, in order to generate images of objects from the information given by the interaction with the muons. The development and application of this techniques requires the production of considerable amounts of simulation data, usually generated with complex and slow particle simulation software. In this work, we explore the use of Generative Adversarial Networks (GAN) as a way of generating simulation data for muon tomography applications in a faster and less computationally expensive way. We have observed that GAN architectures can nicely reproduce the process of propagation of cosmic muons crossing material 50 times faster than other simulation software.