Weekly PhD Newsletter | SIGN UP NOW Weekly PhD Newsletter | SIGN UP NOW

Reconstruction of Transverse Beam Distribution using Machine Learning


   Department of Physics

   Tuesday, January 31, 2023  Competition Funded PhD Project (Students Worldwide)

About the Project

The beam transverse distribution in CERN's high-radiation environment is measured by imaging the light generated by the particle beam hitting a scintillating screen, using cameras produced in-house based on radiation hard tubes. Due to the cessation of rad-hard tube production worldwide, CERN is investigating the transport of the beam image to low radiation areas using radiation tolerant optical fibers coupled to normal CMOS cameras.

In this framework, pioneering work to reconstruct the beam's transverse distribution using a single large-core multimode optical fiber began in 2020. It takes advantage of advances in generative modeling using deep learning methods, such as convolutional neural networks, and attempts to apply them to beam diagnostics. Specifically, conditional adversarial networks for image-to-image translation have been trained to translate the output plane of the fiber, imaged on a CMOS camera, into the image of the beam on the scintillating screen, optically relayed to the input plane of the fiber.

To date, this proof-of-principle work has evolved in two steps. On the theoretical side, a simulated dataset has been created from the image propagation in a simplified model of an optical fiber using a commercial ray-tracing software; it has been used to train the networks to study the feasibility of the technique. In parallel, an experimental setup was installed at CERN’s CLEAR facility and, benefiting from dedicated machine time, allowed to validate the technique by evaluating its potential and establishing a roadmap for improvements.

In this PhD project, you will initially refine the simulated data sets with more realistic optical modeling of the imaging fiber, taking into account environmental factors such as temperature and vibration effects. You will then optimize the fiber’s parameters, such as diameter and numerical aperture, and perform a market survey for available radiation-resistant fibers. In a next step, you will screen available networks for image translation, in particular the convolutional U-Net, widely used for biomedical image segmentation and the already used generative adversarial networks. You will then develop a machine learning-based model using simulated datasets and evaluate its performance. On the basis of these results, you will develop an experimental setup to validate your simulation results and carry out a measurement campaign at CERN’s CHARM irradiation facility to verify and study the accumulated dose related degradation effect to be included in the model. Additional measurements at CLEAR might be used to validate results with an electron beam.

Throughout the project you will have access to the Cockcroft Institute’s comprehensive postgraduate training in accelerator science, as well as to targeted training in data science provided by the University of Liverpool with the Centre for Doctoral Training LIV.INNO. You will also be given the opportunity to carry out an industry placement of six months to broaden your wider research and career skills.

This project will be carried out over 48 months. You will spend years 1 and 4 in the UK, and be based at CERN during years 2 and 3. Whilst in the UK, a standard RKUK PhD stipend will be paid, during the time at CERN, the usual CERN doctoral student allowance will be paid. A mandatory 6-months industry placement forms part of the project.

Supervisors: Prof Carsten Welsch, , Dr Federico Roncarolo, CERN, Switzerland

Applying: https://www.liverpool.ac.uk/study/postgraduate-research/how-to-apply/ Please ensure you quote the following reference on your application: PPPR032 - Reconstruction of Transverse Beam Distribution using Machine Learning


Email Now


Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.

PhD saved successfully
View saved PhDs