Machine Learning Techniques for Uncertainty Estimation in Dynamic Aperture Prediction
The dynamic aperture is an essential concept in circular particle accelerators, providing the extent of the phase space region where particle motion remains stable over multiple turns. The accurate prediction of the dynamic aperture is key to optimising performance in accelerators such as the CERN L...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-07-01
|
| Series: | Computers |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-431X/14/7/287 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849406560079446016 |
|---|---|
| author | Carlo Emilio Montanari Robert B. Appleby Davide Di Croce Massimo Giovannozzi Tatiana Pieloni Stefano Redaelli Frederik F. Van der Veken |
| author_facet | Carlo Emilio Montanari Robert B. Appleby Davide Di Croce Massimo Giovannozzi Tatiana Pieloni Stefano Redaelli Frederik F. Van der Veken |
| author_sort | Carlo Emilio Montanari |
| collection | DOAJ |
| description | The dynamic aperture is an essential concept in circular particle accelerators, providing the extent of the phase space region where particle motion remains stable over multiple turns. The accurate prediction of the dynamic aperture is key to optimising performance in accelerators such as the CERN Large Hadron Collider and is crucial for designing future accelerators like the CERN Future Circular Hadron Collider. Traditional methods for computing the dynamic aperture are computationally demanding and involve extensive numerical simulations with numerous initial phase space conditions. In our recent work, we have devised surrogate models to predict the dynamic aperture boundary both efficiently and accurately. These models have been further refined by incorporating them into a novel active learning framework. This framework enhances performance through continual retraining and intelligent data generation based on informed sampling driven by error estimation. A critical attribute of this framework is the precise estimation of uncertainty in dynamic aperture predictions. In this study, we investigate various machine learning techniques for uncertainty estimation, including Monte Carlo dropout, bootstrap methods, and aleatory uncertainty quantification. We evaluated these approaches to determine the most effective method for reliable uncertainty estimation in dynamic aperture predictions using machine learning techniques. |
| format | Article |
| id | doaj-art-5794d4ca72f4407e9b4cddd50ea1f9ca |
| institution | Kabale University |
| issn | 2073-431X |
| language | English |
| publishDate | 2025-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Computers |
| spelling | doaj-art-5794d4ca72f4407e9b4cddd50ea1f9ca2025-08-20T03:36:19ZengMDPI AGComputers2073-431X2025-07-0114728710.3390/computers14070287Machine Learning Techniques for Uncertainty Estimation in Dynamic Aperture PredictionCarlo Emilio Montanari0Robert B. Appleby1Davide Di Croce2Massimo Giovannozzi3Tatiana Pieloni4Stefano Redaelli5Frederik F. Van der Veken6Department of Physics and Astronomy, The University of Manchester, Manchester M13 9PL, UKDepartment of Physics and Astronomy, The University of Manchester, Manchester M13 9PL, UKCERN, 1211 Geneva, SwitzerlandCERN, 1211 Geneva, SwitzerlandInstitute of Physics, École Polytechnique Fédérale de Lausanne, 1015 Lausanne, SwitzerlandCERN, 1211 Geneva, SwitzerlandCERN, 1211 Geneva, SwitzerlandThe dynamic aperture is an essential concept in circular particle accelerators, providing the extent of the phase space region where particle motion remains stable over multiple turns. The accurate prediction of the dynamic aperture is key to optimising performance in accelerators such as the CERN Large Hadron Collider and is crucial for designing future accelerators like the CERN Future Circular Hadron Collider. Traditional methods for computing the dynamic aperture are computationally demanding and involve extensive numerical simulations with numerous initial phase space conditions. In our recent work, we have devised surrogate models to predict the dynamic aperture boundary both efficiently and accurately. These models have been further refined by incorporating them into a novel active learning framework. This framework enhances performance through continual retraining and intelligent data generation based on informed sampling driven by error estimation. A critical attribute of this framework is the precise estimation of uncertainty in dynamic aperture predictions. In this study, we investigate various machine learning techniques for uncertainty estimation, including Monte Carlo dropout, bootstrap methods, and aleatory uncertainty quantification. We evaluated these approaches to determine the most effective method for reliable uncertainty estimation in dynamic aperture predictions using machine learning techniques.https://www.mdpi.com/2073-431X/14/7/287surrogate modellinguncertainty quantificationMonte Carlo dropoutbootstrap aggregationdynamic apertureaccelerator modelling |
| spellingShingle | Carlo Emilio Montanari Robert B. Appleby Davide Di Croce Massimo Giovannozzi Tatiana Pieloni Stefano Redaelli Frederik F. Van der Veken Machine Learning Techniques for Uncertainty Estimation in Dynamic Aperture Prediction Computers surrogate modelling uncertainty quantification Monte Carlo dropout bootstrap aggregation dynamic aperture accelerator modelling |
| title | Machine Learning Techniques for Uncertainty Estimation in Dynamic Aperture Prediction |
| title_full | Machine Learning Techniques for Uncertainty Estimation in Dynamic Aperture Prediction |
| title_fullStr | Machine Learning Techniques for Uncertainty Estimation in Dynamic Aperture Prediction |
| title_full_unstemmed | Machine Learning Techniques for Uncertainty Estimation in Dynamic Aperture Prediction |
| title_short | Machine Learning Techniques for Uncertainty Estimation in Dynamic Aperture Prediction |
| title_sort | machine learning techniques for uncertainty estimation in dynamic aperture prediction |
| topic | surrogate modelling uncertainty quantification Monte Carlo dropout bootstrap aggregation dynamic aperture accelerator modelling |
| url | https://www.mdpi.com/2073-431X/14/7/287 |
| work_keys_str_mv | AT carloemiliomontanari machinelearningtechniquesforuncertaintyestimationindynamicapertureprediction AT robertbappleby machinelearningtechniquesforuncertaintyestimationindynamicapertureprediction AT davidedicroce machinelearningtechniquesforuncertaintyestimationindynamicapertureprediction AT massimogiovannozzi machinelearningtechniquesforuncertaintyestimationindynamicapertureprediction AT tatianapieloni machinelearningtechniquesforuncertaintyestimationindynamicapertureprediction AT stefanoredaelli machinelearningtechniquesforuncertaintyestimationindynamicapertureprediction AT frederikfvanderveken machinelearningtechniquesforuncertaintyestimationindynamicapertureprediction |