Hongfang Zhang;Adam Stavola;Hal Ferguson;Bence Budavari;Chiman Kwan;Hongyi Wu;Jiang Li
{"title":"用于高能加速器设施辐射估算的深度多任务学习模型","authors":"Hongfang Zhang;Adam Stavola;Hal Ferguson;Bence Budavari;Chiman Kwan;Hongyi Wu;Jiang Li","doi":"10.1109/TNS.2024.3423695","DOIUrl":null,"url":null,"abstract":"Controlling the dose of radiation exposure in potential radioactive facilities is critical for ensuring the safety of staff and the public. In this article, we developed machine learning (ML) models to estimate radiation exposure efficiently at the Thomas Jefferson National Accelerator Facility (JLab), aiming to enhance safety in both accelerator facilities and public areas. Multiple sensors were deployed around the three experimental halls at JLab. Data on single-beam currents, energy levels, and radiation values at the sensor locations were collected during accelerator operation. We proposed a multitask learning (MTL) model for radiation estimation, using either 1-D convolutional neural networks (1D CNNs) or long short-term memory (LSTM) networks as the backbone. The proposed model was trained to simultaneously estimate radiation levels at the sensor locations. Experimental results demonstrated that the proposed model with LSTM backbone achieved the best estimation performance, with an average \n<inline-formula> <tex-math>$R {^{{2}}}$ </tex-math></inline-formula>\n score of 0.7557 for estimation within the same year and 0.7157 for estimation across different years. These results significantly surpassed those of competing models.","PeriodicalId":13406,"journal":{"name":"IEEE Transactions on Nuclear Science","volume":"71 8","pages":"1966-1977"},"PeriodicalIF":1.9000,"publicationDate":"2024-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Deep Multitask Learning Models for Radiation Estimation at High Energy Accelerator Facility\",\"authors\":\"Hongfang Zhang;Adam Stavola;Hal Ferguson;Bence Budavari;Chiman Kwan;Hongyi Wu;Jiang Li\",\"doi\":\"10.1109/TNS.2024.3423695\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Controlling the dose of radiation exposure in potential radioactive facilities is critical for ensuring the safety of staff and the public. In this article, we developed machine learning (ML) models to estimate radiation exposure efficiently at the Thomas Jefferson National Accelerator Facility (JLab), aiming to enhance safety in both accelerator facilities and public areas. Multiple sensors were deployed around the three experimental halls at JLab. Data on single-beam currents, energy levels, and radiation values at the sensor locations were collected during accelerator operation. We proposed a multitask learning (MTL) model for radiation estimation, using either 1-D convolutional neural networks (1D CNNs) or long short-term memory (LSTM) networks as the backbone. The proposed model was trained to simultaneously estimate radiation levels at the sensor locations. Experimental results demonstrated that the proposed model with LSTM backbone achieved the best estimation performance, with an average \\n<inline-formula> <tex-math>$R {^{{2}}}$ </tex-math></inline-formula>\\n score of 0.7557 for estimation within the same year and 0.7157 for estimation across different years. These results significantly surpassed those of competing models.\",\"PeriodicalId\":13406,\"journal\":{\"name\":\"IEEE Transactions on Nuclear Science\",\"volume\":\"71 8\",\"pages\":\"1966-1977\"},\"PeriodicalIF\":1.9000,\"publicationDate\":\"2024-07-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IEEE Transactions on Nuclear Science\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://ieeexplore.ieee.org/document/10599097/\",\"RegionNum\":3,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ENGINEERING, ELECTRICAL & ELECTRONIC\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Transactions on Nuclear Science","FirstCategoryId":"5","ListUrlMain":"https://ieeexplore.ieee.org/document/10599097/","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
Deep Multitask Learning Models for Radiation Estimation at High Energy Accelerator Facility
Controlling the dose of radiation exposure in potential radioactive facilities is critical for ensuring the safety of staff and the public. In this article, we developed machine learning (ML) models to estimate radiation exposure efficiently at the Thomas Jefferson National Accelerator Facility (JLab), aiming to enhance safety in both accelerator facilities and public areas. Multiple sensors were deployed around the three experimental halls at JLab. Data on single-beam currents, energy levels, and radiation values at the sensor locations were collected during accelerator operation. We proposed a multitask learning (MTL) model for radiation estimation, using either 1-D convolutional neural networks (1D CNNs) or long short-term memory (LSTM) networks as the backbone. The proposed model was trained to simultaneously estimate radiation levels at the sensor locations. Experimental results demonstrated that the proposed model with LSTM backbone achieved the best estimation performance, with an average
$R {^{{2}}}$
score of 0.7557 for estimation within the same year and 0.7157 for estimation across different years. These results significantly surpassed those of competing models.
期刊介绍:
The IEEE Transactions on Nuclear Science is a publication of the IEEE Nuclear and Plasma Sciences Society. It is viewed as the primary source of technical information in many of the areas it covers. As judged by JCR impact factor, TNS consistently ranks in the top five journals in the category of Nuclear Science & Technology. It has one of the higher immediacy indices, indicating that the information it publishes is viewed as timely, and has a relatively long citation half-life, indicating that the published information also is viewed as valuable for a number of years.
The IEEE Transactions on Nuclear Science is published bimonthly. Its scope includes all aspects of the theory and application of nuclear science and engineering. It focuses on instrumentation for the detection and measurement of ionizing radiation; particle accelerators and their controls; nuclear medicine and its application; effects of radiation on materials, components, and systems; reactor instrumentation and controls; and measurement of radiation in space.