留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于机器学习的束团横向相空间测量

韩雨涛 李任恺 万唯实

韩雨涛, 李任恺, 万唯实. 基于机器学习的束团横向相空间测量[J]. 强激光与粒子束, 2023, 35: 114005. doi: 10.11884/HPLPB202335.230074
引用本文: 韩雨涛, 李任恺, 万唯实. 基于机器学习的束团横向相空间测量[J]. 强激光与粒子束, 2023, 35: 114005. doi: 10.11884/HPLPB202335.230074
Han Yutao, Li Renkai, Wan Weishi. Measurement of transverse phase space based on machine learning[J]. High Power Laser and Particle Beams, 2023, 35: 114005. doi: 10.11884/HPLPB202335.230074
Citation: Han Yutao, Li Renkai, Wan Weishi. Measurement of transverse phase space based on machine learning[J]. High Power Laser and Particle Beams, 2023, 35: 114005. doi: 10.11884/HPLPB202335.230074

基于机器学习的束团横向相空间测量

doi: 10.11884/HPLPB202335.230074
详细信息
    作者简介:

    韩雨涛,hanyt@shanghaitech.edu.cn

  • 中图分类号: TL506

Measurement of transverse phase space based on machine learning

  • 摘要: 理论上,使用断层扫描技术可以得到真实的横向相空间分布。但是想要更加精确地了解分布的细节,需要解决旋转角度范围受限和投影数目不足的问题。针对这两个问题,提出了在混合域处理的神经网络模型,即组合地在正弦域和断层域分别使用插值和去除伪影神经网络。在简单地测量束线以及投影数目比较少(7个)的情况下,该网络模型也能高质量地重建束团横向相空间分布。并且,由于选择旋转角度的方式和归一化相空间无关,因此,无需测量Twiss参数。采用该方法测量束团横向相空间,一定程度提升了重建质量,简化了测量的方式。
  • 图  1  混合域处理流程.

    Figure  1.  Flow chart of hybrid domain processing

    图  2  Residual U-Net 结构模型.

    Figure  2.  Diagram of Residual U-Net architecture

    图  3  测量束线示意图

    Figure  3.  Layout of tomography section

    图  4  聚焦参数与旋转角度关系

    Figure  4.  Function diagram of focusing parameterk and rotation angle

    图  5  某一个相空间分布和它的归一化相空间分布

    Figure  5.  A TPS distribution and its normalized TPS distribution

    图  6  不同采样方式对应的旋转角度

    Figure  6.  Rotation angles corresponding to different sampling methods

    图  7  不同采样方式的正弦图

    Figure  7.  Sinograms using different sampling methods

    图  8  两个带有细节的激光光斑.

    Figure  8.  Two laser spot with details

    图  9  插值网络的结果正弦图举例

    Figure  9.  An example of the interpolation network results in the form of sinogram

    图  10  插值网路结果断层图举例

    Figure  10.  An example of the interpolation network results in the form of tomography

    图  11  均匀的K值采样及其插值

    Figure  11.  Uniform K value sampling and its interpolation

    图  12  去除伪影网络的结果举例.

    Figure  12.  Examples of the results from removing artifacts network

    表  1  Residual U-Net 网络参数

    Table  1.   Residual U-Net network parameter settings

    name parameters output
    Conv_block_1 1 $ \times $1 conv, 64 200 $ \times $57, 64
    3 $ \times $3 conv, 64
    Conv_block_2 2 $ \times $3 conv, s=2, p=0, 64 100 $ \times $28, 64
    [3 $ \times $3 conv, 64] $ \times $2
    Conv_block_3 2 $ \times $2 conv, s=2, p=0, 64 50 $ \times $14, 64
    [3 $ \times $3 conv, 64] $ \times $2
    Conv_block_4 2 $ \times $2 conv, s=2, p=0, 64 25 $ \times $7, 64
    [3 $ \times $3 conv, 64] $ \times $2
    ConvT_block_1 $2 \times 2$ convT, s=2, p=0, 64 50 $ \times $14, 64
    ConvT_block_2 Conv_block_3, concatenation 100 $ \times $28, 64
    [ $3 \times 3$ conv,64] $ \times $2
    $2 \times 2$ convT, s=2, p=0, 64
    ConvT_block_3 Conv_block_2, concatenation 200 $ \times $57, 64
    [ $3 \times 3$ conv, 64] $ \times $2
    $2 \times 3$ convT, s=2, p=0, 64
    Conv_block_5 Conv_block_1, concatenation 200 $ \times $57, 1
    $3 \times 3$ conv,16
    $3 \times 3$ conv, 1
    shortcut connection
    下载: 导出CSV

    表  2  RED-CNN 网络参数

    Table  2.   RED-CNN network parameter settings

    name parameters output
    Conv_1 [5, Conv, 16] $ \times $2 192 $ \times $192, 16
    Conv_2 [5, Conv, 16] $ \times $2 184 $ \times $184, 16
    Conv_3 [5, Conv, 16] $ \times $2 176 $ \times $176, 16
    ConvT_1 [5, ConvT, 16] $ \times $2 184 $ \times $184, 16
    ConvT_2 Conv_2, addition 192 $ \times $192, 16
    [5, convT, 16] $ \times $2
    ConvT_3 Conv_1, addition 200 $ \times $200, 1
    [5, convT1, 1] $ \times $2
    shortcut connection
    下载: 导出CSV
  • [1] 赵振堂. 先进X射线光源加速器原理与关键技术[M]. 上海: 上海交通大学出版社, 2020: 164-165

    Zhao Zhentang. Principles and key technologies of advanced X-ray light source accelerators[M]. Shanghai: Shanghai Jiao Tong University Press, 2020: 164-165
    [2] Gordon M, Li W H, Andorf M B, et al. Four-dimensional emittance measurements of ultrafast electron diffraction optics corrected up to sextupole order[J]. Physical Review Accelerators and Beams, 2022, 25: 084001. doi: 10.1103/PhysRevAccelBeams.25.084001
    [3] Feist A, Bach N, Rubiano da Silva N, et al. Ultrafast transmission electron microscopy using a laser-driven field emitter: femtosecond resolution with a high coherence electron beam[J]. Ultramicroscopy, 2017, 176: 63-73. doi: 10.1016/j.ultramic.2016.12.005
    [4] 杜应超, 黄文会, 唐传祥, 等. 汤姆逊散射X射线源初步实验中电子束参数测量及其对X射线性能影响的研究[J]. 高能物理与核物理, 2006, 30(s1):84-86 doi: 10.3321/j.issn:0254-3052.2006.z1.027

    Du Yingchao, Huang Wenhui, Tang Chuanxiang, et al. Preliminary experiments of Thomson scattering X-ray source: measurements of electron beam’s parameters and their influence on the X-ray performance[J]. High Energy Physics and Nuclear Physics, 2006, 30(s1): 84-86 doi: 10.3321/j.issn:0254-3052.2006.z1.027
    [5] Brown W J, Anderson S G, Barty C P J, et al. Experimental characterization of an ultrafast Thomson scattering X-ray source with three-dimensional time and frequency-domain analysis[J]. Physical Review Accelerators and Beams, 2004, 7: 060702. doi: 10.1103/PhysRevSTAB.7.060702
    [6] Nagaitsev S, Huang Z, Power J, et al. Accelerator and beam physics research goals and opportunities[R]. FERMILAB-FN-1123, 2021.
    [7] Behrens C, Decker F J, Ding Y, et al. Few-femtosecond time-resolved measurements of X-ray free-electron lasers[J]. Nature Communications, 2014, 5: 3762. doi: 10.1038/ncomms4762
    [8] Rao Y N, Baartman R. Transverse phase space tomography in TRIUMf injection beamline[C]//Proceedings of IPAC2011. 2011.
    [9] Yu Qinglin, Gu Duan, Zhang Meng, et al. Transverse phase space reconstruction study in Shanghai soft X-ray FEL facility[J]. Nuclear Science and Techniques, 2018, 29: 9. doi: 10.1007/s41365-017-0338-0
    [10] McKee C B, O’Shea P G, Madey J M J. Phase space tomography of relativistic electron beams[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 1995, 358(1/3): 264-267.
    [11] Hermann B, Guzenko V A, Hürzeler O R, et al. Electron beam transverse phase space tomography using nanofabricated wire scanners with submicrometer resolution[J]. Physical Review Accelerators and Beams, 2021, 24: 022802. doi: 10.1103/PhysRevAccelBeams.24.022802
    [12] 杜应超, 向导, 黄文会, 等. 基于CT算法的束团横向相空间测量[J]. 高能物理与核物理, 2006, 30(9):888-891 doi: 10.3321/j.issn:0254-3052.2006.09.015

    Du Yingchao, Xiang Dao, Huang Wenhui, et al. Transverse phase space measurement based on computerized tomography[J]. High Energy Physics and Nuclear Physics, 2006, 30(9): 888-891 doi: 10.3321/j.issn:0254-3052.2006.09.015
    [13] Tuy H K. An inversion formula for cone-beam reconstruction[J]. SIAM Journal on Applied Mathematics, 1983, 43(3): 546-552. doi: 10.1137/0143035
    [14] Romanov A. Beam phase space tomography at fast electron linac at Fermilab[R]. FERMILAB-CONF-18-321-AD, 2018.
    [15] Stratakis D, Kishek R A, Li H, et al. Tomography as a diagnostic tool for phase space mapping of intense particle beams[J]. Physical Review Accelerators and Beams, 2006, 9: 112801. doi: 10.1103/PhysRevSTAB.9.112801
    [16] Andersen A H, Kak A C. Simultaneous algebraic reconstruction technique (SART): a superior implementation of the ART algorithm[J]. Ultrasonic Imaging, 1984, 6(1): 81-94. doi: 10.1177/016173468400600107
    [17] Hock K M, Ibison M G. A study of the maximum entropy technique for phase space tomography[J]. Journal of Instrumentation, 2013, 8: P02003. doi: 10.1088/1748-0221/8/02/P02003
    [18] Scheinker A, Cropp F, Paiagua S, et al. Adaptive deep learning for time-varying systems with hidden parameters: predicting changing input beam distributions of compact particle accelerators[J]. 2021.doi: 10.21203/rs.3.rs-373311/v1.
    [19] Wang Minwen, Wang Zhongming, Wang Di, et al. Four-dimensional phase space measurement using multiple two-dimensional profiles[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2019, 943: 162438.
    [20] Hock K M, Ibison M G, Holder D J, et al. Beam tomography in transverse normalised phase space[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2011, 642(1): 36-44.
    [21] 于清林. 束团横向相空间重建技术研究[D]. 上海: 中国科学院大学(中国科学院上海应用物理研究所), 2018

    Yu Qinglin. Study on beam transverse phase space reconstruction technology[D]. Shanghai: Chinese Academy of Sciences (Shanghai Institute of Applied Physics), 2018
    [22] Wang Ge, Ye J C, De Man B. Deep learning for tomographic image reconstruction[J]. Nature Machine Intelligence, 2020, 2(12): 737-748. doi: 10.1038/s42256-020-00273-z
    [23] Wolski A, Johnson M A, King M, et al. Transverse phase space tomography in an accelerator test facility using image compression and machine learning[J]. Physical Review Accelerators and Beams, 2022, 25: 122803. doi: 10.1103/PhysRevAccelBeams.25.122803
    [24] Radon J. Über die Bestimmung von Funktionen durch ihre Integralwerte längs gewisser Mannigfaltigkeiten[J]. Berichte uber die Verhandlungen Gesellshaft der Wissenschaften zu Leipzig. Journal of Mathematical Physics, 1917, 69: 262-277.
    [25] Hounsfield G N. Computerized transverse axial scanning (tomography): Part 1. Description of system[J]. The British Journal of Radiology, 1973, 46(552): 1016-1022. doi: 10.1259/0007-1285-46-552-1016
    [26] Sander O R, Minerbo G N, Jameson R A, et al. Beam tomography in two and four dimensions [100 MeV H+/ beams and 750 keV H beams][R]. LA-UR-79-2540; CONF-790927-20, 1979: 6.
    [27] Lee D, Choi S, Kim H J. High quality imaging from sparsely sampled computed tomography data with deep learning and wavelet transform in various domains[J]. Medical Physics, 2019, 46(1): 104-115. doi: 10.1002/mp.13258
    [28] 刘勇. 插值技术及其在CT重建中的应用[D]. 天津: 天津职业技术师范大学, 2022

    Liu Yong. Interpolation technology and its application in CT reconstruction[D]. Tianjin: Tianjin University of Technology and Education, 2022
    [29] Kalke M, Siltanen S. Sinogram interpolation method for sparse-angle tomography[J]. Applied Mathematics, 2014, 5(3): 423-441. doi: 10.4236/am.2014.53043
    [30] Han Y S, Yoo J, Ye J C. Deep residual learning for compressed sensing CT reconstruction via persistent homology analysis[DB/OL]. arXiv preprint arXiv: 1611.06391, 2016.
    [31] Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation[C]//Proceedings of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention. 2015.
    [32] Yu Yang, Gong Zhiqiang, Zhong Ping, et al. Unsupervised representation learning with deep convolutional neural network for remote sensing images[C]//Proceedings of the 9th International Conference on Image and Graphics. 2017.
    [33] Chen Hu, Zhang Yi, Kalra M K, et al. Low-dose CT with a residual encoder-decoder convolutional neural network[J]. IEEE Transactions on Medical Imaging, 2017, 36(12): 2524-2535. doi: 10.1109/TMI.2017.2715284
    [34] Xiang Dao, Du Yingchao, Yan Lixin, et al. Transverse phase space tomography using a solenoid applied to a thermal emittance measurement[J]. Physical Review Accelerators and Beams, 2009, 12: 022801. doi: 10.1103/PhysRevSTAB.12.022801
    [35] Roussel R, Edelen A, Mayes C, et al. Phase space reconstruction from accelerator beam measurements using neural networks and differentiable simulations[DB/OL]. arXiv preprint arXiv: 2209.04505, 2023.
    [36] ASTRA. Klaus floettmann DESY[CP/OL]. Hamburg, Germany: DESY, 2017. https://www.desy.de/~mpyflo/.
    [37] Gupta L. Measured Data set[Z/OL]. V1. Harvard Dataverse, 2021. https://doi.org/10.7910/DVN/MVE1YC.
    [38] Hock K M, Wolski A. Tomographic reconstruction of the full 4D transverse phase space[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2013, 726: 8-16.
    [39] 杜应超. 基于CT算法的高亮度电子束六维相空间分布的测量研究[R]. 2011: 8

    Du Yingchao. Measurement of six dimensional phase space distribution of high brightness electron beam based on CT algorithm[R]. 2011: 8
  • 加载中
图(12) / 表(2)
计量
  • 文章访问数:  306
  • HTML全文浏览量:  79
  • PDF下载量:  73
  • 被引次数: 0
出版历程
  • 收稿日期:  2023-04-04
  • 修回日期:  2023-08-02
  • 录用日期:  2023-10-19
  • 网络出版日期:  2023-11-05
  • 刊出日期:  2023-11-11

目录

    /

    返回文章
    返回