稀疏表示与多任务学习的复杂核素识别
|
摘要
为提高核探测器在复杂环境下测量的适应性,提出了一种能谱校正和核素识别方法.针对核信号探测过程中,由于环境温度的交替变化会出现γ能谱偏移导致多核素识别率低的问题,提出了一种基于稀疏表示和多任务学习的核素识别方法.首先建立一个用于描述环境变量对于当前测量能谱影响的迁移矩阵,其次对测量能谱进行建模,该模型可以表示为标准能谱中独立核素能谱的瞬时叠加,由此核素识别问题就转化为多种核素能谱稀疏分解的问题,为求解该非凸优化问题采用交替方向乘子法(ADMM)的多任务学习方法同时优化迁移矩阵并进行稀疏分解,实现多核素识别.为验证该方法的可行性和有效性,利用高低温交变试验箱对Cs I(Tl)探测器的测量环境进行模拟,分别测量得到11种核素和典型混合核素的实际放射性元素能谱数据,以及基于蒙特卡洛分析软件Geant4仿真IAEA规定的27种核素的单一与混合核素数据进行实验.结果表明,提出的方法即使在温度为:-20℃~50℃的环境下依然可以准确地识别多种常用核素.
A spectra calibration method and a radionuclide identification method are developed for the improvement of the adaptability to the nuclear detector measurement in complex environment. In view of the low identification rate of multiple nuclides caused by γ-ray energy spectrum shifting with temperature change,we propose a radionuclide identification method based on sparse representation and multi-task learning. Firstly,a transfer matrix was constructed to represent the environment variation affecting currently measured spectra. Then,the model of the measurement spectra was established,which was used to describe the instantaneous superposition of scale copies of individual nuclide sub-spectra in standard spectra library. Thus,the problem of radionuclide identification was transformed into the problem of sparse decomposition of various radionuclides. In order to solve this non-convex optimization problem,the multi-task learning method based on alternating direction multiplier method( ADMM)was developed to optimize the transfer matrix and decompose the sparse matrix simultaneously. The feasibility and effectiveness of the developed method were verified by some experiments,in which the measurement environment of CsI( Tl) detector was simulated by using the programmable temperature and humidity chamber and the real radioactive spectrum data of 11 kinds of nuclides and typical mixed nuclides were measured,respectively.Meanwhile,the single and mixed nuclide data of 27 kinds of nuclides were used,which are specified by the simulation IAEA of Monte Carlo analysis software Geant4. The experiment results show that the developed method can accurately identify a variety of commonly used nuclides even at temperature range of-20 ~ 50℃.
A spectra calibration method and a radionuclide identification method are developed for the improvement of the adaptability to the nuclear detector measurement in complex environment. In view of the low identification rate of multiple nuclides caused by γ-ray energy spectrum shifting with temperature change,we propose a radionuclide identification method based on sparse representation and multi-task learning. Firstly,a transfer matrix was constructed to represent the environment variation affecting currently measured spectra. Then,the model of the measurement spectra was established,which was used to describe the instantaneous superposition of scale copies of individual nuclide sub-spectra in standard spectra library. Thus,the problem of radionuclide identification was transformed into the problem of sparse decomposition of various radionuclides. In order to solve this non-convex optimization problem,the multi-task learning method based on alternating direction multiplier method( ADMM)was developed to optimize the transfer matrix and decompose the sparse matrix simultaneously. The feasibility and effectiveness of the developed method were verified by some experiments,in which the measurement environment of CsI( Tl) detector was simulated by using the programmable temperature and humidity chamber and the real radioactive spectrum data of 11 kinds of nuclides and typical mixed nuclides were measured,respectively.Meanwhile,the single and mixed nuclide data of 27 kinds of nuclides were used,which are specified by the simulation IAEA of Monte Carlo analysis software Geant4. The experiment results show that the developed method can accurately identify a variety of commonly used nuclides even at temperature range of-20 ~ 50℃.
引文
[1]ROSE P B,ERICKSON A S,MAYER M,et al. Uncovering special nuclear materials by low-energy nuclear reaction imaging[J].Scientific Reports,2016(6):24388. DOI:10. 1038/srep24388
[2]GRODZICKA M, MOSZYSKI M, SZCZSNIAK T. 5 Silicon photomultipliers in detectors for nuclear medicine[M]. Boca Raton:CRC Press,2015
[3] LECOQ P, ANNENKOV A, GEKTIN A, et al. Inorganic scintillators for detector systems:physical principles and crystal engineering[M]. Berlin:Springer,2016. DOI:10. 1007/3-540-27768-4
[4]SILVA J,FIORI E,ISAAK J,et al. Temperature gain correction for CsI(Tl)detection systems based on digital pulse shape analysis[J].Nuclear Instruments and Methods in Physics Research Section A:Accelerators,Spectrometers,Detectors and Associated Equipment,2015,776:98. DOI:10. 1016/j. nima. 2014. 12. 064
[5]ZAHN G S,GENEZINI F A,MORALLES M. Evaluation of peakfitting software for gamma spectrum analysis[EB]. Ar Xiv Eprint Ar Xiv:1511. 04362,2015
[6] OCZKOWSKI H L. Calibration standard for use in gamma spectrometry and luminescence dating[M]. Geochronometria:Journal on Methods&Applications of Absolute Chronology,2001
[7]OTTAWAY J,KALIVAS J H,ANDRIES E. Spectral multivariate calibration with wavelength selection using variants of Tikhonov regularization[J]. Applied Spectroscopy,2010,64(12):1388.DOI:10. 1366/000370210793561655
[8]KUNZ M R,OTTAWAY J,KALIVAS J H,et al. Impact of standardization sample design on Tikhonov regularization variants for spectroscopic calibration maintenance and transfer[J]. Journal of Chemometrics,2010,24(3/4):218. DOI:10. 1002/cem. 1302
[9]MERIC I,JOHANSEN G A,MOREIRA I V M. A library leastsquares approach for scatter correction in gamma-ray tomography[J].Radiation Physics and Chemistry,2015,108:39. DOI:10. 1016/j. radphyschem. 2014. 11. 013
[10]GARDNER R P,XU Libai. Status of the Monte Carlo library leastsquares(MCLLS)approach for non-linear radiation analyzer problems[J]. Radiation Physics and Chemistry,2009,78(10):843. DOI:10. 1016/j. radphyschem. 2009. 04. 023
[11]AGOSTINELLI S,ALLISON J,AMAKO K A,et al. GEANT4—a simulation toolkit[J]. Nuclear Instruments and Methods in Physics Research Section A:Accelerators,Spectrometers,Detectors and Associated Equipment,2003,506(3):250. DOI:10. 1016/S0168-9002(03)01368-8
[12]PAN S J,YANG Qang. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering,2010,22(10):1345. DOI:10. 1109/TKDE. 2009. 191
[13]HAN Xiaogang. Development of Monte Carlo code for coincidence prompt gamma-ray neutron activation analysis[D]. Raleigh:North Carolina State University,2005
[14]CHAN K-S,LI Jinzheng,EICHINGER W,et al. A new physicsbased method for detecting weak nuclear signals via spectral decomposition[J]. Nuclear Instruments and Methods in Physics Research Section A:Accelerators,Spectrometers,Detectors and Associated Equipment,2012,667(5):16. DOI:10. 1016/j. nima. 2011. 11. 067
[15]WESTMEIER W. Background subtraction in Ge(Li)gamma-ray spectra[J]. Nuclear Instruments and Methods,1981,180(1):205. DOI:10. 1016/0029-554X(81)90031-8
[16]ZHANG Zheng,XU Yong,YANG Jian,et al. A survey of sparse representation:algorithms and applications[J]. IEEE Access,2015,3:490. DOI:10. 1109/ACCESS. 2015. 2430359
[17]BOYD S,PARIKH N,CHU E,et al. Distributed optimization and statistical learning via the alternating direction method of multipliers[J]. Foundations and Trendsin Machine Learning,2011,3(1):1.DOI:10. 1561/2200000016
[2]GRODZICKA M, MOSZYSKI M, SZCZSNIAK T. 5 Silicon photomultipliers in detectors for nuclear medicine[M]. Boca Raton:CRC Press,2015
[3] LECOQ P, ANNENKOV A, GEKTIN A, et al. Inorganic scintillators for detector systems:physical principles and crystal engineering[M]. Berlin:Springer,2016. DOI:10. 1007/3-540-27768-4
[4]SILVA J,FIORI E,ISAAK J,et al. Temperature gain correction for CsI(Tl)detection systems based on digital pulse shape analysis[J].Nuclear Instruments and Methods in Physics Research Section A:Accelerators,Spectrometers,Detectors and Associated Equipment,2015,776:98. DOI:10. 1016/j. nima. 2014. 12. 064
[5]ZAHN G S,GENEZINI F A,MORALLES M. Evaluation of peakfitting software for gamma spectrum analysis[EB]. Ar Xiv Eprint Ar Xiv:1511. 04362,2015
[6] OCZKOWSKI H L. Calibration standard for use in gamma spectrometry and luminescence dating[M]. Geochronometria:Journal on Methods&Applications of Absolute Chronology,2001
[7]OTTAWAY J,KALIVAS J H,ANDRIES E. Spectral multivariate calibration with wavelength selection using variants of Tikhonov regularization[J]. Applied Spectroscopy,2010,64(12):1388.DOI:10. 1366/000370210793561655
[8]KUNZ M R,OTTAWAY J,KALIVAS J H,et al. Impact of standardization sample design on Tikhonov regularization variants for spectroscopic calibration maintenance and transfer[J]. Journal of Chemometrics,2010,24(3/4):218. DOI:10. 1002/cem. 1302
[9]MERIC I,JOHANSEN G A,MOREIRA I V M. A library leastsquares approach for scatter correction in gamma-ray tomography[J].Radiation Physics and Chemistry,2015,108:39. DOI:10. 1016/j. radphyschem. 2014. 11. 013
[10]GARDNER R P,XU Libai. Status of the Monte Carlo library leastsquares(MCLLS)approach for non-linear radiation analyzer problems[J]. Radiation Physics and Chemistry,2009,78(10):843. DOI:10. 1016/j. radphyschem. 2009. 04. 023
[11]AGOSTINELLI S,ALLISON J,AMAKO K A,et al. GEANT4—a simulation toolkit[J]. Nuclear Instruments and Methods in Physics Research Section A:Accelerators,Spectrometers,Detectors and Associated Equipment,2003,506(3):250. DOI:10. 1016/S0168-9002(03)01368-8
[12]PAN S J,YANG Qang. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering,2010,22(10):1345. DOI:10. 1109/TKDE. 2009. 191
[13]HAN Xiaogang. Development of Monte Carlo code for coincidence prompt gamma-ray neutron activation analysis[D]. Raleigh:North Carolina State University,2005
[14]CHAN K-S,LI Jinzheng,EICHINGER W,et al. A new physicsbased method for detecting weak nuclear signals via spectral decomposition[J]. Nuclear Instruments and Methods in Physics Research Section A:Accelerators,Spectrometers,Detectors and Associated Equipment,2012,667(5):16. DOI:10. 1016/j. nima. 2011. 11. 067
[15]WESTMEIER W. Background subtraction in Ge(Li)gamma-ray spectra[J]. Nuclear Instruments and Methods,1981,180(1):205. DOI:10. 1016/0029-554X(81)90031-8
[16]ZHANG Zheng,XU Yong,YANG Jian,et al. A survey of sparse representation:algorithms and applications[J]. IEEE Access,2015,3:490. DOI:10. 1109/ACCESS. 2015. 2430359
[17]BOYD S,PARIKH N,CHU E,et al. Distributed optimization and statistical learning via the alternating direction method of multipliers[J]. Foundations and Trendsin Machine Learning,2011,3(1):1.DOI:10. 1561/2200000016