人工神经网络解析CdZnTe探测器γ谱

Method in analysis of CdZnTe γ spectrum with artificial neural network

半导体CdZnTe探测器具有体积小、分辨率高、可在室温下工作等优点而适用于野外便携式γ谱仪系统。由于CdZnTe晶体内电荷收集不完全导致γ谱产生低能尾巴,因而增加了常规谱解析的难度。采用了人工神经网络方法进行全γ谱法定性定量分析,可以充分利用γ谱所含信息,迅速准确地得出结果,避免了采用常规谱解析方法时低能尾巴对峰形拟合的影响。

The analysis of gamma-ray spectra to identify lines and their intensities usually requires expert knowledge and time consuming calculations with complex fitting functions. CdZnTe detector often exhibits asymmetric peak shape particularly at high energies making peak fitting methods and sophisticated isotope identification programs difficult to use. This paper investigates the use of the neural network to process gamma spectra measured with CdZnTe detector to verify nuclear materials. Results show that the neural network method gives advantages, in particular, when large low-energetic peak tailings are observed.