Studies On γ-Ray Spectrometry In Determining Uranium Enrichment

Uranium enrichment measurement and attributes determination are important activities in MC&A system and routine safeguards inspections. y-Ray Spectrometry is important one of NDA methods widely used in nuclear safeguards. The y-Ray Spectrometry with self-calibration method and ANN(Artificial Neural Network) methods in uranium measurement are researched in this paper and a surrogate method in measuring shielded uranium samples is also provided.y-Ray Spectrometry is a passive non-destructive assay technology. It's a more accurate method in determining the isotopic abundance of uranium. The paper describes the measurement principles that allow accurate measurements to be taken on samples of arbitrary size,shape,and measurement geometry-and of arbitrary physical and chemical composition -through the use of know nuclear decay data(half-lives and branching intensities). This method relies on internal gamma-ray peaks from the spectrum under analysis to self-calibrate the unknown spectrum for energy and peak shape. PC/FRAM is a code based on this principle and developed by Los Alamos National Laboratory. The versatile-parameter database structure governs all facets of the data analysis. It can't be given a precise value by used of thedefault parameter files. The results of measuring the certified reference material of low enrichment uranium and the corrected parameter files are presented. With the corrected parameter files,the results indicate that the isotopic abundance in the sample can be determined within 2% error using a HPGe detector system.Artificial Neural Networks(ANN) are a class of models based on neural computation and have been applied to the measurement of uranium enrichment. The principles of ANN methods used in uranium measurement are presented in this paper and the conditions of analysis proceedings are described. ANN methods are feasible for the verification measurements in nuclear safeguards. Experimental data sets have been used to study the performance of neural networks involving Radial Basis Function neural network and Generalized Regression neural network(GRNN). The optimization of the parameter SPREADS have been given and the analysis error of GRNN no more than 0.2%. The encoding of the data by ANN approach is a promising method for the measurement of uranium samples.The determination of the shielded uranium,in particular,by passive gamma-ray detection is a formidable challenge. Gamma rays are ideal probes for determining the attributes of uranium in nuclear material transparency monitoring and verification of safeguards regime. It's hard to detect the attributes of inhomogeneous uranium in large compact containers with normal non-destructive assay methods. A non-destructive gamma ray assay method is provided in this paper. The attributes of uranium can be known by determining the high energy gamma ray 2614.75 keV. from 208T1,the daughter of 232U. And a confirmed test was done to verify the method.