摘要
鉴于地质样品岩性变化大，容易引起对Ｘ－射线荧光分析准确度的影响，尤其是主、次量元素受影响更大，因此，作者提出了岩性自动分类，交互有效－基本参数法分析地质样品中主、次量元素，并与原有的经验系数和其它方法进行了比较．结果表明本方法的准确度更好些。
The change of sample category in geological analysis, with the particle size and matrix effects, has the significant effect on the analytical accuracy of major and minor elements based on x ray fluorescence spectroscopy. An algorithm arranging geological samples under categories automatically by a computer program was proposed. It makes use of the category identification and spline function, and has been combined with Cross Validation and fundamental parameters. Compared with the empirical coefficient, hybrid and modified α coefficient methods, the results show that the proposed algorithm has higher flexibility to suit the change of sample types, and the method is of better stability resulted from the noise filtering effect by using cross validation method and of much better prediction accuracy.
出处
《分析科学学报》
CAS
CSCD
1996年第3期189-193,共5页
Journal of Analytical Science
基金
国家自然科学基金资助项目
