期刊文献+

面向轻量级新工具链:GNDS-JSON格式核数据的共振重建

New Lightweight Toolchain:Resonance Reconstruction of GNDS-JSON Formatted Nuclear Data
下载PDF
导出
摘要 为了验证基于GNDS-JSON格式处理评价核数据能否用于粒子输运模拟,构建了一个新的轻量级工具链。新工具链无须先将ENDF-6或GNDS格式的数据转换为自定义的内部数据结构,可以直接在读入的GNDS-JSON格式数据上进行处理。作为该轻量级工具链的第一步,成功实现了可分辨共振截面重建功能,并在ENDF/B-VIII.0数据库上与NJOY程序进行了比较。计算结果表明,除少数核素外,重建截面的最大相对偏差小于10^(-5),均方根相对偏差约为10^(-6),验证了基于GNDS-JSON打造轻量级新工具链的可行性。同时也发现了自研程序和Fudge程序在处理40 Ca的复杂R-Matrix形式共振参数时的错误。工具链的轻量化可为核数据处理工具的敏捷开发提供参考。 The ENDF-6 format has been the current standard for nuclear data,while the more free formatted GNDS format has been proposed,which provides more friendly access for human and machine.In order to solve the problem of the XML-based GNDS format being discrepant from the data structural model of programming languages,the GNDS-JSON format is further proposed.It can reduce the complexity in the development of nuclear data processing tools.To verify the possibility of nuclear data processing from GNDS-JSON format for particle transport,a lightweight toolchain is formulated.Due to the widespread application of JSON format in the field of cloud computing,it has mature and efficient data encoding and decoding programming libraries,where the accessed data are organized in the form of such structures and arrays as in the C programming language.Therefore,the proposed new toolchain is unlike traditional tools that require first conversion of ENDF-6 or GNDS formatted data into internal data structures.Instead,it can directly access and process data in GNDS-JSON format.The new toolchain successfully implements the reconstruction of resolved resonance cross sections,and is verified with the NJOY code on the ENDF/B-VIII.0 library.Except for a few nuclides,the maximum relative deviation of the reconstructed cross section is less than 10^(-5),and the root mean square relative deviation is about 10^(-6).It hence verifies the feasibility of building a new lightweight toolchain based on GNDS-JSON.In the meanwhile,the deviation in the implementation of in-house and Fudge code from the NJOY code for reconstructing 40 Ca with complicated R-Matrix formatted resonance parameters is discovered.
作者 刘畅源 LIU Changyuan(New Compute Laboratory,Beijing 100025,China)
机构地区 新计算实验室
出处 《现代应用物理》 2024年第1期34-43,58,共11页 Modern Applied Physics
关键词 GNDS-JSON ENDF-6 共振截面 R-MATRIX GNDS-JSON ENDF-6 resonance cross section R-Matrix
  • 相关文献

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部