期刊文献+

基于Burgers向量提取与分析的改进PRD算法在位错环模拟中的应用

Application of Improved PRD Algorithm Based on Burgers Vector Extraction and Analysis in Dislocation Loop Simulation
下载PDF
导出
摘要 RAFM钢经中子辐照后会产生两种类型的间隙位错环(1/2〈111〉环和〈100〉环),导致长时间服役后出现辐照硬化与脆化现象,因此理清不同类型位错环的形成与转变过程对材料微结构演化及性能预测具有重要意义。本文基于加速分子动力学中的并行副本复制法(PRD),通过分析位错环转变的机理特性,提取Burgers向量,改进了PRD在位错环模拟上的应用,实现了〈100〉环向1/2〈111〉环的加速转变现象模拟。结果表明:在1000 K下,采用改进PRD算法程序进行模拟,小尺寸〈100〉环会发生向1/2〈111〉环的转变,转变时间比传统分子动力学方法更快;大尺寸〈100〉环会出现〈100〉位错段和1/2〈111〉位错段共存的状态,甚至在带有Ni溶质元素的〈100〉环中还发现了完全转变到1/2〈111〉位错环的现象。此外,对程序模拟的准确性进行验证并分析评估了加速性能与并行性能,通过与原PRD算法对比,发现其并行效率远高于原PRD,并行效率在90%左右,且并行规模的增加对于并行性的影响不大,可以很好地进行大规模并行扩展。 After neutron irradiation,RAFM steel will produce two types of interstitial dislocation loops(1/2〈111〉loop and〈100〉loop),resulting in radiation hardening and embrittlement after long service.And different types of dislocation loops may lead to different hardening phenomena.However,the formation and transformation processes of different types of dislocation loops are still obscure.Therefore,it is of great significance to clarify the formation and transformation processes of different types of dislocation loops for the evolution of materials’microstructures and prediction of their properties.In this paper,we presented an improved parallel replica method(PRD),which improved the application of the original PRD method in dislocation loop evolution simulation.Through analyzing the mechanism characteristics of dislocation loop transition and extracting Burgers vector,the improved PRD method accurately identified the state of dislocation loops,and then calculated the simulation of accelerated transition process from〈100〉loop to 1/2〈111〉loop.By combing the dislocation extraction algorithm in OVITO,the synchronous analysis of Burgers vector in the simulation process was rea-lized.During the dislocation loop transforms,its habit plane will change and eventually affect the change of Burgers vector.So Burgers vector was used to redefine the adjudication for transition events in the original PRD method.We adjudicated whether the system undergoes a transition event,based on the existence of 1/2〈111〉Burgers vector in the simulated system and whether the proportion of 1/2〈111〉Burgers vector increases.With such improvements,it would help us avoid lengthy cyclic iterations at some metastable state in our simulation and realize the simulation of accelerated transition phenomenon of〈100〉loop to 1/2〈111〉loop.Finally,numerical experiments were performed to evaluate the effect of our work.The results show that the improved PRD algorithm successfully simulate the changing process of small size〈100〉loop to 1/2〈111〉loop at 1000 K,and the transition time is faster than the traditional molecular dynamics method.We also observe the coexistence phenomenon of〈100〉dislocation segments and 1/2〈111〉dislocation segments in large size〈100〉loop simulation.So far as to the case of〈100〉loop with Ni solute elements,it is found that the complete transition to 1/2〈111〉dislocation loop is happened.In addition,the accuracy of program simulation was verified,and the parallel performance were also analyzed and evaluated in this paper.We achieve about 90%parallel efficiency in the parallel calculating test,which is found much higher than the original PRD method.Moreover,our program has well-fined scalability,so it can be well expanded for the large-scale parallel calculation.
作者 徐鑫铖 聂宁明 王瑾 贺新福 曾艳 张纪林 王珏 万健 XU Xincheng;NIE Ningming;WANG Jin;HE Xinfu;ZENG Yan;ZHANG Jilin;WANG Jue;WAN Jian(Hangzhou Dianzi University,Hangzhou 310018,China;Computer Network Information Center,Chinese Academy of Sciences,Beijing 100190,China;China Institute of Atomic Energy,Beijing 102413,China)
出处 《原子能科学技术》 EI CAS CSCD 北大核心 2023年第8期1614-1624,共11页 Atomic Energy Science and Technology
基金 国家自然科学基金(U1867217) 国家重点研发计划(2018YFE0308104)。
关键词 位错环 Burgers向量 加速分子动力学 大规模并行计算 位错环转变 dislocation loop Burgers vector accelerated molecular dynamics large scale parallel computing transition of dislocation loop
  • 相关文献

参考文献1

二级参考文献3

共引文献4

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