Application of Hierarchical Matrices to Large-Scale Electromagnetic Field Analyses of Coils Wound With Coated Conductors
Application of Hierarchical Matrices to Large-Scale Electromagnetic Field Analyses of Coils Wound With Coated Conductors
复制标题
分层矩阵在涂层导体绕制线圈大规模电磁场分析中的应用
DOI：
10.1109/tasc.2017.2780821
复制
发表时间：
2018
影响因子：
1.8
通讯作者：
Amemiya Naoyuki
中科院分区：
文献类型：
--
作者：
Tominaga Naoki;Mifune Takeshi;Ida Akihiro;Sogabe Yusuke;Iwashita Takeshi;Amemiya Naoyuki
由于采用电流矢量势的涂层导体电磁场分析模型中系数矩阵密集，因此大型线圈的分析需要大量的内存和较长的计算时间。在本文中，我们引入层次矩阵使用HACApK库，以减少内存消耗和计算时间。在数值试验中，使用一个扁平线圈缠绕涂层导体的方法相对于内存消耗和计算时间的效果进行检查。并行计算的效果也检查。最后，我们分析了一个大规模的磁铁，有150万个未知值。
Because the coefficient matrix is dense in the electromagnetic-field-analysis models of coated conductors using current vector potentials, the analyses of large-scale coils require a large amount of memory and long computation time. In this paper, we introduce hierarchical matrices by using the HACApK library in order to reduce both memory consumption and computation time. The effect of the method with respect to memory consumption and computation time is examined in numerical tests using a pancake coil wound with a coated conductor. The effect of parallel computations is also examined. Finally, we analyzed a large-scale magnet, with 1.5 million unknown values.