摘要
针对泊松方程的数值解,提出了一种基于截断奇异值分解(TSVD)的正则化和径向基函数(RBF)的改进的无网格方法.由于通过RBF拟合方程所产生的系数矩阵经常是病态的,TSVD正则化方法可以改善RBF无网格方法而获得更精确的数值解,与传统的RBF方法相比能够获得更好的数值结果,而且通过选择恰当的径向基函数,也能够提高数值解的精度.
An improved gridless method based on radical basis function( RBF) for the numerical solution of Poisson' equation is proposed. Since the coefficient matrix generated by the RBF approximation is usually ill-conditioned,the truncated singular value decomposition( TSVD) regularization method is used to obtain a more accurate numerical solution. Compared to common RBF,better numerical results will be achieved. What's more,the accuracy of numerical solution can be improved by choosing proper radial basis functions.
出处
《江西师范大学学报(自然科学版)》
CAS
北大核心
2017年第1期42-45,共4页
Journal of Jiangxi Normal University(Natural Science Edition)
基金
国家自然科学基金(61271337)资助项目
关键词
TSVD正则化
径向基函数
泊松方程
TSVD regularization
radial basis function
Poisson's equation