求解线性积分方程的压缩投影离散的Landweber算法  

A Compressed Multiscale Projection Method for Solving Linear Integral Equtions Via Landweber Method

作　　者：罗兴钧[1] 黄静月 张荣[1] LUO Xing-jun;HUANG Jing-yue;ZHANG Rong(School of Mathematics and Computer Science,Gannan Normal University,Ganzhou 341000,Jiangxi,China)

机构地区：[1]赣南师范大学数学与计算机科学学院,江西赣州341000

出　　处：《韶关学院学报》2021年第12期1-7,共7页Journal of Shaoguan University

基　　金：国家自然科学基金资助项目(11761010).

摘　　要：研究了求解线性积分方程的多尺度压缩投影离散的Landweber算法,提出了Landweber迭代终止指标的停止准则,与传统的投影算法比较,减少了计算量,确保了近似解的最优收敛性.算例表明了该方法的有效性.In this paper,a compressed multiscale projection method for solving linear integral equations is studied and a stopping rule is proposed to choose the termination index of Landweber iteration.Compared with the traditional projection algorithm,it reduces the amount of calculation and ensures the optimal convergence of the approximate solution.Numerical examples are given to demonstrate the efficiency and accuracy of the proposed.

关 键 词：线性积分方程 多尺度压缩投影方法 参数选择策略 Landweber迭代 

分 类 号：O242.2[理学—计算数学]