摘要
通过介绍一类线性微分方程两点边值问题的神经网络方法,分析神经网络模型中网络拓扑结构及样本量对计算结果的影响,并通过数值实验进行验证,为该方法在求解两点边值问题的推广提供参考.
A neural network method for a class of two-point boundary value problems of linear differential equations was introduced in this paper,the influence of the network topology and sample size in the neural network model on the calculation results was analyzed and then verified with numerical experiments,which provided a reference for the application of this method in solving two-point boundary value problems.
作者
杨云磊
张天乐
侯木舟
罗建书
YANG Yunlei;ZHANG Tianle;HOU Muzhou;LUO Jianshu(School of Mathematics and Statistics,Central South University,Changsha 410083,China;College of Science,National University of Defense Technology,Changsha 410073,China)
出处
《徐州工程学院学报（自然科学版）》
CAS
2018年第2期57-63,共7页
Journal of Xuzhou Institute of Technology(Natural Sciences Edition)
基金
国家自然科学基金项目(61375063
61271355
11301549
11271378)
关键词
两点边值问题
神经网络方法
网络拓扑结构
样本量
two-Gpoint boundary value problem
neural network method
net workt opology
samplesize
