A New Constructing Rational Functions Method For Solving Lane-Emden Type Equations

被引：3
|
作者：
He, Jilong [1 ]
Zheng, Zhoushun [1 ]
Du, Changfa [1 ]
机构：
[1] Cent South Univ, Sch Math & Stat, Changsha 410083, Hunan, Peoples R China
来源：
NEURAL PROCESSING LETTERS | 2023年 / 55卷 / 02期
基金：
中国国家自然科学基金;
关键词：
Rational Chebyshev polynomials; Lane-Emden equation; Error back propagation algorithm; Machine learning; NEURAL-NETWORK; NUMERICAL-SOLUTION; DIFFERENTIAL-EQUATIONS; COLLOCATION APPROACH; WAVELETS METHOD; ALGORITHM; PHYSICS; MODEL;
D O I：
10.1007/s11063-022-10968-6
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
In this paper, a new construction machine learning method is proposed based on Rational Chebyshev polynomials to solve Lane-Emden type differential equations. The method is used for training, and the parameters are obtained by the error back propagation principle. By comparing with Chebyshev neural network proposed in [Appl. Math. Comp. 247:100-114(2014)][Appl. Soft. Comp. 43:347-356(2016)], our results are more accurate.
引用
收藏
页码：1889 / 1918
页数：30
相关论文
共 50 条
12345