基于支持向量回归的边值问题求解方法

张慧菁; 莫艳

doi:10.12052/gdutxb.230121

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于支持向量回归的边值问题求解方法

微分方程及其应用 | 更新时间：2026-05-27

- 基于支持向量回归的边值问题求解方法
- Support Vector Regression Based Method to Solve Boundary Value Problems
- 广东工业大学学报 2024年41卷第5期页码：111-118
- 作者机构：
  
  广东工业大学数学与统计学院,广东,广州,510520
- 作者简介：
  
  张慧菁(1998–)，女，硕士研究生，主要研究方向为计算及应用调和分析，E-mail：zhanghj9801@163.com
  莫艳(1986–)，女，副教授，主要研究方向为计算及应用调和分析，E-mail：marvel-2008@163.com
- 基金信息：
  
  广州市科技计划项目(202102020704)
- DOI：10.12052/gdutxb.230121
  中图分类号： O175.8
- 网络首发：2024-09-25，
  
  纸质出版：2024
- 稿件说明：
移动端阅览
张慧菁, 莫艳. 基于支持向量回归的边值问题求解方法[J]. 广东工业大学学报, 2024, 41(5): 111-118.

Zhang Hui-jing, Mo Yan. Support Vector Regression Based Method to Solve Boundary Value Problems[J]. 2024, 41(5): 111-118.
张慧菁, 莫艳. 基于支持向量回归的边值问题求解方法[J]. 广东工业大学学报, 2024, 41(5): 111-118. DOI： 10.12052/gdutxb.230121.

Zhang Hui-jing, Mo Yan. Support Vector Regression Based Method to Solve Boundary Value Problems[J]. 2024, 41(5): 111-118. DOI： 10.12052/gdutxb.230121.

摘要

边值问题是方程问题中的一个热门领域，对常微分方程边值问题的研究已相当成熟。然而，当已知条件为离散点而非给定函数时如何求解边值问题仍有很大的研究空间。支持向量回归机是一种基于统计学习理论的机器学习方法，在逼近问题上有独特的优势，在经验风险最小化同时保证了泛化能力。因此，本文结合正则化、再生核理论和支持向量回归机对边值问题展开研究。将边值问题视为算子方程问题，利用再生核空间的性质得到方程解与已知条件的关系式，将问题转化为逼近问题后正则化为二次规划问题，然后利用支持向量回归机进行求解，最终得到一个由支持向量构成的稀疏解。通过Sobolev空间中的范数关系对所得数值解进行误差分析，给出了数值解与解析解的误差上界。以二阶三点边值问题为例，仅有离散值作为已知条件求解方程，实验结果表明该方法优于传统再生核方法和W-POAFD方法，验证了该方法的高精度和有效性。

Abstract

Boundary value problems are a hot research area in the field of equation problems

with the study of boundary value problems for ordinary differential equations being quite mature. However

there is significant research space in solving boundary value problems when the known conditions are discrete points rather than given functions. Support Vector Regression (SVR) is a machine learning method based on statistical learning theory

which shows unique advantages in approximation problems by minimizing empirical risk while ensuring generalization capability. Therefore

this paper combines regularization

reproducing kernel theory

and SVR to investigate boundary value problems. Treating the boundary value problem as an operator equation problem

the relationship between the solution of the equation and the known conditions is obtained using the properties of reproducing kernel spaces. The problem is then transformed into an approximation problem

regularized into a quadratic programming problem

and solved using SVR to obtain a sparse solution composed of support vectors. Error analysis of the numerical solution obtained is conducted using norms in Sobolev spaces

providing an upper bound for the error between the numerical solution and the analytical solution. Taking a second-order three-point boundary value problem as an example

where only discrete values are given as known conditions for solving the equation

experimental results demonstrate that this method outperforms traditional reproducing kernel methods and W-POAFD methods

confirming its high accuracy and effectiveness.

关键词

Keywords

references

ILIN V A, MOISEEV E I. Nonlocal boundary value of the first kind for a sturm-liouville operator in its differential and finite difference aspects [J]. Differential Equations, 1987, 23: 803-810.

MOHAMED M S, SEAID M, BOUHAMIDI A. Iterative solvers for generalized finite element solution of boundary value problems [J]. Numerical Linear Algebra with Applications, 2018, 25(6): e2205.

GEORGE K, TWIZELL E H. Stable second-order finite-difference methods for linear initial-boundary-value problems [J]. Applied Mathematics Letters, 2006, 19(2): 146-154.

GROZA G, POP N. Approximate solution of multipoint boundary value problems for linear differential equations by polynomial functions [J]. Journal of Difference Equations and Applications, 2008, 14(12): 1289-1309.

MCFALL K S, MAHAN J R. Artificial neural network method for solution of boundary value problems with exact satisfaction of arbitrary boundary conditions [J]. IEEE Transactions on Neural Networks, 2009, 20(8): 1221-1233.

BEIDOKHTI R S, MALEK A. Solving initial-boundary value problems for systems of partial differential equations using neural networks and optimization techniques [J]. Journal of the Franklin Institute, 2009, 346(9): 898-913.

CYBENKO G. Approximation by superpositions of a sigmoidal function [J]. Mathematics of Control, Signals and Systems, 1989, 2(4): 303-314.

CORTES C, VAPNIK V. Support-vector networks [J]. Machine Learning, 1995, 20: 273-297.

VAPNIK V, GOLOWICH S, SMOLA A. Support vector method for function approximation, regression estimation and signal processing [J]. Advances in Neural Information Processing Systems, 1996, 9: 281-287.

富坤, 汪友华, 何平, 等. 应用支持向量机回归求解不规则边界边值问题[J]. 河北省科学院学报, 2005, 022(z1): 92-93.

MO Y, QIAN T. Support vector machine adapted tikhonov regularization method to solve dirichlet problem [J]. Applied Mathematics and Computation, 2014, 245: 509-519.

WANG D R, MO Y. Support vector machines based method to solve discrete linear differential equations [J]. Journal of Guangdong University of Technology, 2020, 37(2): 87-93.

LIN Y Z, NIU J, CUI M G. A numerical solution to nonlinear second order three-point boundary value problems in the reproducing kernel space [J]. Applied Mathematics and Computation, 2012, 218(14): 7362-7368.

LYU X Q, CUI M G. Existence and numerical method for nonlinear third-order boundary value problem in the reproducing kernel space [J]. Boundary Value Problems, 2010, 2010: 1-19.

CUI M G, LIN Y Z. Nonlinear numerical analysis in reproducing kernel space[M]. New York: Nova Science Publishers, Inc. , 2009.

GAO E, SONG S H, ZHANG X J. Solving singular second-order initial/boundary value problems in reproducing kernel hilbert space [J]. Boundary Value Problems, 2012(1): 1-11.

CASTRO L, SAITOH S, SAWANO Y, et al . Discrete linear differential equations [J ] . International Mathematical Journal of Analysis Its Applications, 2012, 32(3): 181-191.

CASTRO L P, SAITOH S, SAWANO Y, et al . General inhomogeneous discrete linear partial differential equations with constant coefficients on the whole spaces [J ] . Complex Analysis and Operator Theory, 2012, 6(1): 307-324.

QIAN T. Reproducing kernel sparse representations in relation to operator equations [J]. Complex Analysis and Operator Theory, 2020, 14(2): 36.

BAI H F, LEONG I T, DANG P. Reproducing kernel representation of the solution of second order linear three-point boundary value problem [J]. Mathematical Methods in the Applied Sciences, 2022, 45(17): 11181-11205.

RIEGER C, ZWICKNAGL B. Deterministic error analysis of support vector regression and related regularized kernel methods [J]. Journal of Machine Learning Research, 2009, 10(9): 2115-2132.

BRENNER S C. The mathematical theory of finite element methods[M]. New York: Springer, 2008.

WENDLAND H, RIEGER C. Approximate interpolation with applications to selecting smoothing parameters [J]. Numerische Mathematik, 2005, 101: 729-748.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于主成分分析与支持向量回归的精明增长建模与预测

面向灾后救援的多无人机通感算一体化资源分配优化方法

基于数字混频的脉冲式激光测距方法的研究与实现

双模态迭代交叉注意力融合集成框架

融合沟道倾斜角的3D NAND存储单元建模及编程策略优化