Learning solutions to two dimensional electromagnetic equations using LS-SVM

Abstract In this paper, a new approach based on least squares support vector machines (LS-SVM) is proposed for solving the electromagnetic equations. Firstly, the cubic spline function is employed to smooth the discontinuous boundary. LS-SVM is used to solve the modified problem. Secondly, nonlinear electromagnetic equation is solved by LS-SVM. Finally, multimedia electromagnetic equation is solved by LS-SVM. Same as to the artificial neural networks (ANN), the approximate solutions are composed of two parts. The first part is a known function that satisfies the boundary conditions. The second part is the product of two terms. One term is also a known function which vanished on the boundary. The left part is the combination of kernel functions containing regression parameters. The parameters can be obtained by solving a system of equations. The numerical results show that the proposed method in this paper is feasible.

[1]  Hasan Alli,et al.  The solutions of vibration control problems using artificial neural networks , 2003, J. Frankl. Inst..

[2]  Mohsen Hayati,et al.  Multilayer perceptron neural networks with novel unsupervised training method for numerical solution of the partial differential equations , 2009, Appl. Soft Comput..

[3]  Johan A. K. Suykens,et al.  Parameter estimation of delay differential equations: An integration-free LS-SVM approach , 2014, Commun. Nonlinear Sci. Numer. Simul..

[4]  Johan A. K. Suykens,et al.  Approximate Solutions to Ordinary Differential Equations Using Least Squares Support Vector Machines , 2012, IEEE Transactions on Neural Networks and Learning Systems.

[5]  Johan A. K. Suykens,et al.  LS-SVM approximate solution to linear time varying descriptor systems , 2012, Autom..

[6]  Yan Li,et al.  Multi-Layer Support Vector Machine and its Application , 2006, 2006 International Conference on Machine Learning and Cybernetics.

[7]  A. R. Fonseca,et al.  The element-free Galerkin method in three-dimensional electromagnetic problems , 2006, IEEE Transactions on Magnetics.

[8]  Ju-Hong Lee,et al.  Comparison of generalization ability on solving differential equations using backpropagation and reformulated radial basis function networks , 2009, Neurocomputing.

[9]  Alaeddin Malek,et al.  Solving initial-boundary value problems for systems of partial differential equations using neural networks and optimization techniques , 2009, J. Frankl. Inst..

[10]  Kevin Stanley McFall,et al.  Automated design parameter selection for neural networks solving coupled partial differential equations with discontinuities , 2013, J. Frankl. Inst..

[11]  Euripidis Glavas,et al.  Solving differential equations with constructed neural networks , 2009, Neurocomputing.

[12]  Dimitris G. Papageorgiou,et al.  Neural-network methods for boundary value problems with irregular boundaries , 2000, IEEE Trans. Neural Networks Learn. Syst..

[13]  Johan A. K. Suykens,et al.  Learning solutions to partial differential equations using LS-SVM , 2015, Neurocomputing.

[14]  Yang Zou,et al.  Hybrid Approach of Radial Basis Function and Finite Element Method for Electromagnetic Problems , 2015, IEEE Transactions on Magnetics.

[15]  Kevin Stanley McFall,et al.  Artificial Neural Network Method for Solution of Boundary Value Problems With Exact Satisfaction of Arbitrary Boundary Conditions , 2009, IEEE Transactions on Neural Networks.

[16]  Li Fule,et al.  Approximate solutions to one-dimensional backward heat conduction problem using least squares support vector machines , 2016 .

[17]  Jin-Long An,et al.  Study on the solving method of electromagnetic field forward problem based on support vector machine , 2010, 2010 International Conference on Machine Learning and Cybernetics.

[18]  Satish S. Udpa,et al.  Finite-element neural networks for solving differential equations , 2005, IEEE Transactions on Neural Networks.

[19]  Dimitrios I. Fotiadis,et al.  Artificial neural networks for solving ordinary and partial differential equations , 1997, IEEE Trans. Neural Networks.