Backpropagation and Levenberg-Marquardt Algorithm for Training Finite Element Neural Network

被引：63

作者：

Reynaldi, Arnold ^{[1
]}

Lukas, Samuel ^{[2
]}

Margaretha, Helena ^{[1
]}

机构：

[1] Pelita Harapan Univ, Fac Sci & Math, Tangerang, Banten, Indonesia

[2] Pelita Harapan Univ, Fac Comp Sci, Tangerang, Banten, Indonesia

来源：

2012 SIXTH UKSIM/AMSS EUROPEAN SYMPOSIUM ON COMPUTER MODELLING AND SIMULATION (EMS) | 2012年

关键词：

finite element method; artificial neural network; backpropagation algorithm; Levenberg-Marquardt algorithm; inverse differential problem;

D O I：

10.1109/EMS.2012.56

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

In this paper, finite element based neural network is developed. The purpose is to solve differential equation and inverse problem of differential equation. Inverse problem of differential equation is a problem to solve for parameters of differential equation, assuming that the solution of the differential equation is already known beforehand. Inverse problem mainly used to approximate physical parameters of material. Finite element method will be combined with artificial neural network using backpropagation algorithm to solve differential equation and Levenberg-Marquardt training algorithm to solve inverse differential problem. By using proposed method, invers matrix calculation will not be needed for solving both differential equation and inverse differential problem. From any given differential equation, the solution will be solved first. And the solution is used to validate the parameter in differential equation, namely to solve inverse problem of that differential equation.

引用

页码：89 / 94

页数：6

共 9 条

[1]

Fish J., 2007, A First Course in Finite Elements

[2] TRAINING FEEDFORWARD NETWORKS WITH THE MARQUARDT ALGORITHM [J].

HAGAN, MT ;

MENHAJ, MB .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1994, 5 (06) :989-993

[3]

Haykin S., 1999, Neural Networks: A Comprehensive Foundation, DOI DOI 10.1017/S0269888998214044

[4]

Mehrotra K., 1996, ELEMENT ARTIFICIAL N

[5] Finite-element neural networks for solving differential equations [J].

Ramuhalli, P ;

Udpa, L ;

Udpa, SS .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 2005, 16 (06) :1381-1392

[6]

Ramuhalli P., 2002, THESIS

[7]

Reddy J., 1984, An Introduction to the Finite Element

[8]

Sreekumar N., 2011, INT J COMPUTER SCI T, V2

[9] NEURAL-NETWORK REPRESENTATION OF FINITE-ELEMENT METHOD [J].

TAKEUCHI, J ;

KOSUGI, Y .

NEURAL NETWORKS, 1994, 7 (02) :389-395

← 1 →