You are here

Home » Content

COMPUTATIONAL INVESTIGATION ON THE USE OF FEM AND MLP NEURAL NETWORK IN THE INVERSE PROBLEM OF DEFECTS IDENTIFICATIONS

Journal Name:

  • Istanbul University Journal of Electrical & Electronics Engineering

Publication Year:

  • 2008

Key Words:

Author NameUniversity of Author
Abstract (2. Language): 
In this article an attempt is made to study the applicability of a general purpose, supervised feed forward neural network, namely multilayer perceptron (MLP) neural network and finite element method (FEM) to solve the inverse problem of defect identification. The approach is used to identify unknown defects in metallic walls. The methodology used in this study consists in the simulation of a large number of defects in a metallic wall, using the finite element method. Both variations in with and height of the defects are considered. Then, the obtained results are used to generate a set of vectors for the training of MLP neural network model. Finally, the obtained neural network is used to identify a group of new defects, simulated by the finite element method, but not belonging to the original dataset. Noisy data, added to the probe measurements is used to enhance the robustness of the method. The reached results demonstrate the efficiency of the proposed approach, and encourage future works on this subject.
Bookmark/Search this post with
FULL TEXT (PDF): 
PDF icon arastirmax-computational-investigation-use-fem-and-mlp-neural-network-inverse-problem-defects-identifications.pdf
  • 1
537-548
  • English

REFERENCES

References: 

[1] Ramuhalli. P., Neural network based
iterative algorithms for solving electromagnetic
NDE inverse problems, Ph.D. dissertation, Dept.
Elect. Comp. Eng, Iowa Univ, USA, 2002.
[2] Coccorese. E., Martone. R., Morabito. F. C.,
"A neural network approach for the solution of
electric and magnetic inverse problems", IEEE
Trans. Magnetics, Vol: 30, No: 5, pp. 2829-
2839, 1994.
[3] Hoole. S. R. H., "Artificial neural networks
in the solution of inverse electromagnetic field
problems", IEEE Trans. Magnetics, Vol: 29, No:
2, pp. 1931-1934, 1993.
[4] Ramuhalli. P., Udpa. L., Udpa. S. S., "Finite
element neural networks for solving differential
equations", IEEE Trans. Neural Networks, Vol:
16, No: 6, pp. 1381-1392, 2005.
[5] Wong. P. M., Nikravesh. M., "Field
applications of intelligent computing
techniques", J. Petrol Geolog, Vol: 24, No: 4,
pp. 381-387, 2001.
[6] Fanni. A., Montisci. A., "A neural inverse
problem approach for optimal design", IEEE
Trans. Magnetics, Vol: 39, No: 3, pp. 1305-
1308, 2003.
[7] Haykin. S., Neural networks: A
comprehensive foundation, Englewood Cliffs,
NJ: Prentice-Hall, New York, 1999.
[8] Turchenko. I. V., "Simulation modelling of
multi-parameter sensor signal identification
using neural networks", Proc. 2nd IEEE Int Conf.
Intelligent Systems, Bulgaria, 2004, pp. 48-53.
[9] De Alcantara. N.P., Alexandre. J., De
Carvalho. M., "Computational investigation on
the use of FEM and ANN in the non-destructive
analysis of metallic tubes", Proc. 10th Biennial
Conf. Electromagnetic Field Computation, Italy,
2002.
[10] Jain. A. K., Mao. J., Mohiuddin. K. M.,
Artificial neural networks: a tutorial, Computer,
pp. 31-44, 1996.
[11] Mehrotra. K., Mohan. C. K., Ranka. S.,
Elements of artificial neural networks, MA: MIT
Press, Cambridge, 1997.
[12] Cherubini. D., Fanni. A., Montisci. A.,
Testoni. P., "A fast algorithm for inversion of
MLP networks in design problems", COMPEL.
Int. J. Comp and Math in Electric and Electro
Eng, Vol: 24, No: 3, pp. 906-920, 2005.
[13] Chari. M. V. K., Salon. S. J., Numerical
methods in electromagnetism, CA: Academic,
San Diego, 2000.
COMPUTATIONAL INVESTIGATION ON THE USE OF FEM AND MLP NEURAL NETWORK IN THE INVERSE PROBLEM
OF DEFECTS IDENTIFICATIONS
T. HACIB ,M. R. MEKIDECHE , N. FERKHA
548
[14] Silvester. P. P. and Ferrari. R. L., Finite
elements for electrical engineers, Univ Press,
Cambridge, 1996.
[15] Raida. Z., Modeling EM tructures in the
neural network toolbox of MATLAB, IEEE
Antenna’s and propagation Magazine, Vol: 44,
No: 6, pp. 46-67, 2002.
[16] Partial Differential Equation Toolbox
user’s guide, for use with MATLAB, The Math
Works Inc.
[17] Han. W., Que. P., "2D defect
reconstruction from MFL signals based on
genetic optimization algorithm", Proc. 1st IEEE
Int Conf. Industrial Technology, China, 2005,
pp. 508-513.
[18] Chady. T., Enokizono. M., Sikora. R.,
Todaka. T., Tsuchida. Y., "Natural crack
recognition using inverse neural model and
multi-frequency eddy current method", IEEE
Trans. Magnetics, Vol: 37, No: 4, pp. 2797-
2799, 2001.
[19] Hagan. M. T., Menhaj. M., "Training
feed-forward networks with the Levenberg-
Marquardt algorithm", IEEE Trans Neural
Networks, Vol: 5, No: 6, pp. 989-993, 1994.

Thank you for copying data from http://www.arastirmax.com