Third M.I.T. Conference on Computational Fluid and Solid Mechanics June 14–17, 2005  

Coupled evolutionary algorithm and artificial neural network in defects identification

T. Burczyńskia,b, A. Skrobola,*
aDepartment for Strength of Materials and Computational Mechanics, Silesian University of Technology, Konarskiego 18A, Gliwice 44-100, Poland bInstitute of Computer Modelling, Artificial Intelligence Division, Cracow University of Technology, Cracow, Poland

  Full Text
This paper is devoted to a method based on computational intelligence for non-destructive defect identification. In the paper an elastic body with an unknown number of internal defects is considered.

The Evolutionary Algorithm (EA) is combined with the Artificial Neural Network (ANN) into one computational intelligence system. The EA is applied to identify the number of defects and their parameters by minimizing the fitness function, which is expressed as a difference between measured and computed displacements on the boundary and a difference between measured and computed eigenfrequences of the investigated structure. The fitness function is computed by means of a Fuzzy-Artificial Neural Network (FANN).

Keywords:  Identification; Evolutionary algorithm; Fuzzy artificial neural network; Boundary element method; Finite element method; Internal defect

* Corresponding author. Tel.: +48 (32) 237 1074; Fax: +48 (32) 237 1282; E-mail: