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

Identification of stiffness reduction in beams using parameter-dependent frequency changes and neural networks

A. Borowiec*, L. Ziemiański
Rzeszów University of Technology, Department of Structural Mechanics, ul. W. Pola 2, PL-35959 Rzeszów, Poland

  Full Text
This paper presents the application of artificial neural networks (ANN) in the identification of damage in simple engineering structures. Three numerical examples are presented considering the identification of damage in a cantilever beam with and without an additional support. The application of ANNs expands the nondestructive damage identification method using an additional parameter introduced to the structure. The input vector of the ANNs consists of the dynamic responses of a structure with additional mass. The output vector is composed of the position of damage and, in the last example, the extent of damage.

Keywords:  Artificial neural networks; Identification; Dynamics; FEM

* Corresponding author. Tel.: + 4817 865 1535; Fax: + 4817 865 1173; E-mail: