Experimental Examples for Identification of Structural Systems Using Neural Network and DOF-Based Reduction Method

Abstract

Various identification methods have been introduced in many manners using numerical techniques to validate a complicated structures described in FEM by comparing experimentally measured modal data. The objection of this study is to propose a methodology to identify a perturbed structure by comparing measured modal data to the original FEM data. Identified structures will improve the accuracy to the numerical model by minimizing the differences between those two models. Base-line model will be constructed by using FEM and will be compared to perturbed model by solving inverse problem. Measured modal responses, which are eigenvalues and eigenvectors, will be applied to satisfy the equilibrium and to minimize the differences of modal responses between the original model and the perturbed model. In this study, a neural networks-based detection method using modal properties is presented as a method for the identification which can effectively consider the modeling errors. Also experimental examples will be followed. Due to lack of number of the sensors, DOF-based reduction method is used to restore full model. As neural network is used for identification method, detailed schemes will be reported. In the present study, neural networks-based identification method will be proposed and will be verified by experimental examples. Experimental examples will demonstrate that the proposed method have efficiencies in accuracy of identifying structural model.