Study on the structural damage identification method with combined parameters based on RBF neural network

Localized damage in a structure affects its dynamic properties. The change is characterized by changes in the eigenparameters such as natural frequencies, the mode shapes associated with each natural frequency etc. and much work has been undertaken by investigating the single parameter which is assigned as input parameter to neural network to determine the damage location and the damage size. In this paper, a structural damage identification method with combined parameters based on RBF neural network is presented. To overcome the disadvantages of single input parameter, combined parameters, which are obtained by combining natural frequencies, mode shape data and changes in curvature mode shape, are assigned as input parameters to neural network. The simulation results to a lumped-mass system of six degrees show that the method is effective and applicable.