Vibration Modal Analysis And Neural Network Technology-based Structural Damage Identification

In recent years the Structure Health Monitoring (SHM) has become one of the hottest topics both in engineering world and academic world. The major research work has been done in the identifications, locations and severities of structure damage. Research of structure damage diagnosis is the key point and difficult part of structure health monitoring. Due to the characteristic suitable to solve the problem of structure damage diagnosis, the technique of vibration modal analysis and artificial neural network become two active research fields.Based on the systematic summary and categorization of domestic and international researches on structure damage diagnosis, the research of combination of the techniques of vibration modal analysis and artificial neural network in structure damage diagnosis is carried on in the paper. A plate model and a simple wing model are chosen as the research object, on which damages of different location and severity are simulated. Modal parameters are obtained by the modal analysis. Three damage indices of different categories are made as the plate model input parameter of neural network to train the network and identify the damage and two damage indices for wing model. The result of damage identification is satisfactory. Dependent on the analysis of the result, the sensitivity of different damage indices is compared and the conclusion that the incompletion of modal information have nearly no effect to damage identification is drawn, which have certain practical engineering significance. At last, the method of combination of structure dynamic characteristics analysis and neural network in structure damage diagnosis is summarized and the future research direction is presented in the paper.