The Research On Dynamic Diagnostic Of Sluice Structural Damage

The sluice is an important public infrastructure of flood control security, water adjustment and storage of irrigation water. The effect of sluice can be compared to the front door of hydro-junction. Nevertheless, the working characteristic of sluice is comparative complexity. It is difficult to diagnose the local damage of sluice structure under the working condition with water. Therefore, it is very necessary to research the dynamic diagnosis about sluice structure damage by using intelligent method. The intelligent diagnosis model of sluice structure damage is established as the purpose of this paper. The method, using impulse technique, to obtain the modal parameters of small and medium-sized sluice is presented. The unitarity of mathematic model with experimental model of sluice structure is formed. Finally,the conclusion that the damage of physical model can instead of the experimental model damage is concluded.In this paper, the current research status of structural damage detection method and the advanced technology of hydraulic structural damage detection are discussed. The intelligent diagnosis method in the field of structure damage diagnosis is introduced also. In addition, the least squares support vector machine approach with kernel parameters based on HPSO optimization is proposed. Secondly, several pretreatment methods of vibration signal are expounded, and the signal with noise with the aid of Matlab language is constructed in order to check them. And the dynamic principle of structural damage detection is introduced, and the sensitive modal parameters to structural damage are summarized. Then the modal parameters of physical model of sluice structure are obtained by numerical calculation; meanwhile, the experimental model of sluice model is tested by the dynamic testing experiment; the modal parameters of experimental model are obtained by the experiment; the numerical calculation method is equivalent with the experimental method through contrasting the results by these methods. In next chapter, the basic principle of LS-SVM method with kernel parameters using the HPSO optimization method is introduced, and the model is integrated into the problem about diagnosing structure damage. The dynamic diagnosis model based on HPSO-SVM method is formed. According to the incomplete monitoring data of structure damage, a new diagnostic method based on HPSO-SVM pattern recognition is proposed. All these models can quickly and accurately lock structure damage through the proof of damage samples of concrete beam. While sluice structure is in different damage situations, the modal parameters of physical model are obtained by numerical calculation. The influence of the structure about different positions and degrees of damage are analyzed. The dynamic diagnosis model of sluice structure damage is established based on HPSO-SVM pattern recognition according to the input and output of damage samples of sluice as intelligent diagnostic methods.In summary, the proposed numerical calculation method can get structure damage samples under the situation of the structure without damage in this paper; the established dynamic diagnosis model based on HPSO-SVM pattern recognition can accurately diagnose the damage of sluice structure.