Modal Parameters Identification of Bridge Structure Based on Improved Hilbert-huang Transformation
The identification of structural modal parameters is the foundation and precondition of a series of studies and applications in bridge engineering, such as bridge health monitoring, damage detection, dynamical character analysis, etc.However, the conventional modal parameter identification method is basically grounded on the frequency domain or time domain, and the ambient test data with nonlinear and non-station ary characters could not be treated properly by the conventional modal parameter identification method.Hil bert-Huang Transformation (HHT), one kind of modern time-frequency signal processing method with pow erful nonlinear and non-stationary signal processing capacity, is adopted to identify the frequencies, damp ratio and mode shapes for bridge structure.Mirror extension and support vector regression machine are en gaged to suppress the end effects in HHT method, and then the improved modal parameter identification method based on HHT is brought forward to figure out the modal parameters of bridge structure.A software package of this innovational algorithm is work out on MATLAB platform to identify the bridge structural mo dal parameters, and it is verified by the experiment of one certain cable-suspension bridge model.The modal parameters of this model bridge are identified from the measurements under different testing conditions, and the identified modal parameters are compared with the result of FEM and experiment.The validity, practica bility and superiority of the proposal method are verified.The results show that the proposed modal parameter identification algorithm can be used in the real bridge engineering to identify their modal parameters.
Deshan Shan Yanwen Su Qiao Li Zhaoxu Wu
Bridge Engineering Department, Southwest Jiaotong University, Chengdu, China
国际会议
哈尔滨
英文
811-820
2013-07-06（万方平台首次上网日期，不代表论文的发表时间）