Research On Bridge Damage Identification Method Based On Wavelet Analysis

In view of probable partial damage of bridge structures in operation due to outside complex factors, the sudden variation information is detected using self-adaptiveness of time-frequency window in wavelet analysis to identify damage in structures. For the simple beam, continuous beam and cantilever beam, the damage and its identification are simulated numerically, the damage alarming, positioning and quantifying are studied for wavelets-transform-based damage identification. The main subjects are summarized as follows:(1) The damage alarming in structure is studied for wavelet packet nodal energy. First, the thesis analyzes the links between nodal energy information and structural damage, and then proposes the relative value of wavelet packet nodal energy as indicators of damage alarming. The continuous beam is simulated numerically with the help of ANSYS, and the change of modal frequencies between damaged and undamaged structure is obtained; Mean while, the acceleration of ANSYS transient analysis is studied in wavelet packet analysis in MATLAB, and the change of the relative value of energy is obtained. Comparing the two, the change of energy is more sensitive to the damage, and it needs no more information of structure parameters, so the online damage alarming can be better achieved.(2) The thesis studies the methods of damage positioning based on curvature mode and wavelet transform. First, the methods of damage identification based on curvature mode and wavelet transform are analyzed respectively. The cantilever beam is simulated numerically, and the curve of curvature mode is obtained, then we can find the damage point easily; the modal displacement of the simple beam is studied by wavelet transform, and the coefficient curve is obtained, then we can find the different point of damage signal clearly. The singularity of the signal can be found in curvature easily and identificated in wavelet analysis clearly, so we can get better effect for the research of damage detecting by combining the two.(3) The thesis studies the relation between the sizes of Lipschitz exponent and the extent of the damage. The simple beam is simulated numerically, and its modal displacements of different damage conditions are analyzed by Mexican wavelet transform, and we can get the maximum coefficient modulus of 1-6 scale, from the logarithm of which Lipschitz exponent curve is obtained, and we can find that different extent of the damage match with different Lipschitz exponent. The thesis also studies the change of Lipschitz exponent in different place, different mode and the condition with other damage.(4) The thesis studies the application of wavelet analysis in the damage identification of the bridge and gets a good result from the analysis, which is of great significance to the health monitoring of the bridge.