Structural damage identification based on moving load and time-frequency entropy analysis

Moving load could enhance damage structural nonlinear character. And time-frequency analysis described character of non-stationary signal more exactly than traditional fourier spectrum. Affine class time-frequency analysis was good at wide-band signal analysis and cross terms suppression. The result of time-frequency analysis was two dimensional , and time-domain entropy or frequency-domain entropy can extract information and characters from it. Using these characters the structural damage pattern could be recognized by neural network. The validity of this method was reviewed by a beam under moving load. When structure has damage, time-frequency entropy of structural vibration can figure the nonlinear grade of signal availably, and it is not sensitive to noise. It can be used as eigenvectors of pattern identification based on neural network.