Study On Identification Of Non-stationary Characteristics Of Ground Motion And Its Effect On Random Response Of Bridges

Dynamic response of bridge structures under random earthquakes is always a hot topic.The seismic ground motion is a typical non-stationary process,which shows significant changes in both the time domain and the frequency domain.Usually,in the seismic analysis of the existing bridge,the former is considered.However,the influence of the latter is seldom studied.Therefore,it is of great theoretical and practical significance to identify the non-stationary characteristics of seismic ground motion by using the existing seismic records and the wavelet transform,and then to investigate the influence on the dynamic response of long-span bridges.The main contents and conclusions of this paper are as follows:(1)Theories based on the continuous wavelet transform and the Stockwell transform are summarized in order to extract non-stationary characteristics of excitation.A program is compiled and verified for identifying the evolutionary power spectrum of earthquake records,in which the Morlet wavelet transform,the generalized harmonic wavelet transform and the S-transform are employed.Three identifying methods have good effects.The constant coefficient matrix with 15 orders is proposed to get the result in the Morlet wavelet method.The generalized harmonic wavelet method is recommended in the following analysis.(2)The Priestley spectral model considering the fully non-stationary characteristics,the spatially variation model and the pseudo excitation method(PEM)are introduced.(3)A finite element model of a self-anchored suspension bridge is established and the modal analysis is carried out The evolutionary power spectrums of three real ground motion time histories the amplitudes of which are modulated are identified using the generalized harmonic wavelet transform The PEM is employed to calculate the evolutionary power spectral densities and variances of dynamic response,and the results show that the difference in the non-stationary characteristics of excitation results in the variation of the peak value of bridge response by nearly 20%.