Research On Analysis Method Of Bridge Damage Signal By Different Wavelet Base Function

In recent years,the non-destructive testing of bridge damage is a hot issue in bridge research.And modal parameter identification through vibration signal analysis is the basis.Therefore,how to identify modal parameters accurately attract the attention of scholars.The traditional Fourier transform is to get the continuous spectrum and then analyze the signal in the frequency domain.It have high precision.But this method have unobvious effect about de-noising,especially for short-time and high-intensity noise.Fourier transform average the energy of the signal to the entire spectrum,then affect the identification of the parameters.The modal parameter identification based on wavelet transform is a time-frequency domain identification method.Because of the time-positioning capability and multi-resolution characteristic,the wavelet analysis has been widely use.However,due to the selection of wavelet bases,the further development of wavelet transform is restricted.The main contents of this paper include:Firstly,de-noise the signal and analyze the de-noising effect by using different wavelet bases.To determine wavelet base which have the best effect of the vibration signal de-noising.Identify frequencies from the de-noising signal by using Fourier transform.Secondly,introduce the principle about identifying each natural frequency of the bridge with the method of wavelet time-frequency diagram.Identifying each natural frequency by wavelet time-frequency diagram with different complex wavelet base function.Compare the solution of finite element,Fourier spectrum and wavelet time-frequency diagram.In short,this paper is to use the de-noising ability and time-frequency analysis capability of wavelet analysis,dislodge the effect of the noise signal and make the result of identification more accurate and credible.