Influence Of Rail Wear Condition On Environmental Vibration Source Intensity Induced By Metro Trains

The train-induced environmental vibration has become a big problem with the rapid development of urban transit.Track irregularity is the source of wheel-rail vibration.Short-wave irregularity like rail surface roughness causes high-frequency vibration.High-frequency vibration is easy to break fasteners,increase the vibration of tunnel wall and make rolling noise exceed the limits.Besides,track irregularity is a vital input excitation model of the vehicle-track model and it is very important to accurately simulate the short-wave irregularity for predicting the influence of environmental vibration.To quantitatively calculate and analyze the effect of rail roughness on train-induced vibration,it is necessary to simulate the real rail surface roughness excitation under different wear states in the vehicle-track coupling model.However,there is a lack of research on metro rail surface roughness spectrums and description methods of different rail wear situations.Because of this,this thesis collected 439 samples of real Beijing metro rail surface roughness,which cover various working conditions of the Beijing subway,with the financial support of National Natural Science Foundation of China(No.51978043,No.51778049).The classic periodogram method was used to obtain the Power spectral density(PSD)function of all data.Then the condition of rail wear was classified into 5 levels,Q1-Q5(Q1 corresponds to the best condition of rail and Q5 corresponds to the worst condition of rail),according to the characteristic of the PSD curves and the functions of the Beijing metro 5-level spectrums(Beijing spectrum)were obtained by fitting.Among the 2? rad/m-200? rad/m wavenumber range,FRA1 spectrum(American class 1 track irregularity spectrum),Wanglan spectrum(roughness spectrum proposed by Wanglan)and Sato spectrum(roughness spectrum proposed by Sato)fully lie between Beijing Q1 and Beijing Q5 spectrum.The curve of Wanglan spectrum is higher than the curve of Q4 and lower than the curve of Q5,and the band of the Sato spectrum falls between the Q1 and Q3 spectrum curves.Moreover,the curve of Wanglan spectrum and Sato spectrum drop in the same velocity with the Beijing 5-level spectrums,but the curve of FRA1 and FRA6(American class 6 track irregularity spectrum)decrease far more quickly.Then the spatial signals of Beijing spectrum were obtained by triangle series method.Compared with the in-site test roughness level in 1/3 octave wavelength,the all fitted signal roughness levels are more ‘smooth' and cannot represent the characteristic wavelength of different tracks,but the Q1 and Q5 can be used as the theoretic boundary of the in-site test roughness level because the curves of Q1 and Q5 can completely cover all in-site test roughness levels.The rail acceleration levels of Beijing spectrum were obtained after inputting the spectrums into a vehicle-track model directly.It is found the in-site test rail acceleration level is completely covered by the acceleration levels of Beijing spectrum.Between 12.5Hz and 63 Hz,the in-site test acceleration level is close to the acceleration levels of Q2 and Q3 spectrums.Between 63 Hz and 200 Hz,the in-site test acceleration level drops slowly to the acceleration of Q1 spectrum.Between 200 Hz and 500 Hz,the in-site test acceleration level grows gradually to the acceleration level of Q5 spectrum.It is proved that the Beijing spectrum proposed in this paper can cover all the energy distribution of actual rail wear.Finally,the reaction force of the fastener calculated by the vehicle-rail coupling model is brought into the track-tunnel-soil finite element model to calculate the tunnel wall acceleration.In the 1/3 octave frequency domain,from 1 Hz to 80 Hz,the energy of the tunnel wall acceleration spectrum gradually increases,and there is a very distinct gradient distribution,which is consistent with the gradient of Beijing spectrum.The biggest differences of maximum Z vibration level of the tunnel wall acceleration between the Q2 level and Q4 level are 8d B,8d B and 4d B of the slab track with directly fixed fastening,slab track with Cologne Egg fastening and steel-spring floating slab track respectively,which is same as in-site test results in research.The research proves that: 1 it is reasonable to divide the Beijing roughness spectrum into different levels.2 The rail wear condition has a great influence on the vibration response of the tunnel wall.