In this paper, the method of time series analysis——the maximum entropy spectrum estimation is used in setting up the track crosswise roughness power spectrum density functions (call simple “the track spectra”), th...In this paper, the method of time series analysis——the maximum entropy spectrum estimation is used in setting up the track crosswise roughness power spectrum density functions (call simple “the track spectra”), the track spectra are obtained by process the data measured from the ground. The results obtained in the way are prove to be superior to those obtained with the "FFT" methods.展开更多
Because the existing spectral estimation methods for railway track irregularity analysis are very sensitive to outliers, a robust spectral estimation method is presented to process track irregularity signals. The prop...Because the existing spectral estimation methods for railway track irregularity analysis are very sensitive to outliers, a robust spectral estimation method is presented to process track irregularity signals. The proposed robust method is verified using 100 groups of clean/contaminated data reflecting he vertical profile irregularity taken from Bejing-Guangzhou railway with a sampling frequency of 33 data every ~10 m, and compared with the Auto Regressive (AR) model. The experimental results show that the proposed robust estimation is resistible to noise and insensitive to outliers, and is superior to the AR model in terms of efficiency, stability and reliability.展开更多
文摘In this paper, the method of time series analysis——the maximum entropy spectrum estimation is used in setting up the track crosswise roughness power spectrum density functions (call simple “the track spectra”), the track spectra are obtained by process the data measured from the ground. The results obtained in the way are prove to be superior to those obtained with the "FFT" methods.
文摘Because the existing spectral estimation methods for railway track irregularity analysis are very sensitive to outliers, a robust spectral estimation method is presented to process track irregularity signals. The proposed robust method is verified using 100 groups of clean/contaminated data reflecting he vertical profile irregularity taken from Bejing-Guangzhou railway with a sampling frequency of 33 data every ~10 m, and compared with the Auto Regressive (AR) model. The experimental results show that the proposed robust estimation is resistible to noise and insensitive to outliers, and is superior to the AR model in terms of efficiency, stability and reliability.
