Research Of Short Wave Track Irregularity Recognition Method Based On Axlebox Spectrum

With the increase of running speed, the influence of the short wave track irregularities on dynamic action between wheel and rail become a very important problem on high speed rail lines. The current methods of track measurement are the traditional measurement, the track inspection car and light duty motor trolley. They are not very suitable for frequent measurement on high speed and high density lines. So, the research of other supplementary means for inspection of short wave track irregularity is needed.In this paper, the mechanical fault diagnosis theory is applied to the field of short wave track irregularity recognition for the first time. The short wave track irregularity recognition method based on axlebox spectrum is proposed. This method is from the angle of the frequency for the inspection of short wave track irregularity, which is different from the angle of the displacement adopted by various countries now.The purpose of the study is to summarize signal characteristics and the state of corresponding short wave track irregularity through the processing and analysis of axlebox acceleration. That is to recognize the type and state of short wave track irregularity utilizing axlebox acceleration.Harmonic irregularity model, parameter lumped simple wheel-track model and wheel-track model after introducing track irregularity are established based on dynamics in the paper. And the equal relationship between the frequency of harmonic irregularity and the frequency of axlebox acceleration is deduced. Simultaneously, the preprocessing, filter processing of experimental data and comparison recognition are done. What have been proved in theoretical calculations and experimental data processing are that short wave track irregularity will lead to the changes of axlebox acceleration, the recognition of short wave track irregularity such as corrugation, weld depression, loose sleeper etc. utilizing the short wave track irregularity recognition method based on axlebox spectrum is feasible.