THE TIME-FREQUENCY ANALYSIS OF THE TRAIN AXLE BOX ACCELERATION SIGNALS USING EMPIRICAL MODE DECOMPOSITION

Xingjie Chen, Xiaodong Chai, Xining Cao

College of Urban Railway Transportation, Shanghai University of Engineering Science, Shanghai, China, 201620

Rail defects usually result in lots of problems such as affecting the comfort of passengers, increasing the wheel-rail forces, exacerbating the train axle boxes vibration and track wear, even threatening the safe operation of trains. In this paper, the characteristic frequency distribution of the changing axle box acceleration caused by defects is analysed by empirical mode decomposition and Hilbert-Huang Transform is used to analyse the time-frequency changes of axle box acceleration. As a result, rail defects can be effectively positioned and the short wave irregularities within a certain degree can be detected. The research provides timely protection for the maintenance of the track.