Crack Damage Identification And Localization Of Rail Vehicle Wheel Web Based On Guided Waves

With the continuous advancement of the transportation industry,China has become a strong rail transit country.The rise of rail transit will bring great social and economic benefits.But at the same time,it should pay more attention to the safety of vehicles and the safety during operation.Among them,wheel safety is the basis of the safety of rail vehicles.Once the wheels are in trouble,the consequences are unimaginable.The severity of major accidents caused by wheel failures has spurred the research into wheel damage detection methods.By investigating the crack failure of the web,cracks in the wheel web may cause the wheel to crack,which causes the train to derail.The consequences are extremely serious.And there are few studies on its detection methods.Therefore,this paper proposes a guided wave-based damage detection method for wheel web cracks,which detects and identifies the damage.The main work of this paper is as follows:(i)According to the study of guided waves and the condition of wheel webs,the Lamb wave suitable for the plate structure is selected for crack damage detection.The group velocity and phase velocity of the Lamb were calculated and the dispersion curves of different materials were solved.The excitation signal of Lamb wave is optimized and selected.Finally the 5-peak narrow-band signal of the sine wave modulated by Hanning window is selected as the excitation signal.The Lamb wave signal collected by the experiment is analyzed by short-time Fourier transform.The research on the curved steel plate shows that the existence of the curved surface does not cause the reflection of the Lamb wave,and the wave will continue to propagate forward in the plate.Waveform signals corresponding to 19 different sizes of damage were collected.The energy amplitude of the damage reflected wave was the damage indicator,and the damage size was identified by the support vector machine.(ii)For the complexity for wheel webs,this paper proposes a damage identification technology based on minimum difference,constructs the difference index function.Combining the genetic algorithm to obtain the damage related information by reconstructing the damage reflected wave.Thereby using geometric positioning method to achieve damage localization.The reconstruction algorithm was verified by single damage reflection experiment and non-single damage reflection experiment using aluminum plate and steel plate structural materials.The results show that the difference between the reconstructed signal and the measured reflection signal is small and the effect is good.The damage localization experiment was carried out on the single damage reflection source,and the results showed that less error compared to the actual situation.(iii)Verified by the previous algorithm,this paper applies it to the crack damage detection of rail vehicle wheel webs(including web holes).By experimenting to compare the magnitude of the direct wave amplitude of different excitation frequencies,260kHz was chosen as the experimental frequency.The experimental analysis of the 2cm crack damage in the wheel web is carried out.The short-time Fourier processing is performed on the different path signals.The result shows that the calculated propagation distance is larger error compared to the actual situation.The damage signal reconstruction method is used to locate the damage.The results show that although there is a certain error,it can basically realize the localization identification of crack damage.Finally,the design scheme of the wheel web structure health monitoring system is proposed.