| The wheel flat will cause impact excitation to the train-track dynamic system,which causes severe vibration of the wheel-rail and leads to the diseased occurrence of vehicle and track system and accelerates expansion.Realizing the early identification of wheel flat damage and tracking its development process,automatically obtaining the key geometric parameters of the wheel flat,and predicting the maintenance requirements,which is an important part of the intelligent identification system of rail transit fault.The theoretical basis of the wheel flat recognition is studied in this paper.It is determined that the appropriate dynamic response signal is selected as the carrier of the wheel flat diseased information in the vehicle-track dynamics system.The dynamic response signal processing method which can meet the needs of the intelligent recognition is explored to realize the quantitative identification of the wheel flat diseased information.Based on the theory of vehicle-track coupled dynamics,a dynamic model of the interaction between a certain type of high-speed train and non-ballast track is established.Considering the two working conditions of discrete support mode(fastener)and continuous support mode,the dynamic response of vehicle and track system is calculated.The vibration acceleration of rail is used as the carrier of the wheel flat scar diseased information,and studying the recognition effect of various time-frequency analysis methods on wheel flat scar.Finally,under the condition of continuous rail support,the VMD-envelope spectrum analysis method is used to quantitatively identify the wheel flat scar.The main research contents are as follows:(1)Based on the vehicle-track coupled dynamics theory,the CRH380 B high-speed train-track coupling dynamics model is established,and the calculation model is verified by the experimental data.The dynamic response of the vehicle and track system is calculated by the established dynamic model.(2)Using the discrete support condition of rail,the dynamic response of vehicle-track system under the wheel flat damage is calculated.By comparing the acceleration time domain signal characteristics and energy characteristics of the car body,frame,axle box,wheelset and rail,the energy value is selected as the evaluation index.It is found that there is a linear increasing relationship between the wavelet packet energy value of the rail and the depth of the wheel flat scar.Therefore,it is better to use the rail vibration acceleration response as the carrier of the wheel flat scar diseased information to identify.(3)In order to reduce the influence of fastener spacing on the recognition effect,the influence of rail continuous support is further considered,and the vertical vibration acceleration of rail under two working conditions of rail discrete support and continuous support is compared and analyzed by wavelet packet decomposition method.The results show that the impact of the wheel flat extracted under the continuous support of the rail is more obvious,and the energy characteristics are more prominent.It can effectively weaken the existence of the side peak,reduce the influence of the fastener spacing on the recognition effect,and facilitate the identification of the wheel flat.(4)To improve the ability to quantitatively identify the geometric parameters of the wheel flat,the signal is decomposed by VMD under the continuous support of the rail.The decomposed modal components are analyzed by fast spectral kurtosis to retain the maximum spectral kurtosis component,and the remaining components are reconstructed according to their respective kurtosis weights.Envelope spectrum analysis is performed on the reconstructed signal.The results show that the wheel flat scar damage occurs on the wheel in the rail evaluation area,and the geometric parameters of the wheel flat scar damage can be quantitatively identified.The effect of setting trackside information acquisition system under continuous support of rail is better. |
PDF Full Text Request