Diagnosis Method Of Short-wave Irregularity Of Heavy-haul Railway Track Based On Vehicle Dynamic Response And Applied Research

The track health status of heavy-haul railway is the key factor for the safe and smooth operation of the rail-wheel system.Poor welded joints,rail corrugation and other shortwave irregularities of the track will produce abnormal wheel-rail force and cause highfrequency vibration.In serious cases,it will damage the parts of the track-vehicle system,and even affect the safety of the system.In order to eliminate and mitigate the adverse effects caused by the track short-wave irregularities,each public works section has invested significant manpower and financial resources for regular on-line maintenance.However,for heavy-haul lines which spans hundreds of kilometers,it is impossible to conduct a comprehensive inspection due to the prohibitive costs.Therefore,an effective short-wave track irregularity diagnosis method,one that could monitor the track state of the whole road while reducing expenses,is especially important for maintenance.This purpose can be achieved by using the methods of vehicle dynamic response signal and signal processing.The research of this thesis is about the short-wave irregularity diagnosis method of heavy haul railway based on vehicle dynamic response,including: finite element simulation of heavy haul railway,time frequency analysis method of signals with various frequency components,diagnosis method of welded joint and corrugation,etc.To study the change of wheel-rail vertical force when the train passes through corrugation sections with different wavelengths,the high frequency finite element model was established by commercial finite element analysis software ABAQUS,and the wheel-rail vertical force response of the heavy-haul railway under different working conditions was simulated and calculated.The finite element simulation model was established and the correctness of the simulation results was verified by the existing measured data.The influence characteristics of wheel-rail dynamics under different vehicle axle loads,running speeds,fastener stiffness and other parameters were calculated by using the control variable method.By comparing wheel-rail forces at different wavelengths,the results show that under the same working conditions,the short wavelength can cause a larger range of wheel-rail force fluctuation amplitude.When the vehicle passes through a corrugation section mixed with poor welded joints,the box acceleration signal contains two components: steady component and impact component.In order to solve the problem of limited window length of short-time Fourier transform for synchronous compression of this kind of signal,and obtain the time frequency distribution with high resolution,an adaptive synchro-squeezed short time Fourier transform by wavelet packet decomposition(WPD-ASSTFT)was proposed.The method is applied to the simulated signal with steady component and impact component and the measured vehicle dynamic response signal,and compared with the time frequency analysis methods such as SSTFT and VNCMD,it is found that WPD-ASSTTFT has higher time frequency resolution.There are four types of heavy-haul railway welded joint irregularity: convex type,concave type,M type and staggered type.Through time trequency analysis of axle box acceleration in different joint types,it is found that the impact caused by welded joint irregularity has a short duration and a wide frequency band,and the frequency distribution of impact components caused by the four types is about 10?900 Hz.On this basis,in order to diagnose the irregularity of welded joints of heavy-haul railways,the WPD-ASSTTFT marginal index method is proposed.This method is applied to the diagnosis of welded joint irregularity of heavy-haul railway,which can stably diagnose the welded joint irregularity of heavy-haul railway.Moreover,this method can not only identify the irregularity of a single welded joint,but also the irregularity of the joint in the wave grinding section.Rail corrugation of heavy-haul railway will aggravate wheel-rail vibration,lead to damage of parts of rail-wheel system,and endanger operation safety.In practical maintenance,corrugation section should be found and polished in time.If the train wheels are out of round,the detection data will contain its corresponding periodic components,which will affect the diagnosis of rail wave wear.In data preprocessing,a joint analysis method of envelope spectrum and WPD-ASSTFT is proposed to diagnose the wheel out of round.If there is wheel out of round,the data should be filtered before preprocessing.In the diagnosis of single rail corrugation,both the characteristic of amplitude increase,and the characteristic of periodicity should be considered.If the marginal spectrum is calculated only in the time domain,the periodicity is lost.The marginal spectrum is improved and the time-frequency marginal spectrum is calculated.In the early stage,a WPD-ASSTFT time-frequency marginal spectrum method based on axle box acceleration was proposed to diagnose rail corrugation of heavy-haul railway according to the field data collected and time frequency characteristics analysis.According to the results of diagnosis and recheck,this method can effectively diagnose the rail corrugation of single component in heavy-haul railway.Combining timefrequency analysis and machine learning techniques,a method of rail corrugation diagnosis based on WPD-ASTFT and SVM was proposed and the prediction accuracy rate reaches 93.33%.Through monitoring rail corrugation of heavy-haul railway for many years,it is found that the axle box acceleration signals corresponding to some corrugation have multiple frequency components,which is called multicomponent rail corrugation.According to the characteristics of multicomponent corrugation,a corrugation index-multiple energy factor method is proposed for the diagnosis of multicomponent rail corrugation of heavy-haul railway.This method is applied to the diagnosis of rail corrugation in heavyhaul railway.According to the analysis and on-site review for four consecutive months,it is found that the method of corrugation index-multiple energy factors can effectively identify single and multicomponent rail corrugation in heavy-haul railway.Based on the diagnosis results and field review for four consecutive months,it is found that this method can effectively identify single and multicomponent rail corrugations of heavyhaul railways.In order to obtain the depth information of the corrugation section,an estimation method of rail corrugation amplitude of heavy-haul railway is proposed.The experimental results show that the estimated results are consistent with the measured geometric data.This method is helpful to estimate the severity of corrugation,arrange on-line grinding sequence,and inhibit further deterioration of corrugation.