Research On The Evolvement Model Of Long Railway Service Status And The Estimation Algorithm Of Key Parameters

High-speed railways are complex large systems,and the service performance of vehicles and railway infrastructures changes continually in the process of long-term operations.Due to the substantial increase of vehicle speed,the wheel/rail interaction force increases which enhance the impact effect on the track.With the development of large-scale railway network,the lines route through multiple climatic zones,which makes the evolvement of service performance more complicated.The degradation of railway infrastructure service performance leads to more abruptness and severity of potential disease,which influences significantly operation safty of the railway.Therefore,the formation and distribution regularities are the critical question for accurately perceiving the variation trend of component status parameters in the critical areas,and also establishing and perfecting the scientific security system of railway health service status,which contribute to ensuring the safety and steady operation of the high-speed railway.The rail stress is the key parameter for reflecting the railway safety,which is taken as the starting point in thesis.Concentrating on the evolvement process of long railway service performance and considering the slow change effects of temperature factor and the transient effects of vehicle impact,the CWR(continuous welded rail)track dynamic and static model is established and the research on the intelligent estimation algorithm of relevant parameters has been carried out.The evolvement process of rail stress in complicated section has been simulated.The influence of rail stress on dynamic responses of vehicle-track has been analyzed,which provides theoretical basis for the detection and evaluation of the rail stress.Firstly a progressive homogenization method for calculating the average performance parameters of large structures was proposed and the multi-scale static model of long railway under the change effect of circadian and seasonal temperature was established.According to the structure characteristics of railway in different sections,the evolvement rules of railway status parameters in each section were analyzed by adopting different simulation scales.A refined model that retains key details was established for the focus sections of railway.For the extended sections of railway,according to the periodic characteristics of the structure,the average performance of the typical structure was extracted and the macro scale model was established.According to the principle of the consistent boundary behavior,the local models of different scales were combined into multi-scale overall railway model and the process of rail stress redistribution caused by local structure change was simulated.The improved long rail units were applied in the model,which considered the nonlinear influence of unit section’s rotation on the vertical and horizontal displacement field and improved the deformation calculation accuracy of long rail units under multiple loads.The nonlinear linkage unit and the contact element which can reflect the skid condition were applied to simulate the restraint relationship of these layers in the track,which achieved the calculation of the cumulative deformation of track under the cyclic loads.By comparing the simulation results and the field measurement data from the Beijing-Hangzhou canal railway bridge and the Xinbianhe bridge of Beijing-Shanghai high-speed railway,the effectiveness and accuracy of this proposed multiscale model were verified.Secondly,the vehicle-track coupling dyancmic model under rail longitudinal stress was established in this paper and the influence of the rail longitudinal stress on the dynamic responses of vehicle-track was analyzed.The vibration differential equation of rail was proposed with the rail longitudinal stress considered on the basis of the complete ballastless track model.According to the segmental analysis and combination solution of the track model,an exact analytical form of track inherit vibration characteriscs under the rail stress was deduced,and the regularities of track inherit vibration characteristics under different rail longitudinal stress were analyzed.Based on the Zhai-sun vehicle-track dynamic model,the dynamic responses of the vehicle and track along with the changes of rail longitudinal stress were calculated according to introducing the modified track mode and frequency.At last,on the basis of grasping the distribution regularities of rail stress of long railway,according to the significant correlation between wheel/rail dynamic responses and rail longitudinal stress,the research on intelligent estimation of rail stress based on the wheel/rail dynamic response were carried out for the state sensitive area of railway.The Support Vector Machine classified model was proposed for the fast stress recognition.According to measuring the dynamic responses of wheel/rail,the fast classification and real-time location of rail stress using sensors mounted on in-service vehicles.Meanwile,the stress parameters estimation model based on the Genetic Algorithm was proposed for resisting disturbance of track irregularity.By using the continuity of the stress distribution,the computational efficiency and accuracy of the estimation algorithm were improved.The recognition accuracy and applicable conditions of the two methods were analyzed by simulation test.The research contents have been conducive to establish the comprehensive monitoring system of China’s high-speed railway equipment service status,providing the theoretical and methodological support for guiding maintenance and repair of railway scientifically.