Modelling Research Of Railway Track Geometry Degradation Law Based On Hierarchical Bayesian Models

With the continuous development of China's railway industry,the continuous increase of railway operating mileage,the continuous improvement of train speed,put forward higher requirements for railway infrastructure management.The maintenance and repair methods of railway infrastructure have gradually changed from cycle repair to state repair and preventive repair management modes,and put forward higher requirements for accurately grasping the current state of the infrastructure and the future state.The track geometry state is very important to the operation safety of the train.This paper will use the Hierarchical Bayesian model to model the geometric degradation process of the railway track.The Hierarchical Bayesian Model is a flexible statistical model that determines the spatial correlation between the degradation rate of different quality parameters and the initial standard deviation between successive track segments.In this paper,a Hierarchical Bayesian model is constructed for the standard deviation of longitudinal level defects and the standard deviation of horizontal alignment defects of railway track geometry.The main research work is as follows:Firstly,aiming at the limitations of the current railway track geometry degradation law model,the modelling idea of applying the Hierarchical Bayesian model to the track geometry degradation law is proposed.The advantages of the Hierarchical Bayesian model are analyzed and based on the Hierarchical Bayesian model.The applicability of the railway track geometry degradation law model was studied.Then,a model of railway track geometry degradation law based on Hierarchical Bayesian model is constructed.Through the correlation analysis,the model hypothesis is discussed.The influence of the spatial interaction factors of the track segment on the track geometry degradation is considered.The conditional probability structure is introduced to carry out the spatial correlation analysis.The parameters in the track geometric degradation law model are defined.The prior distribution of the joints is definited,and the joint posterior distribution is derived.The geometrical deterioration law model of railway track based on the Hierarchical Bayesian model is established.Four kinds of the Hierarchical Bayesian model of track geometric deterioration law obeying different prior distribution functions are constructed.The regular level Bayesian model was compared.Finally,a case study was conducted with Lanxin high-speed railway as an example.Based on the track detection data of Lanxin high-speed railway,the Markov chain Monte Carlo method is used to simulate the Hierarchical Bayesian model of the railway track geometry degradation law,and the Markov chain Monte Carlo method process of the model is coded.The simulation process and simulation results are analyzed in detail.The sensitivity of the prior distribution parameters in the model is analyzed,and the Hierarchical Bayesian model of the optimal railway track geometry degradation law under different prior distribution conditions is determined.