Study On Structural Health Diagnosis And Early Warning Of Railway Tunnel During Service Period

As our railway tunnel continues to develop at a high speed,the issue of tunnel structure diseases has become increasingly prominent.This will directly affect the efficiency and quality of railway operation,so it is particularly necessary to carry out the study of railway tunnel disease diagnosis and structural health state early warning during service period.This study is based on the key research project of the Education Department of Liaoning Province,"Research on Diagnosis and Early Warning Technology of Structural Diseases of Railway Operation Tunnel".Combining actual tunnel research subjects with the health assessment standards of railway tunnels during the service period,this study analyzes tunnel disease characteristics and influencing factors,and statistically analyzes field monitoring data.The ultimate goal is to build an integrated platform for comprehensive diagnosis and early warning of tunnel structure health.(1)To begin with,this study analyzes the primary types of railway tunnel diseases during the service period.Taking into account the entire life cycle of the tunnel,the influencing factors and characteristics of these diseases are identified.The main factors contributing to tunnel diseases during the service period are summarized,including complex and variable geological conditions,inadequate maintenance of the tunnel structure,and substandard construction quality.Based on this,find out the characteristics of project research tunnel disease,including water seepage,frost heaving and so on.(2)Furthermore,the diagnostic index system of railway tunnel diseases during service period was established based on the disease characteristics.There were 7 first-level indicators in total,including lining cracks,holes behind lining,leakage,lining material deterioration,lining crushing or peeling,lining deformation and lining corrosion.Each first-level indicator contained 1～3 second-level indicators,and there were 13 second-level indicators.Based on the field special monitoring data of tunnel diseases of a tunnel in Daqin Railway,13 subdivided diseases of railway tunnels during the service period proposed in this study were monitored and analyzed according to the characteristic parameters of different diseases,such as size,area,location,type and damage,etc.,and the disease monitoring system of railway tunnels in the Internet of Things was established.(3)Next,a comprehensive weight vector is defined.Taking into account the impact of water leakage and freezing on tunnel structures during winter,this study combines an improved group G2 method to calculate the weight of tunnel diseases in both freezing and non-freezing seasons.Using this approach,a diagnosis and early warning method for railway tunnel structure health based on its service period is proposed.The accuracy of predictions is verified using RMSE value and MAPE value.To further reduce errors and optimize the early warning model,a combined model based on SARIMA-BP neural network is proposed.This model achieves higher long-term prediction accuracy than the sub-model,confirming the accuracy and stability of the combined model.(4)Lastly,this study designs a health diagnosis and early warning system for railway tunnels during the service period.The system uses My SQL software to build a railway tunnel disease diagnosis database.Then,Visual Studio development tool is used to create an integrated platform that includes disease statistics,monitoring,diagnosis,and warning systems.This platform allows for visual management of railway tunnel health diagnosis and warning,ensuring efficient and stable maintenance of railway tunnel structures during their service period.