Abstract

Drifting snow, the flow of dispersed snow particles near ground level under the action of wind, is a major form of snow damage. When drifting snow occurs on railways, highways, and other transportation lines, it seriously affects their operational safety and results in drifting snow disasters. Drifting snow disasters frequently occur in the high latitudes of northwest China. At present, most scholars are committed to studying the prevention and control measures of drifting snow, but the prerequisite for prevention is to effectively evaluate the susceptibility of drifting snow along railways and highways to identify areas with a high risk of occurrence. Taking the Xinjiang Afukuzhun Railway as an example, this study uses a geographic information system (GIS) combined with on-site monitoring and surveys to establish a drifting snow susceptibility evaluation index system. The drifting snow susceptibility index (DSSI) is calculated through the weight of an evidence (WOE) model, and a genetic algorithm backpropagation (GA-BP) algorithm is used to obtain optimised evaluation index weights to improve the accuracy of model evaluation. The results show that the accuracies of the WOE model, WOE backpropagation (WOE-BP) model, and weight of evidence genetic algorithm backpropagation (WOE-GA-BP) model are 0.747, 0.748, and 0.785, respectively, indicating that the method can be effectively applied to evaluate drifting snow susceptibility.

Highlights

  • Drifting snow is an atypical gas–solid two-phase flow in which dispersed snow particles move near the ground under the action of wind; referred to as snowstorm flow [1,2], it is a major form of snow damage

  • The method uses statistical analysis of the contribution of the evidence-level factors to the research goal to predict whether the event will occur; in this manner, the influence of subjective factors can effectively be avoided [42]. This method was first applied in the field of medicine and introduced by geologists Bonham-Carter et al [43] and Ahterberg et al [44] into the field of mineral research. It has been widely used in research on landslides, debris flows and other geological hazards [45]; it has seldom been applied to drifting snow disaster evaluation

  • Snow area equal the area the of the drifting snow in the secondary of the indicator factor/the total areadrifting of the drifting snow in the secondary state ofstate the indicator factor/the total area of the drifting in the wholethe region; the graded area of each state secondary snow in snow the whole region; graded area ratio is theratio area is of the eacharea secondary of the state of the index factor compared to the total area of the index factor

Read more

Summary

Introduction

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. From February 29 to 13, 2016, severe drifting snow disasters frequently occurred in Maytas, Xinjiang Region, causing traffic jams and trapping hundreds of cars and passengers in disaster areas. The influencing factors of linear engineering drifting snow are notably different from natural snow disasters, but recent studies do not establish a risk evaluation index system for railway drifting snow disasters [32,33,34]. This study uses the Xinjiang Afuzhun Railway as the research object, combined with a site survey and monitoring data based on a GIS platform, establishes a railway drifting snow susceptibility evaluation system, and uses the weight of evidence (WOE) model to calculate the susceptibility index. The result shows that this method can provide a reference for other similar railway projects

Evidence Weight Method
Coupling Model
Overview of the Study
Monitoring
Evaluation Index System
Index Factor Classification
Elevation
Relief amplitude
Surfacestatistical roughness analysis statistical analysis
Regional Snow Field Conditions
14. Included angle statistical
GA-BP Algorithm Optimisation
Susceptibility
Accuracy Evaluation
19. ROC curve and comparison
Conclusions
Taking

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.