Rockfall disasters occur frequently in mountainous areas of western China, and the rockfall disasters along a railway line will seriously affect the safety and normal operation of railways, causing great economic and property losses. Existing rockfall monitoring and early warning methods still have shortcomings, such as accurate warning of single-point disasters and vulnerability to the natural environment. In this study, a rockfall test of a flexible safety protection net along the slope of a railway and a rockfall test of the railway track were carried out, and the vibration signals of the falling rock hitting the different sites of the protective net and hitting different positions of the rails were obtained. Using the signal analysis methods such as Fast Fourier Transformation and Short-Time Fourier Transform, the basic characteristics of the rockfall vibration signal and the vibration signal when the train passes and the propagation law of the rockfall vibration signal are obtained. Finally, a set of monitoring and early warning systems for rockfall disasters along the railway based on the analysis of vibration signal characteristics is established. The monitoring and early warning method has the advantages of all-weather, high-time, semi-automatic and high efficiency performance.

No abstract

Western China is rich in oil and gas resources, and many oil and gas pipelines are under construction or have been completed. However, many water-related natural hazards, such as landslides, collapses, rockfalls, and debris flows, have developed in the areas passed through by oil and gas pipelines and seriously threaten the operational safety of these pipelines. Therefore, it is urgent to carry out large-scale identification and assessment of pipeline geological hazards. At present, conventional on-site investigation, evaluation, monitoring, and early warning methods are difficult to apply for rapid identification and evaluation of pipeline geological hazards across large-scale areas. Based on this, this study takes the pipeline of Sinopec Marketing South China Branch in Yunnan Province as the research area. In this research, unmanned aerial vehicle (UAV) and photogrammetry technology were used to quickly and accurately obtain multi-phase images of an oil pipeline passing through the study area, and the images were post-processed to obtain multi-phase high-resolution, high-precision digital orthophoto maps and digital terrain models (DTMs) to identify landform changes and deformation. The focus of this research is to propose a set of technical methods for UAV point cloud filtering. The DTMs obtained based on this method can effectively identify unstable areas of oil pipelines. In addition, we have carried out numerical simulations under different motion scenarios in unstable regions, providing scientific support for future geological hazard prevention and mitigation and engineering practices in oil pipeline areas.