Study On Slope Safety Evaluation Method Of High Speed Railway In Mountainous Area

Slope instability and dangerous rockfall sometimes are occurred in the process of mountainous high speed railway operation, which are great threats to the railway. Therefore, a scientific and reasonable method for slope safety evaluation along the railway is very important.The subject comes from the former Ministry of Railways science and technology development project-the research of monitoring and alarming system for high steep slope operation safety (2010G018-E-4). This paper establishes a method to evaluate the slope safety of high speed railway in mountainous area by studying the main affecting factors of slope instability and dangerous rockfall and applying rough set and support vector machine theory to provide scientific basis for the slope safety evaluation and ensure the railway’s operation safety.This paper firstly summarizes the following characters of high speed railway slope in mountainous area:distribution and scale, geographic and geomorphic, lithology, geological structure, hydrogeology, unfavorable geology and supporting measures, etc. and analyses the influence mechanism of height, slope, classification of rock mass, shearing strength parameters, dip relationship between slope and structural plane, weathering effect, action of water, slope steps, supporting conditions, rock types, development of fissure, vegetation effect and occurred geological disasters for the slope safety. The safety evaluation indexes are put forward respectively in rock slope, soil slope and rockfall fields.Secondly, according to the safety evaluation indexes, data of slope sites are collected. The decision tables based on rough set theory are formed after the discretization of data. By applying the concept of reduction and significance in rough set theory and the integration of practice, the main affecting factors for slope safety and their weight coefficients are acquired, which can avoid the influence of redundancy factors for the evaluation results and reflect the influence degree of the main affecting factors quantitatively.On this basis, slope safety evaluation models of high speed railway in mountainous area are established respectively in rock slope, soil slope and rockfall fields in which the weight coefficients of the main affecting factors are considered. The support vector machine theory is used to train these models. Examples prove that the method can reflect the complex and nonlinear mapping relationship between slope safety and the main affecting factors and its accuracy can be ensured. By this method, the slope safety classification can be decided from the related main affecting factors and the potential geological hazard can be warned in time. Therefore, the evaluation method of slop safety based on rough set and support vector machine theory is verified to be practical.