Research On The Method Of Intelligent Response Surface Reliability Analysis Of Slope Stability

Landslide is an important geological disaster, and it is a major threat on human life and property. There are a lot of uncertainty factors in the design, construction, process of using, such as load, resistance and other structural design variables, so, it is necessary to consider all the uncertainties, in order to ensure the safety and application of the slope. Slope reliability analysis is based on the known statistical parameters of random variables and probability distribution models, and functional status of a given slope, it is estimated slope in the prescribed conditions and be completed within the stipulated time the probability of intended function. This article uses Intelligent Response Surface Method in the analysis of slope stability, builds Intelligent Response Surface model, combines with Monte Carlo simulation, analys slope stability reliability. For the structure and function of high non-linear implicit can be a larger reduction of the frequency of finite element analysis,this method greatly improved access to meet the accuracy requirements of the structural failure probability estimates and its variance, and has important practical value.At present, there are three kinds of normal reliability analysis methods, such as, Mean First Order Reliability Method (MFORM), First Order Reliability Method (FORM) and Response Surface Method (RSM), MFORM's precision is lower, while FORM and Response Surface Method has a great of calculation. So, Intelligent Response Surface Method is proposed. The method adhere to the integration of RBF and SVM with Monte Carlo simulation method, then analysis the reliability of the slope. The calculations indicates this method has a higher precision just as FORM method, while the efficiency is much more higher than FORM method and RSM method through a example of landslide in Da Shu-shan of Wang-jiang west Road. It indicates that Intelligent Response Surface Method is more superior in reliability analysis in slope engineering.The research is supported by the National Natural Science Foundation of China (No.40972194).