Font Size: a A A

Slope Reliability Analysis With Monte Carlo Method

Posted on:2007-03-15Degree:MasterType:Thesis
Country:ChinaCandidate:P P HeFull Text:PDF
GTID:2120360182482554Subject:Geological Engineering
Abstract/Summary:PDF Full Text Request
Because the influence factors of slope stability are uncertain obviously, traditional safety factor can not show the failure probability of landslides any more. Reliability theory is induced in slope stability analysis, which is called slope reliability analysis. The purpose of slope reliability analysis is calculating the slope failure probability. Many different methods for calculating the failure probability and reliability index are put forward gradually these years, such as Monte Carlo method, first-order second-moment method, Rosenblueth method, etc. In these methods, Monte Carlo method which has brief principle and accurate results, is used abroad to analyze slope reliability.The sensitivity of calculating parameter on slope failure is analyzed, with the angle of friction and cohesion force of slope failure surface necessarily being considered as factor of uncertainty. Basing upon mechanical model of imbalance thrust force method and considering two vital induced factor: water and seism, reliability analysis model on slope is set up by inducing uncertainty factors and using reliability analysis principle of Monte Carlo method. The methods of error estimate and simulation times are brought forward, and precision qualification is given out.If the stability or reliability of slope is being analyzed, the location of failure surface must be confirmed initially, which can be done by two ways. One is drilling prospecting while the failure surface exists exactly;and the other is optimization method or numerical simulation for seeking failure surface while it does not exist or drilling can not be achievable. Genetic algorithm is induced to slope stability analysis for seeking the most dangerous arc failure surface in soil slope, choosing the coordinates of circle center and radius of arc as searching variable. The results show that the precision of solution and searching efficiency have been improved.Reliability analysis of the Xiakou slope in Yucheng district of Ya'an City in Sichuan Province is accomplished by Monte Carlo slope reliability analysis method based on limitequilibrium. In the calculation, angle of friction and cohesion force are considered as stochastic and independent variables. According to the data of test, the numeral characteristics of variables are acquired by statistic analysis, and then random numbers which are as same distribution as the variables are generated. Through calculating under four compounding conditions of water and seism, it shows that water must be considered in stability analysis of rainfall-induced landslide. The safety factor probability histograms show that the probability of a certain safety factor in different range is unequal, and the safety factor which probability value is maximal has correlation with the one obtained in stability analysis. The probability of slope failure and distribution of safety factor are ascertained, which can show the slope stability as better.
Keywords/Search Tags:slope reliability, slope stability, Monte Carlo method, genetic algorithm
PDF Full Text Request
Related items