Non-probabilistic Reliability Analysis of Slopes Based on Blind Number Theory
Aiming at the problems that the single non-probability reliability index provides limited in-formation and the distribution of slope sliding surface is not involved,a non-probability reliability a-nalysis method for slope based on blind number theory is proposed.Considering the geotechnical pa-rameters as the blind numbers,the sample data is arranged in ascending order and divided into subin-tervals.The subintervals of different parameters are combined,the corresponding ellipsoid models are constructed,and then the non-probability reliability indexes and their blind number credibility are ob-tained,and the critical sliding surfaces with the minimum safety factors are searched.The blind num-ber credibility for which the non-probability reliability indexes are less than 1 is counted as the failure possibility of the slope.The feasibility of the proposed method is verified by example and case analysis.The results show that the division of parameter subintervals has a certain influence on the calculation results,and it is recommended to divide 10 subintervals.Compared with the reliability method which is based on blind number theory,the calculation results of proposed method are closer to those of Monte Carlo method.This method can provide the non-probabilistic reliability index of slope and the distribution of each sliding surface with the blind number credibility,which is suitable for application in practical engineering.
slopenon-probabilistic reliabilityblind number theoryblind number credibilityslip surface
万方数据
维普
