| The Loess Plateau has a unique loess-paleosol sedimentary strata,with 42 layers of loess and 41 layers of paleosol interbedded in the fully developed profile.The division of strata in the actual slope project is simplified,such as no stratification or the use of geological age in Q1,Q2 and Q3.The simplification of stratification inevitably results in the calculation of stability factor not accurately reflecting the actual situation,and it is more difficult to stratify and assign parameters according to the actual situation.To address this issue,this paper selects the loess-paleosol strata deposited above L9 in Wangcun,Gaozhuang Town,Jing Yang South Plateau.Graded exploration well sampling and study at the slope.Through laboratory tests,the parameters related to steady-state seepage and stability analysis were modelled in relation to the sedimentary age.An Excel-VBA program was used for the development of the steady-state seepage program,inverse analysis of the main factors based on the actual water content,and calibration of the parameter estimation model considering the sedimentary time effects was carried out.Coupling the finite element meshing function with the parameter model to develop a finite element analysis program for slope stability that takes into account the effect of sedimentary time.The results of the stability factor calculations are analyzed in comparison with the stratigraphic division methods commonly used in engineering.The following conclusions are mainly obtained.(1)A parameter estimation model for loess that systematically considers the effect of deposition time is establishedThe typical loess-paleosol strata of the southern plateau of Jing Yang was sampled and tested.A total of nine parameters,including saturated coefficient of permeability(ks),unsaturated permeability curve parameters(A1、A2、x0、p),soil-water characteristic parameters(a、m、n)and dry density(ρd),were determined in relation to the sedimentary age.The model of soil-water characteristic parameters was also calibrated by the measured water content.The problem of assigning stratigraphic parameters to multi-strata loess-paleosols was solved.(2)The mechanism of the natural water content fugacity pattern of the loess-paleosol was revealedThe water content of the entire profile was tested and found to have a sawtooth pattern of increasing water content with increasing depth of burial.This means that the water content in the loess layer gradually increases with depth,while the water content in the paleosol layer gradually decreases.The results of the principal component analysis of the factors influencing the water content show that the soil-water characteristics parameters and dry density are the main influencing factors,independent of the saturated coefficient of permeability and unsaturated coefficient of permeability.The higher water content of the paleosol layer compared to the loess layer is due to the high water holding capacity of the paleosol and is not due to the low coefficient of permeability of the paleosol,but to the strong water barrier effect.(3)A finite element algorithm for unsaturated slopes considering the parameter estimation model of sedimentary time effect is established.A finite element slope stability analysis program is developed using the FORTRAN language to consider the variation of gravity and shear strength caused by moisture in unsaturated soils,coupling the parameter estimation model considering the sedimentary time effect with the finite element mesh.The problem of inconsistency between stratigraphic division and actual loess-paleosol strata has been solved,and the input of parameters has been greatly reduced.(4)The influence of engineering layering method on slope stability is discussedThe finite element analysis program for slope stability in this paper assigns values to a parameter estimation model with sedimentary time effects,taking into account the loess-paleosol strata,moisture changes in the slopes,and unsaturated shear strength theory.The calculation of slope stability factor for four layered models with constant total number of parameters is carried out by thickness weighted averaging,i.e.1-layer,3-layer,L-S layer(17layers with the same parameters in each layer)and unsaturated parameter model considering sedimentary time effect.It was found that as the number of stratified layers increased,the slope stability deteriorated and the failure mode changed from global failure to local failure.Compared with models estimating parameters considering the sedimentary time effect,the L-S layering underestimates the stability factor by 20%and the commonly used 3-layer model for engineering(strength parameter based on natural water content)underestimates the stability factor by 5.5%. |