A multi-feature fusion transfer learning method for displacement prediction of rainfall reservoir-induced landslide with step-like deformation characteristics

The Green-Ampt (GA) model is one of the most widely used analytical methods of slope stability under rainfall. However, it may overestimate the soil water content above the wetting front. In this study, a novel approach to evaluate the time-dependent slope stability during intense rainfall based on a modified GA model is presented, and is known as the stratified Green-Ampt (SGA) model. By considering the stratified soil water content above the wetting front, the soil above the wetting front can be divided into saturated and transitional layers, and the SGA model is used to analyze the infiltration process of intense rainfall into slopes. Thereafter, safety factors (Fs) of infinite and finite slopes are derived using the SGA model. In the analysis of an infinite slope, the conventional limit equilibrium method is adopted to calculate the safety factor; as for a finite slope, the residual thrust method is introduced to obtain the safety factor with sliding mass divided into multiple soil slices. The performance of the SGA model is illustrated in two cases: an infinite slope and the Majiagou landslide as a finite slope. The results indicate that compared to the GA model, the calculated wetting front based upon the SGA model moves faster, and the wetting front depth shows a positive correlation with the slope surface angle and rainfall intensity. The evolution of the safety factor above the sliding surface can be divided into three phases, while the evolution of the safety factor above the wetting front can be divided into two phases. The critical time of the slope reaching a less stable state (safety factor is 1.05) or unstable state (safety factor is 1.00) decreases exponentially with an increase in rainfall intensity. In addition, the rainfall has a significant influence on the design of stabilizing piles for the Majiagou landslide. The presented SGA model appears to be accurate to investigate slope stability during intense rainfall events.

Time-dependent slope stability during intense rainfall with stratified soil water content

The slake durability; grain quantity, size, roundness, and slake degree of red-bed mudstones are experimentally investigated using the slake durability test and the cyclic wetting and drying test. ...

Experimental study of slaking properties of red-bed mudstones from the Three Gorges Reservoir area

Application of a multi-smartphone measurement system in slope model tests

A fast and practical method for determining particle size and shape by using smartphone photogrammetry

An increasing number of landslides are occurring during the construction of highways in mountainous areas all over the world. A recent highway landslide located in the mountainous region of Southwest China is taken as a representative example to carry out the initiation mechanism and deformation characteristics analysis, and study the intelligent partitioning method in this paper. The geological conditions of the landslide are obtained through detailed geological surveys and investigation. A three-dimensional (3D) monitoring network with a total of 72 multi-points, including 53 surface displacement monitoring points and 19 deep displacement monitoring holes, is established to examine the spatial deformation characteristics of the landslide for nearly 25 months. The monitoring results of the surface displacement show that the dynamic deformation of the landslide can be divided into three stages, namely initial deformation stage, accelerated deformation stage, and stabilization stage. Based on the strata distribution obtained by the borehole investigation and the results of deep displacement monitoring, the depth of the slip surface can be reliably determined. The smart contract method in blockchain technology combined with the multi-point monitoring dataset is novelty applied to realize the intelligent automatic partition of the highway landslide without any human involvement. This method can provide the basis for further independent analysis of different landslide zones. According to the investigation and analysis, the initiation mechanism for the highway landslides is attributed to both internal poor geological conditions and external intense precipitation and engineering excavation.

Mechanism analysis and partition characteristics of a recent highway landslide in Southwest China based on a 3D multi-point deformation monitoring system

The paper presents an improved plane layout for stabilizing piles based on a proposed piecewise function expression for the irregular driving force. Based on the specific morphological characteristics of a highway landslide, the piecewise function is used to calculate the irregular driving force by dividing the landslide into several sub-areas. Furthermore, the reasonable layout range and pile spacing can be obtained based on the piecewise function expression of the irregular driving force and on relevant research results of the plane layout for stabilizing piles. Therefore, an improved plane layout of stabilizing piles is presented in consideration of a piecewise function expression of the irregular driving force. A highway landslide located in eastern Guizhou Province, China, is analyzed as a case study using the proposed method. The results demonstrate that the theory presented in this paper provides improved economic benefits and can reduce the required number of stabilizing piles by 28.6% compared with the conventional plane layout scheme.

Improved plane layout of stabilizing piles based on the piecewise function expression of the irregular driving force

https://doi.org/10.1016/j.idm.2022.04.004

Qun Li

Papers

Mechanism analysis and partition characteristics of a recent highway landslide in Southwest China based on a 3D multi-point deformation monitoring system

Improved plane layout of stabilizing piles based on the piecewise function expression of the irregular driving force

Correlation between mumps and meteorological factors in Xiamen City, China: A modelling study