Data Mining and Deep Learning for Predicting the Displacement of “Step-Like” Landslides

Landslide displacement prediction is one of the unsolved challenges in the field of geological hazards, especially in reservoir areas. Affected by rainfall and cyclic fluctuations in reservoir water levels, a large number of landslide disasters have developed in the Three Gorges Reservoir Area. In this article, the Baishuihe landslide was taken as the research object. Firstly, based on time series theory, the landslide displacement was decomposed into three parts (trend term, periodic term, and random term) by Variational Mode Decomposition (VMD). Next, the landslide was divided into three deformation states according to the deformation rate. A data mining algorithm was introduced for selecting the triggering factors of periodic displacement, and the Fruit Fly Optimization Algorithm–Back Propagation Neural Network (FOA-BPNN) was applied to the training and prediction of periodic and random displacements. The results show that the displacement monitoring curve of the Baishuihe landslide has a “step-like” trend. Using VMD to decompose the displacement of a landslide can indicate the triggering factors, which has clear physical significance. In the proposed model, the R2 values between the measured and predicted displacements of ZG118 and XD01 were 0.977 and 0.978 respectively. Compared with previous studies, the prediction model proposed in this article not only ensures the calculation efficiency but also further improves the accuracy of the prediction results, which could provide guidance for the prediction and prevention of geological disasters.

Stability analysis of Baishuihe landslide under the combined effect of reservoir and rainfall

Reservoir drawdown and rainfall have important influence on bank landslides, but existing research on these two factors is too idealistic. A new reservoir drawdown model was proposed for the rapid drawdown stage based on the consideration of reduction, navigation and power generation. A rainfall model was proposed considering actural rainfall and rainfall time based on fifty years of daily rainfall data. At last, taking Baishuihe landslide as an example, the landslide stability was analyzed under the combined influenced of rainfall and reservoir drawdown. Results show that the Baishuihe landslide is mainly influenced by reservoir drawdown. The terminal reservoir drawdown model can reduce the effect of continuous decline of reservoir on landslide and the stability decreases about 0.7%~1.2% compared with normal scenario. The reservoir drawdown model proposed in this paper is of significance to the reservoir operation in the Three Gorges Reservoir.

Direct Interval Prediction of Landslide Displacements Using Least Squares Support Vector Machines

Accurate and reliable predictions of landslide displacements are difficult to perform using traditional point prediction approaches due to the associated uncertainty. Prediction intervals are effective tools for quantifying the uncertainty of point predictions by estimating the limit of future landslide displacements. In this paper, under the framework of the original lower upper bound estimation method, a direct interval prediction approach is proposed for landslide displacements based on the least squares support vector machine (LSSVM) and differential search algorithms. Two LSSVM models are directly implemented to generate the interval of future displacements, and the optimal model parameters are derived by the differential search algorithm. The Baishuihe landslide and the Tanjiahe landslide located on the shoreline of the Three Gorges Reservoir, China, are used to test the proposed approach. Compared with other models, the proposed method performed best and presented the smallest coverage width-based criterion values of 0.8927 and 1.0562 at monitoring stations XD01 and ZG118 for the Baishuihe landslide, respectively, and 0.1316 and 0.1191 at monitoring stations ZG289 and ZG287 for the Tanjiahe landslide, respectively. The results indicate that the proposed approach can provide high-quality prediction intervals for landslide displacements in the Three Gorges Reservoir area.

Study on the Spatially Variable Saturated Hydraulic Conductivity and Deformation Behavior of Accumulation Reservoir Landslide Based on Surface Nuclear Magnetic Resonance Survey

Saturated hydraulic conductivity (Ks) is spatially variable in accumulation landslide sites that exert significant effort onto landslide seepage and deformation behavior. To better understand spatial variability and the effect of Ks on the slide mass of an accumulation landslide, this study introduced the surface nuclear magnetic resonance (SNMR) technology to study a representative reservoir accumulation landslide field in the Three Gorges Reservoir area (TGRA), the Baishuihe landslide, to obtain a series of relative reliable spatial measurements of Ks effectively on the basis of calibration in terms of the field tests measurements. The estimated Ks values were distributed log-normally for the overall landslide mass site with a wide range of 3.00 × 10−6∼7.80 × 10−3 cm/s, which reaches about 3 orders of magnitude. Variogram analysis indicated that the Ks values have the range (A) of 295.89 m and 65.56 m for the overall site and major cross-sectional analysis, respectively. A finite-element seepage-stress analysis associated with a Kriging-interpolated spatial Ks variable calculation model based on the best-fitted theoretical variogram was subsequently performed to study the seepage and deformation behavior of the landslide. The available monitored data and simulated results of the finite-element seepage-stress analysis indicated that the Baishuihe landslide is a progressive landslide, and the main factor influencing the deformation is rainfall and reservoir water fluctuation. This study provides an unconventional framework for studying the heterogeneous geomaterial and contributes to a better understanding of the spatial variation of the hydraulic property of accumulation reservoir landslides at a field scale.