Search

Rainfall and reservoir water level are commonly regarded as the two major influencing factors for reservoir landslides and are employed for landslide displacement prediction, yet their daily data are readily available with current monitoring technology, which makes a more refined analysis possible. However, until now, few efforts have been made to predict landslide displacements using daily data, which is likely to substantially improve accuracy and is crucial for landslide early warning. A novel feature enhancement approach for extracting critical characteristics from daily rainfall and reservoir water level data for use in landslide displacement prediction is proposed in this study. Six models, including gated recurrent units (GRUs), long short-term memory (LSTM), and support vector regression (SVR) with an unenhanced dataset and GRU-E, LSTM-E, and SVR-E with an enhanced dataset, were employed for displacement predictions at four GPS monitoring stations on the Baijiabao landslide, a typical step-like reservoir landslide. The results show that the accuracy values of all the enhanced models were significantly improved, and the GRU-E model achieved the most significant improvement, with the RMSE decreasing by 24.39% and R2 increasing by 0.2693, followed by the LSTM-E and SVR-E models. Further, the GRU-E model consistently outperformed the other models, achieving the highest R2 of 0.6265 and the lowest RMSE of 16.5208 mm, significantly superior than the others. This study indicates the feasibility of improving the accuracy of landslide monthly displacement predictions with finer monitoring data and provides valuable insights for future research. [ABSTRACT FROM AUTHOR]

Accurate landslide displacement prediction has great practical significance for mitigating geohazards. Traditional deterministic forecasting methods can provide only a single point value and cannot give the degree of uncertainty associated with the forecast, thereby failing to provide information on predictive confidence. This study applied interval prediction for landslide displacement. Taking the Tanjiahe landslide of the Three Gorges Reservoir Area as an example and considering the impact of seasonal variations in reservoir level and rainfall, the uncertainties associated with landslide displacement prediction were quantified into prediction intervals (PIs) by a bootstrapped least-square support vector machine (LSSVM) method (B-LSSVM). The proposed method consists of three steps: First, the LSSVM and bootstrapping were combined to estimate the true regression means of landslide displacement and the variance with respect to model misspecification uncertainties. Second, a new LSSVM model optimized by a genetic algorithm (GA) was implemented to estimate the noise variance. Finally, the point prediction was derived from the regression means, and the PIs were constructed by combining the regression mean, the model variance, and the noise variance. We applied the proposed method to predict the displacement of four GPS monitoring points of the Tanjiahe landslide, and we comprehensively compared the prediction accuracy and the quality of the constructed PIs with benchmark methods. A simulation and performance comparison showed that the proposed method is a promising technique for providing accurate and reliable prediction results for landslide displacement. [ABSTRACT FROM AUTHOR]

There hides a certain relationship among various monitoring data in a landslide, and the mining of this relationship is of significance to landslide research. In this paper, we first collect multiple monitoring data of riverside 1# slump-mass of Huangtupo landslide, the Three Gorges Reservoir Region, China, including Global Positioning System (GPS) monitoring data, inclinometer data, reservoir water level, rainfall, water content, crack width, groundwater level and temperature data, etc. By adopting the combination of quantitative statistics and qualitative simulation method for multi-sensor fusion monitoring data analysis, we overcome the one-sidedness of using a single method or single data type. The result of fusion analysis has indicated that in time periods with low rainfall or when the rainfall is not the major factor, main factors affecting landslide movement are crack development, water content of the landslide and water level of the Three Gorges Reservoir. Compared with the actual monitoring data, the fusion analysis results has a maximum error of 1.9%, which shows a good effect. [ABSTRACT FROM AUTHOR]

This paper compares the performance of five popular machine learning methods, namely, particle swarm optimization–extreme learning machine (PSO–ELM), particle swarm optimization–kernel extreme learning machine (PSO–KELM), particle swarm optimization–support vector machine (PSO–SVM), particle swarm optimization–least squares support vector machine (PSO–LSSVM), and long short-term memory neural network (LSTM), in the prediction of reservoir landslide displacement. The Baishuihe, Shuping, and Baijiabao landslides in the Three Gorges reservoir area of China were used for case studies. Cumulative displacement was decomposed into trend displacement and periodic displacement by the Hodrick–Prescott filter. The double exponential smoothing method and the five machine learning methods were used to predict the trend and periodic displacement, respectively. The five machine learning methods are compared in three aspects: highest single prediction accuracy, mean prediction accuracy, and prediction stability. The results show that no method performed the best for all three aspects in the three landslide cases. LSTM and PSO–ELM achieved better single prediction accuracy, but worse mean prediction accuracy and stability. PSO–KELM, PSO–LSSVM, and PSO–SVM always yielded consistent predictions with slight variations. On the whole, PSO–KELM and PSO–LSSVM are recommended for their superior mean prediction accuracy and prediction stability. • The performances of machine learning methods depend on the selected evaluation index and reservoir landslide case. • Using single prediction accuracy for evaluating the superiority of machine learning methods may be unreliable. • PSO-KELM and PSO-LSSVM are recommended for their superior mean prediction accuracy and prediction stability. [ABSTRACT FROM AUTHOR]

With the widespread application of machine learning methods, the continuous improvement of forecast accuracy has become an important task, which is especially crucial for landslide displacement predictions. This study aimed to propose a novel prediction model to improve accuracy in landslide prediction, based on the combination of multiple new algorithms. The proposed new method includes three parts: data preparation, multi-swarm intelligence (MSI) optimization, and displacement prediction. In the data preparation, the complete ensemble empirical mode decomposition (CEEMD) is adopted to separate the trend and periodic displacements from the observed cumulative landslide displacement. The frequency component and residual component of reconstructed inducing factors that related to landslide movements are also extracted by the CEEMD and t-test, and then picked out with edit distance on real sequence (EDR) as input variables for the support vector regression (SVR) model. MSI optimization algorithms are used to optimize the SVR model in the MSI optimization; thus, six predictions models can be obtained that can be used in the displacement prediction part. Finally, the trend and periodic displacements are predicted by six optimized SVR models, respectively. The trend displacement and periodic displacement with the highest prediction accuracy are added and regarded as the final prediction result. The case study of the Shiliushubao landslide shows that the prediction results match the observed data well with an improvement in the aspect of average relative error, which indicates that the proposed model can predict landslide displacements with high precision, even when the displacements are characterized by stepped curves that under the influence of multiple time-varying factors. [ABSTRACT FROM AUTHOR]

In the Three Gorges Reservoir area (TGRA), landslides typically exhibit slow-moving characteristic deformations in response to seasonal changes in precipitation and reservoir level. This study proposed an optimal combination of methods in which the frequency components of decomposed inducing factors are considered for predicting landslide displacement. In preprocessing, the monitored surface displacements are decomposed into trend displacement and periodic displacement components via ensemble empirical mode decomposition (EEMD); the time series of the inducing factors (rainfall and reservoir level) were reconstructed and decomposed into high- and low-frequency time series via complete ensemble empirical mode decomposition (CEEMD) and t -test methods, among which two dominant inducing factors are identified by analyzing the longest common subsequence (LCSS) distance of the normalized data. In the prediction of the periodic displacement component, three optimal methods that use the support vector regression (SVR) model have been considered for comparison: the artificial bee colony (ABC) algorithm, the genetic algorithm (GA) and particle swarm optimization (PSO). The results demonstrate that the combination of CEEMD-LCSS reconstruction and ABC-SVR yields a robust model with remarkable prediction performance, which can increase the prediction accuracy of landslide displacements by 54% at most in the evaluation of RMSE. [ABSTRACT FROM AUTHOR]

Accurately predicting the surface displacement of the landslide is important and necessary. However, most of the existing research has ignored the frequency component of inducing factors and how it affects the landslide deformation. Therefore, a hybrid displacement prediction model based on time series theory and various intelligent algorithms was proposed in this paper to study the effect of frequency components. Firstly, the monitoring displacement of landslide from the Three Gorges Reservoir area (TGRA) was decomposed into the trend and periodic components by complete ensemble empirical mode decomposition (CEEMD). The trend component can be predicted by the least square method. Then, time series of inducing factors like rainfall and reservoir level was reconstructed into high frequency components and low frequency components with CEEMD and t-test, respectively. The dominant factors were selected by the method of dynamic time warping (DTW) from the frequency components and other common factors (e.g., current monthly rainfall). Finally, the ant colony optimization-based support vector machine regression (ACO-SVR) is utilized for prediction purposes in the TGRA. The results demonstrate that after considering the frequency components of landslide-induced factors, the accuracy of the displacement prediction model based on ACO-SVR is better than that of other models based on SVR and GA-SVR. [ABSTRACT FROM AUTHOR]

Time series prediction is an important tool for system analysis. Traditional forecasting methods usually achieve prediction by establishing static model or only mining the information from the sequence itself, but without considering the dynamics of the system or its triggering factors. Elman neural network (ENN) is a dynamical model that can remember historical states. In order to better establish the real multivariate time series model and improve the single-step prediction accuracy, this study proposes an improved regularized ENN method based on piecewise time weighted gradient (PWRENN). Each time the parameters of PWRENN model are updated, the weighted gradient is calculated by weighing the current gradient and the previous gradients according to a monotonically increasing temporal function. By considering both the current gradient and previous gradients, PWRENN is able to simulate the time-lag effect in the time series prediction problem. Moreover, the system environment may change with time, and the historical data of the same slot at a different time should have different effects on the predicted data. This work uses a piecewise time function with different parameters at different time periods to more accurately model the real system. In PWRENN, L2 regularization is adopted to solve the overfitting problem and enhance the generalization performance. Furthermore, the effectiveness of the proposed method is verified in an artificial Mackey-Glass time series prediction and three landslide displacement predictions. [ABSTRACT FROM AUTHOR]

The mechanical behavior of in situ sliding zone soil is unique and is significant for the state identification and prediction of landslides. This study focuses on interpreting the anisotropic deformation observed on the sliding zone soil of the Huangtupo landslide during an in situ triaxial creep test; specifically, the deformation of the A-Surface (towards 35° dip direction) of the soil cube, which points approximately to the landslide sliding orientation, is significantly large. Such anisotropy could be attributed to damaged structure, assumed as pre-existing directional cracks, of the in situ soil. Therefore, a Discrete Element Method (DEM) model containing prefabricated cracks is established to investigate the deforming behavior of compressed soil during in situ creep test on the mesoscopic scale. Results indicated that the specimen's anisotropic deformation is intimately related to the presence of prefabricated cracks. Subjected to the prefabricated cracks, microcracks generated during the loading process of the specimen initiated and developed in a directional way, subsequently forming directed cracks. Then the directed cracks tend to form a complete slip zone. However, due to the insufficient vertical load, the directed cracks were unable to coalesce, and the soil mass did not fail during the in situ creep test. Thereafter, additional load was applied in the simulation model. The directed cracks then merged and a full slip zone formed, and the soil was completely destroyed. As a consequence, the anisotropic characteristics of compressed in situ sliding zone soil can be well explained in view of the cracks development. This study may help perceive the landslide deformation characteristics and facilitate the evaluation and prediction methods of landslide stability. • Anisotropic deformation of the in situ sliding zone during compression • Formation of cracked slip zone in a compressed soil sample • Brand new recognition on the in situ sliding zone soil with prefabricated cracks [ABSTRACT FROM AUTHOR]

For landslide displacement, interval predictions are generally more realistic and reliable compared with traditional point predictions. This paper presents a new interval prediction method for landslide displacement integrating dual-output least squares support vector machine (DO-LSSVM) and particle swarm optimization (PSO) algorithms. In this new method, the PSO algorithm is employed to optimize coefficients of the least squares support vector machine (LSSVM) model for obtaining point prediction results, and the interval prediction of the landslide displacement is made based on the dual-outputs obtained from the DO-LSSVM model. To assess the rationality of the predictions, three performance evaluation indicators, including the prediction interval coverage probability (PICP), normalized mean prediction interval width (NMPIW), and coverage width-based criterion (CWC), are established. Case studies of the Tanjiahe landslide and the Baishuihe landslide in the Three Gorges Reservoir region are then used to demonstrate the effectiveness of the proposed method in predicting the landslide displacement interval. The case study results demonstrate that this new method has the best overall performance compared with other existing methods, and this new method can provide accurate and reliable results for the medium- to long-term interval prediction of landslide displacement. [ABSTRACT FROM AUTHOR]

Monitoring data show that many landslides in the Three Gorges region, China, undergo step-like displacements in response to the managed, quasi-sinusoidal annual variations in reservoir level. This behavior is consistent with motion initiating when the reservoir water level falls below a critical level that is intrinsic to each landslide, with the subsequent displacement rate of the landslide being proportional to the water depth below that critical level. Most motion terminates when the water level rises back above the critical level, so the annual step size is the time integral of the instantaneous displacement rate. These responses are incorporated into a differential equation that is easily calibrated with monitoring data, allowing prediction of landslide movement from actual or anticipated reservoir level changes. Model successes include (1) initiation and termination of the annual sliding steps at the critical reservoir level, producing a series of steps; (2) prediction of variable step size, year to year; and (3) approximate prediction of the shape and size of each annual step. Annual rainfall correlates poorly with step size, probably because its effect on groundwater levels is dwarfed by the 30 m annual variations in the level of the Three Gorges Reservoir. Viscous landslide behavior is suggested. [ABSTRACT FROM AUTHOR]

Interval prediction is an efficient approach to quantifying the uncertainties associated with landslide evolution. In this paper, a novel method, termed lower upper bound estimation (LUBE), of constructing prediction intervals (PIs) based on neural networks (NNs) is applied and extended to landslide displacement prediction. A random vector functional link network (RVFLN) is adopted as the NN used in the improved LUBE. A hybrid evolutionary algorithm, termed PSOGSA, that combines particle swarm optimization (PSO) and gravitational search algorithm (GSA) is utilized to train LUBE. The loss function of LUBE is redesigned by considering the quality of PI centre, which allows for a more comprehensive evaluation of PIs. The population initialization in the training process of LUBE is implemented by transferring the weights of a series of pre-trained RVFLNs. The performance of the improved LUBE method is validated by considering a comprehensive set of cases using seven benchmark datasets. In addition, a hybrid method that integrates ensemble empirical mode decomposition (EEMD) with the improved LUBE is proposed for the special case of landslide displacement prediction. Six real-world reservoir-induced landslides are considered to validate the capability and merit of the proposed hybrid method. [ABSTRACT FROM AUTHOR]

Data preprocessing is an indispensable step for landslide displacement forecasting, which is an effective approach for predicting the deformation and failure behaviors of landslides. However, most studies focus on the construction of displacement forecast models and ignore the influence of data preprocessing on the forecasting results. Data normalization is an important part of data preprocessing; however, the selection of a data normalization method is subjective and arbitrary. In this study, four types of normalization methods for data preprocessing are presented, and these methods are applied in forecasting the displacement of bank landslides in the Three Gorges Reservoir area with various deformation mechanisms for comparison. The results demonstrate that (1) the selected normalization method substantially influences the forecast performance; (2) the normalization method is closely related to the selected forecasting model and is less dependent on the landslide deformation mechanism; and (3) the commonly used max–min normalization approach is not the optimal method, and the zero-mean normalization method is optimal for the particle swarm optimizer of support vector machine (PSO-SVM) method, while the logarithmic normalization method is optimal for the extreme learning machine method. The obtained results suggest that the data preprocessing methods must be carefully selected in landslide displacement forecasting. [ABSTRACT FROM AUTHOR]