A hybrid machine learning ensemble approach based on a Radial Basis Function neural network and Rotation Forest for landslide susceptibility modeling: A case study in the Himalayan area,India

In this paper, a hybrid machine learning ensemble approach namely the Rotation Forest based Radial Basis Function (RFRBF) neural network is proposed for spatial prediction of landslides in part of the Himalayan area (India). The proposed approach is an integration of the Radial Basis Function (RBF) neural network classifier and Rotation Forest ensemble, which are state-of-the art machine learning algorithms for classification problems. For this purpose, a spatial database of the study area was established that consists of 930 landslide locations and fifteen influencing parameters (slope angle, road density, curvature, land use, distance to road, plan curvature, lineament density, distance to lineaments, rainfall, distance to river, profile curvature, elevation, slope aspect, river density, and soil type). Using the database, training and validation datasets were generated for constructing and validating the model. Performance of the model was assessed using the Receiver Operating Characteristic (ROC) curve, area under the ROC curve (AUC), statistical analysis methods, and the Chi square test. In addition, Logistic Regression (LR), Multi-layer Perceptron Neural Networks (MLP Neural Nets), Naïve Bayes (NB), and the hybrid model of Rotation Forest and Decision Trees (RFDT) were selected for comparison. The results show that the proposed RFRBF model has the highest prediction capability in comparison to the other models (LR, MLP Neural Nets, NB, and RFDT); therefore, the proposed RFRBF model is promising and should be used as an alternative technique for landslide susceptibility modeling.     

Application of back-propagation neural network on bank destruction forecasting for accumulative landslides in the three Gorges Reservoir Region,China

In recent years, a large number of bank destruction occur in the reservoir area under the effect of water fluctuation, which may be lead to reservoir accumulative landslide geological hazards finally. The paper conducted the bank destruction forecasting study for accumulative landslides in the Three Gorges Reservoir Region, China utilizing back-propagation (BP) neural network approach. A representative scenario of Jinle landslide is then taken for analysis purposes. On the basis of the existing data sets of bank destruction cases, the BP neural network forecasting model and the corresponding programs for bank destruction are both presented, whose forecasting result is validated by two independent approaches, namely empirical method and numerical modeling method. Furthermore, the BP neural network model had obvious advantages over the convention approaches in the aspects of the fast calculation speed and high convenience. According to the bank destruction forecasting scale presented above, the corresponding revetment measures can be proposed to prevent the occurring of the bank destruction, whose effectiveness has been further validated by the actual engineering practice.     

Forecasting effective drought index using a wavelet extreme learning machine (W-ELM) model

A drought forecasting model is a practical tool for drought-risk management. Drought models are used to forecast drought indices (DIs) that quantify drought by its onset, termination, and subsequent properties such as the severity, duration, and peak intensity in order to monitor and evaluate the impacts of future drought. In this study, a wavelet-based drought model using the extreme learning machine (W-ELM) algorithm where the input data are first screened through the wavelet pre-processing technique for better accuracy is developed to forecast the monthly effective DI (EDI). The EDI is an intensive index that considers water accumulation with a weighting function applied to rainfall data with the passage of time in order to analyze the drought-risk. Determined by the autocorrelation function (ACF) and partial ACFs, the lagged EDI signals for the current and past months are used as significant inputs for 1 month lead-time EDI forecasting. For drought model development, 97 years of data for three hydrological stations (Bathurst Agricultural, Wilsons Promontory and Merredin in Australia) are partitioned in approximately 90:5:5 ratios for training, cross-validation and test purposes, respectively. The discrete wavelet transformation (DWT) is applied to the predictor datasets to decompose inputs into their time–frequency components that capture important information on periodicities. DWT sub-series are used to develop new EDI sub-series as inputs for the W-ELM model. The forecasting capability of W-ELM is benchmarked with ELM, artificial neural network (ANN), least squares support vector regression (LSSVR) and their wavelet-equivalent (W-ANN, W-LSSVR) models. Statistical metrics based on agreement between the forecasted and observed EDI, including the coefficient of determination, Willmott’s index, Nash–Sutcliffe coefficient, percentage peak deviation, root-mean-square error, mean absolute error, and model execution time are used to assess the effectiveness of the models. The results demonstrate enhanced forecast skill of the drought models that use wavelet pre-processing of the predictor dataset. Based on statistical measures, W-ELM outperformed traditional ELM, LSSVR, ANN and their wavelet-equivalent counterparts (W-ANN, W-LSSVR). It is found that the W-ELM model is computationally efficient as shown by a faster running time with the majority of forecasting errors in lower frequency bands. The results demonstrate the usefulness of W-ELM over W-ANN and W-LSSVR models and the benefits of wavelet transformation of input data to improve the performance of drought forecasting models.     

Real‐time recurrent learning neural network for stream‐flow forecasting

Various types of neural networks have been proposed in previous papers for applications in hydrological events. However, most of these applied neural networks are classified as static neural networks, which are based on batch processes that update action only after the whole training data set has been presented. The time variate characteristics in hydrological processes have not been modelled well. In this paper, we present an alternative approach using an artificial neural network, termed real‐time recurrent learning (RTRL) for stream‐flow forecasting. To define the properties of the RTRL algorithm, we first compare the predictive ability of RTRL with least‐square estimated autoregressive integrated moving average models on several synthetic time‐series. Our results demonstrate that the RTRL network has a learning capacity with high efficiency and is an adequate model for time‐series prediction. We also investigated the RTRL network by using the rainfall–runoff data of the Da‐Chia River in Taiwan. The results show that RTRL can be applied with high accuracy to the study of real‐time stream‐flow forecasting networks. Copyright © 2002 John Wiley & Sons, Ltd.     

Predicting catchment flow in a semi‐arid region via an artificial neural network technique

A model of rainfall–runoff relationships is an essential tool in the process of evaluation of water resources projects. In this paper, we applied an artificial neural network (ANN) based model for flow prediction using the data for a catchment in a semi‐arid region in Morocco. Use of this method for non‐linear modelling has been demonstrated in several scientific fields such as biology, geology, chemistry and physics. The performance of the developed neural network‐based model was compared against multiple linear regression‐based model using the same observed data. It was found that the neural network model consistently gives superior predictions. Based on the results of this study, artificial neural network modelling appears to be a promising technique for the prediction of flow for catchments in semi‐arid regions. Accordingly, the neural network method can be applied to various hydrological systems where other models may be inappropriate. Copyright © 2004 John Wiley & Sons, Ltd.     

Multi-step water quality forecasting using a boosting ensemble multi-wavelet extreme learning machine model

The use of electrical conductivity (EC) as a water quality indicator is useful for estimating the mineralization and salinity of water. The objectives of this study were to explore, for the first time, extreme learning machine (ELM) and wavelet-extreme learning machine hybrid (WA-ELM) models to forecast multi-step-ahead EC and to employ an integrated method to combine the advantages of WA-ELM models, which utilized the boosting ensemble method. For comparative purposes, an adaptive neuro-fuzzy inference system (ANFIS) model, and a WA-ANFIS model, were also developed. The study area was the Aji-Chay River at the Akhula hydrometric station in Northwestern Iran. A total of 315 monthly EC (µS/cm) datasets (1984–2011) were used, in which the first 284 datasets (90% of total datasets) were considered for training and the remaining 31 (10% of total datasets) were used for model testing. Autocorrelation function (ACF) and partial autocorrelation function (PACF) demonstrated that the 6-month lags were potential input time lags. The results illustrated that the single ELM and ANFIS models were unable to forecast the multi-step-ahead EC in terms of root mean square error (RMSE), coefficient of determination (R²) and Nash–Sutcliffe model efficiency coefficient (NSC). To develop the hybrid WA-ELM and WA-ANFIS models, the original time series of lags as inputs, and time series of 1, 2 and 3 month-step-ahead EC values as outputs, were decomposed into several sub-time series using different maximal overlap discrete wavelet transform (MODWT) functions, namely Daubechies, Symlet, Haar and Coiflet of different orders at level three. These sub-time series were then used in the ELM and ANFIS models as an input dataset to forecast the multi-step-ahead EC. The results indicated that single WA-ELM and WA-ANFIS models performed better than any ELM and ANFIS models. Also, WA-ELM models outperformed WA-ANFIS models. To develop the boosting multi-WA-ELM and multi-WA-ANFIS ensemble models, a least squares boosting (LSBoost) algorithm was used. The results showed that boosting multi-WA-ELM and multi-WA-ANFIS ensemble models outperformed the individual WA-ELM and WA-ANFIS models.     

基于降雨入渗的Newmark模型改进及地震滑坡危险性预测研究

滑坡是一种破坏性非常强的地质灾害,其中地震与降雨均为诱导滑坡发生的关键因素。从降雨期间发生地震的角度考虑,基于Green-Ampt降雨入渗模型对Newmark模型进行改进,推导两因素耦合作用下的边坡安全系数F_S。以云南省鲁甸县某一区域为例,分别开展无降雨、降雨无积水与降雨积水三种情况下的地震滑坡危险性预测及坡度与入渗深度因子对位移影响分析。通过比较上述三种情况,得到研究区域内的Newmark累积位移分布及危险性区划。结果表明：与未降雨情况相比,后两种情况下地震滑坡高危险程度区域面积占比计算区域随着降雨时间的增加从1%分别提高至9%、12%,滑坡低危险程度区域面积从51%分别降低至35%、33%;坡度值与入渗深度值越大,滑坡位移越大,危险性越高。Newmark改进模型充分考虑了降雨对地震滑坡产生的促进作用,能更好地反映出研究区每个场点相对的滑坡危险性,对滑坡危险性预测具有一定指导意义。     

Simulation of streamflow with statistically downscaled daily rainfall using a hybrid of wavelet and GAMLSS models

ABSTRACT

A rainfall–streamflow model is proposed, in which a downscaled rainfall series and its wavelet-based decomposed sub-series at optimum lags were used as covariates in GAMLSS (Generalized Additive Model in Location, Scale and Shape). GAMLSS is applied in climate change impact assessment using CMIP5 general climate model to simulate daily streamflow in three sub-catchments of the Onkaparinga catchment, South Australia. The Spearman correlation and Nash-Sutcliffe efficiency between the observed and median simulated streamflow values were high and comparable for both the calibration and validation periods for each sub-catchment. We show that the GAMLSS has the capability to capture non-stationarity in the rainfall–streamflow process. It was also observed that the use of wavelet-based decomposed rainfall sub-series with optimum lags as covariates in the GAMLSS model captures the underlying physics of the rainfall–streamflow process. The development and application of an empirical rainfall–streamflow model that can be used to assess the impact of catchment-scale climate change on streamflow is demonstrated.     

Investigating the impact of the Chi‐Chi earthquake on the occurrence of debris flows using artificial neural networks

Debris flows have caused enormous losses of property and human life in Taiwan during the last two decades. An efficient and reliable method for predicting the occurrence of debris flows is required. The major goal of this study is to explore the impact of the Chi‐Chi earthquake on the occurrence of debris flows by applying the artificial neural network (ANN) that takes both hydrological and geomorphologic influences into account. The Chen‐Yu‐Lan River watershed, which is located in central Taiwan, is chosen for evaluating the critical rainfall triggering debris flows. A total of 1151 data sets were collected for calibrating model parameters with two training strategies. Significant differences before and after the earthquake have been found: (1) The size of landslide area is proportioned to the occurrence of debris flows; (2) the amount of critical rainfall required for triggering debris flows has reduced significantly, about half of the original critical rainfall in the study case; and (3) the frequency of the occurrence of debris flows is largely increased. The overall accuracy of model prediction in testing phase has reached 96·5%; moreover, the accuracy of occurrence prediction is largely increased from 24 to 80% as the network trained with data from before the Chi‐Chi earthquake sets and with data from the lumped before and after the earthquake sets. The results demonstrated that the ANN is capable of learning the complex mechanism of debris flows and producing satisfactory predictions. Copyright © 2009 John Wiley & Sons, Ltd.     

Indian summer monsoon rainfall prediction using artificial neural network

Forecasting the monsoon temporally is a major scientific issue in the field of monsoon meteorology. The ensemble of statistics and mathematics has increased the accuracy of forecasting of Indian summer monsoon rainfall (ISMR) up to some extent. But due to the nonlinear nature of ISMR, its forecasting accuracy is still below the satisfactory level. Mathematical and statistical models require complex computing power. Therefore, many researchers have paid attention to apply artificial neural network in ISMR forecasting. In this study, we have used feed-forward back-propagation neural network algorithm for ISMR forecasting. Based on this algorithm, we have proposed the five neural network architectures designated as BP1, BP2, $\ldots, $ … , BP5 using three layers of neurons (one input layer, one hidden layer and one output layer). Detail architecture of the neural networks is provided in this article. Time series data set of ISMR is obtained from Pathasarathy et al. (Theor Appl Climatol 49:217–224 1994) (1871–1994) and IITM (http://www.tropmet.res.in/, 2012) (1995–2010) for the period 1871–2010, for the months of June, July, August and September individually, and for the monsoon season (sum of June, July, August and September). The data set is trained and tested separately for each of the neural network architecture, viz., BP1–BP5. The forecasted results obtained for the training and testing data are then compared with existing model. Results clearly exhibit superiority of our model over the considered existing model. The seasonal rainfall values over India for next 5 years have also been predicted.     

A comparative study of extensive machine learning models for predicting long-term monthly rainfall with an ensemble of climatic and meteorological predictors

Rainfall prediction is of vital importance in water resources management. Accurate long-term rainfall prediction remains an open and challenging problem. Machine learning techniques, as an increasingly popular approach, provide an attractive alternative to traditional methods. The main objective of this study was to improve the prediction accuracy of machine learning-based methods for monthly rainfall, and to improve the understanding of the role of large-scale climatic variables and local meteorological variables in rainfall prediction. One regression model autoregressive integrated moving average model (ARIMA) and five state-of-the-art machine learning algorithms, including artificial neural networks, support vector machine, random forest (RF), gradient boosting regression, and dual-stage attention-based recurrent neural network, were implemented for monthly rainfall prediction over 25 stations in the East China region. The results showed that the ML models outperformed ARIMA model, and RF relatively outperformed other models. Local meteorological variables, humidity, and sunshine duration, were the most important predictors in improving prediction accuracy. 4-month lagged Western North Pacific Monsoon had higher importance than other large-scale climatic variables. The overall output of rainfall prediction was scalable and could be readily generalized to other regions.     

Research on progressive failure process of Baishuihe landslide based on Monte Carlo model

To the progressive landslide, development of the internal deformation and failure situation can’t be accurately reflected by the overall stability of coefficients and failure probability. But this problem can be solved by utilizing the principle of progressive failure by slices. Taking the warning area of Baishuihe landslide as an example, 5 days accumulated rainfall in different reappearing period is computed by Gumbel model. The failure probability of each slice is calculated by progressive failure principle, which is based on Monte Carlo model. The following results can be revealed through calculation: Overall stability and failure probability can’t reflect real situation of Baishuihe landslide warning area. Through building the calculation of progressive failure model of each slice, the stability of each part in the Baishuihe landslide warning area is quite different. Unstable region mainly lies in vicinity of the middle and posterior warning area. The front of the warning area remains stable. Deformation characteristics of the warning area are consistent with the investigation report. The scope of unstable area increased gradually with rainfall and the decline of reservoir water. Under 5 day’s accumulated rainfall of 50 years, the poor stable and unstable region reached 75 %, there is a large possibility of local deformation slip. Under the joint action of rainfall and reservoir water level, the warning area of Baishuihe landslide shows a progressive failure mode from top to bottom.     

基于ε-SVR的滑坡位移预测

我国是一个地质灾害多发的国家,特别是滑坡发生的次数比较多、危害性比较大。因此对滑坡的位移进行监控预测有着十分重要的意义。对于滑坡位移变化的非线性问题,可以利用支持向量机在回归分析中的方法——ε-支持向量回归机（ε-SVR）进行预测,该方法基于统计学理论,在处理小样本、非线性、高维数等问题上有一定的优势。以福建八尺门滑坡的监测数据为例,将前面的17个位移数据作为学习样本,后面的6个位移数据作为预测样本,采用不同的核函数分别进行位移预测来与原始监测值进行对比,比较其预测精度。结果显示,该方法产生的预测值与原始监测值之间的误差比较小,其位移变化趋势与原始数据的变化趋势也基本一致,这说明该方法预测精度高,实用性强。     

Multi-step prediction of strong earthquake ground motions and seismic responses of SDOF systems based on EMD-ELM method

This paper proposes a new multi-step prediction method of EMD-ELM (empirical mode decomposition-extreme learning machine) to achieve the short-term prediction of strong earthquake ground motions. Firstly, the acceleration time histories of near-fault ground motions with nonstationary property are decomposed into several components of intrinsic mode functions (IMFs) with different characteristic scales by the technique of EMD. Subsequently, the ELM method is utilized to predict the IMF components. Moreover, the predicted values of each IMF component are superimposed, and the short-term prediction of ground motions is attained with low error. The predicted results of near-fault acceleration records demonstrate that the EMD-ELM method can realize multi-step prediction of acceleration records with relatively high accuracy. Finally, the elastic and inelastic acceleration, velocity and displacement responses of single degree of freedom (SDOF) systems are also predicted with satisfactory accuracy by EMD-ELM method.     

Forecasting monthly precipitation using sequential modelling

In the hydrological cycle, rainfall is a major component and plays a vital role in planning and managing water resources. In this study, new generation deep learning models, recurrent neural network (RNN) and long short-term memory (LSTM), were applied for forecasting monthly rainfall, using long sequential raw data for time series analysis. “All-India” monthly average precipitation data for the period 1871–2016 were taken to build the models and they were tested on different homogeneous regions of India to check their robustness. From the results, it is evident that both the trained models (RNN and LSTM) performed well for different homogeneous regions of India based on the raw data. The study shows that a deep learning network can be applied successfully for time series analysis in the field of hydrology and allied fields to mitigate the risks of climatic extremes.     

Analysing the Relationship Between Typhoon‐Triggered Landslides and Critical Rainfall Conditions

Many landslides are triggered by rainfall. Previous studies of the relationship between landslides and rainfall have concentrated on deriving minimum rainfall thresholds that are likely to trigger landslides. Though useful, these minimum thresholds derived from a log–log plot do not offer any measure of confidence in a landslide monitoring or warning system. This study presents a new and innovative method for incorporating rainfall into landslide modelling and prediction. The method involves three steps: compiling radar reflectivity data in a QPESUMS (quantitative precipitation estimation and segregation using multiple sensors) system during a typhoon (tropical hurricane) event, estimating rainfall from radar data and using rainfall intensity and rainfall duration as explanatory variables to develop a landslide logit model. Given the logit model, this paper discusses ways in which the model can be used for computing probabilities of landslide occurrence for a real‐time monitoring system or a warning system, and for delineating and mapping landslides. Copyright © 2007 John Wiley & Sons, Ltd.     

Reservoir permeability prediction by neural networks combined with hybrid genetic algorithm and particle swarm optimization

Reservoir characterization involves describing different reservoir properties quantitatively using various techniques in spatial variability. Nevertheless, the entire reservoir cannot be examined directly and there still exist uncertainties associated with the nature of geological data. Such uncertainties can lead to errors in the estimation of the ultimate recoverable oil. To cope with uncertainties, intelligent mathematical techniques to predict the spatial distribution of reservoir properties appear as strong tools. The goal here is to construct a reservoir model with lower uncertainties and realistic assumptions. Permeability is a petrophysical property that relates the amount of fluids in place and their potential for displacement. This fundamental property is a key factor in selecting proper enhanced oil recovery schemes and reservoir management. In this paper, a soft sensor on the basis of a feed‐forward artificial neural network was implemented to forecast permeability of a reservoir. Then, optimization of the neural network‐based soft sensor was performed using a hybrid genetic algorithm and particle swarm optimization method. The proposed genetic method was used for initial weighting of the parameters in the neural network. The developed methodology was examined using real field data. Results from the hybrid method‐based soft sensor were compared with the results obtained from the conventional artificial neural network. A good agreement between the results was observed, which demonstrates the usefulness of the developed hybrid genetic algorithm and particle swarm optimization in prediction of reservoir permeability.     

Short-term prediction of influent flow in wastewater treatment plant

Predicting influent flow is important in the management of a wastewater treatment plant (WWTP). Because influent flow includes municipal sewage and rainfall runoff, it exhibits nonlinear spatial and temporal behavior and therefore makes it difficult to model. In this paper, a neural network approach is used to predict influent flow in the WWTP. The model inputs include historical influent data collected at a local WWTP, rainfall data and radar reflectivity data collected by the local weather station. A static multi-layer perceptron neural network performs well for the current time prediction but a time lag occurs and increases with the time horizon. A dynamic neural network with an online corrector is proposed to solve the time lag problem and increase the prediction accuracy for longer time horizons. The computational results show that the proposed neural network accurately predicts the influent flow for time horizons up to 300 min.     

A two‐step‐ahead recurrent neural network for stream‐flow forecasting

In many engineering problems, such as flood warning systems, accurate multistep‐ahead prediction is critically important. The main purpose of this study was to derive an algorithm for two‐step‐ahead forecasting based on a real‐time recurrent learning (RTRL) neural network that has been demonstrated as best suited for real‐time application in various problems. To evaluate the properties of the developed two‐step‐ahead RTRL algorithm, we first compared its predictive ability with least‐square estimated autoregressive moving average with exogenous inputs (ARMAX) models on several synthetic time‐series. Our results demonstrate that the developed two‐step‐ahead RTRL network has efficient ability to learn and has comparable accuracy for time‐series prediction as the refitted ARMAX models. We then investigated the two‐step‐ahead RTRL network by using the rainfall–runoff data of the Da‐Chia River in Taiwan. The results show that the developed algorithm can be successfully applied with high accuracy for two‐step‐ahead real‐time stream‐flow forecasting. Copyright © 2003 John Wiley & Sons, Ltd.     

Prediction of rainfall-triggered landslides: a test of the Antecedent Water Status Model

A rainfall-based landslide-triggering model, developed from previous landslide episodes in Wellington City, New Zealand, is tested for its ability to provide a 24-hour forecast of landslide occurrence. The model, referred to as the Antecedent Water Status Model, calculates an index of soil water, by running a daily water balance and applying a soil drainage factor to excess precipitation, over the preceding ten days. Together with the daily rainfall input, the soil water status has been used empirically to identify a threshold condition for landslide triggering. The prediction process provides a daily update of the soil water status and thereby the amount of rainfall required on the following day to equal or exceed the triggering threshold. The probability that this triggering rainfall will occur is then determined from the frequency/magnitude distribution of the local rainfall record. The model produces a satisfactory level of prediction, particularly for periods of concentrated landslide activity. Copyright © 1999 John Wiley & Sons, Ltd.