The comparison of the performance of ELM,BRNN, and SVM methods for the prediction of compression index of clays

The compression index (Cc) is a necessary parameter for the settlement calculation of clays. However, determination of the compression index from oedometer tests takes a relatively long time and leads to a very demanding experimental working program in the laboratory. Therefore, geotechnical engineering literature involves many studies based on indirect methods such as multiple regression analysis (MLR) and soft computing methods to determine the compression index. This study is aimed to predict the compression index by using extreme learning machine (ELM), Bayesian regularization neural network (BRNN), and support vector machine (SVM) methods. The selected variables for each method are the natural water content (w_n), initial void ratio (e₀), liquid limit (LL), and plasticity index (PI) of clay samples. Many trials were carried out in order to get the best prediction performance with each model. The application results obtained from the models were also compared based on the correlation coefficient (R), coefficient of efficiency (E), and mean squared error (MSE). The results indicate that the BRNN method has better success on estimation of the compression index compared to the ELM and SVM methods.     

Mapping horizontal displacement of soil nail walls using machine learning approaches

Machine learning (ML) approaches have stormed nearly all engineering fields since recent years. However, the situation is somehow subtle in civil engineering practice, especially in the sub-field of geotechnical engineering where data from real-life projects are usually scarce, which in turn prevents development of meaningful mapping tools based on ML techniques. This study first shares a database containing a total of 376 measured horizontal displacements from instrumented soil nail walls reported in the literature. Then, these data are utilized to develop three types of ML models for mapping the wall horizontal displacement along depth, including artificial neural network (ANN), random forest (RF), and support vector machine (SVM). The uncertainties of the ANN, RF, and SVM models are then quantitatively evaluated using bias statistics where bias is defined as the ratio of measured to predicted horizontal displacement. The three ML models are proved to be accurate on average with medium dispersions in prediction, which outperform the existing simple empirical regression models. Probability distribution functions for those biases are also characterized. This study demonstrates that introduction of machine learning approaches into the reliability-based design framework for soil nail walls and other geotechnical structures is promising.

     

基于机器学习模型的淮河流域中长期径流预报研究

准确可靠的中长期径流预报是支撑水资源科学调配、提高水资源利用效率的关键。本研究采用AdaBoost模型(AdB)、随机森林模型(RF)和支持向量机模型(SVM)进行淮河流域王家坝和蚌埠站当年11月至次年10月共12个月的中长期径流预报研究。采用置换准确度重要性度量法从130项气象-气候因子及前期降雨/流量构建的1 562个因子变量中筛选出影响各月径流的关键因子,构建了基于AdB、RF和SVM模型的各月径流预报模型,模型参数采用随机搜索技术并结合交叉验证方式确定。采用变幅误差合格率和等级(五级)预报合格率指标对模型的预报精度进行了评估。变幅误差合格率指标表明,王家坝12个月的平均合格率分别为99.8%(AdB)、96.6%(RF)和95.9%(SVM),蚌埠站分别为100%(AdB)、94.8%(RF)和93.8%(SVM);等级预报合格率指标表明,王家坝12个月的平均合格率分别为79.0%(AdB)、76.4%(RF)和79.9%(SVM),蚌埠站分别为81.0%(AdB)、75.6%(RF)和76.6%(SVM)。模型均具有较好的预报效果,但RF和SVM模型对于高流量值的预报存在偏低现象,AdB模型整体上优于RF和SVM模型。     

滑坡易发性预测不确定性:环境因子不同属性区间划分和不同数据驱动模型的影响

对于滑坡易发性预测建模,连续型环境因子在频率比分析时的属性区间划分数量（attribute interval numbers,AIN）和不同易发性预测模型是两个重要不确定性因素.为研究这两个因素对建模的影响规律,以江西省上犹县为例,考虑5种连续型环境因子AIN划分（4、8、12、16及20）和5种数据驱动模型（层次分析法（analytic hierarchy process,AHP）、逻辑回归（logistic regression,LR）、BP神经网络（back-propagation neural network,BPNN）、支持向量机（support vector machine,SVM）和随机森林（random forest,RF））,总计25种不同工况下的滑坡易发性预测研究.再开展滑坡易发性指数的不确定性（包括精度评价和统计规律等）分析.结果表明:（1）对于同一模型,随着AIN值从4增加至8再到20时,易发性预测精度先逐渐提升,然后缓慢提升直至稳定;（2）对于同一AIN值,RF模型预测精度最高,其后依次为SVM、BPNN、LR和AHP模型;（3）在25种组合工况下,AIN=20和RF模型的预测精度最高,AIN=4和AHP模型精度最低,但在AIN=8和RF模型组合下的易发性建模效率较高且精度也较高;（4）更大的AIN值和更先进的机器学习模型预测出的滑坡易发性指数的不确定性相对较低,更符合实际的滑坡概率分布特征.在环境因子属性区间划分为8和RF模型工况下高效准确地构建滑坡易发性预测模型.      

基于机器学习和测井数据的碳酸盐岩孔隙度与渗透率预测

准确预测碳酸盐岩储层孔隙度和渗透率对于碳酸盐岩油气藏储层评价具有重要意义。碳酸盐岩储层裂缝与溶孔广泛发育,基于经验公式从测井曲线预测储层孔隙度和渗透率具有较大误差。以中东某碳酸盐岩油藏为研究对象,选取914块取心井岩心,测定孔隙度与渗透率,利用随机森林（RF）、K-近邻（KNN）、支持向量机（SVM）和长短期记忆网络（LSTM）4种不同机器学习方法,通过测井数据进行孔隙度与渗透率预测,优化机器学习参数,筛选出适用于碳酸盐岩油藏的测井孔隙度与渗透率预测方法。研究结果表明：4种机器学习方法预测储层孔隙度结果差异不大,通过调整输入参数种类,可进一步提高孔隙度与渗透率预测效果,当以补偿中子（NPHI）、岩性密度（RHOB）和声波时差（DT）3种测井参数数据作为输入时,基于LSTM的储层孔隙度预测精度最高,孔隙度预测结果均方根误差（RMSE）为4.536 2;由于碳酸盐岩储层的强非均质性,基于机器学习的测井储层渗透率预测效果较差,相对而言,仅以NPHI作为机器学习输入参数时,基于RF的储层渗透率预测精度最高,渗透率预测结果的RMSE为45.882 3。     

How do machine learning techniques help in increasing accuracy of landslide susceptibility maps?

Landslides are abundant in mountainous regions.They are responsible for substantial damages and losses in those areas.The A1 Highway,which is an important road in Algeria,was sometimes constructed in mountainous and/or semi-mountainous areas.Previous studies of landslide susceptibility mapping conducted near this road using statistical and expert methods have yielded ordinary results.In this research,we are interested in how do machine learning techniques help in increasing accuracy of landslide susceptibility maps in the vicinity of the A1 Highway corridor.To do this,an important section at Ain Bouziane(NE,Algeria) is chosen as a case study to evaluate the landslide susceptibility using three different machine learning methods,namely,random forest(RF),support vector machine(SVM),and boosted regression tree(BRT).First,an inventory map and nine input factors were prepared for landslide susceptibility mapping(LSM) analyses.The three models were constructed to find the most susceptible areas to this phenomenon.The results were assessed by calculating the receiver operating characteristic(ROC) curve,the standard error(Std.error),and the confidence interval(CI) at 95%.The RF model reached the highest predictive accuracy(AUC=97.2%) comparatively to the other models.The outcomes of this research proved that the obtained machine learning models had the ability to predict future landslide locations in this important road section.In addition,their application gives an improvement of the accuracy of LSMs near the road corridor.The machine learning models may become an important prediction tool that will identify landslide alleviation actions.     

不同环境因子联接和预测模型的滑坡易发性建模不确定性

拟深入探讨滑坡与其环境因子间的非线性联接计算以及不同数据驱动模型等因素,对滑坡易发性预测建模不确定性的影响规律.以江西省瑞金市为例共获取370处滑坡和10种环境因子,通过概率统计（probability statistics,PS）、频率比（frequency ratio,FR）、信息量（information value,Ⅳ）、熵指数（index of entropy,IOE）和证据权（weight of evidence,WOE）等5种联接方法分别耦合逻辑回归（logistic regression,LR）、BP神经网络（BP neural networks,BPNN）、支持向量机（support vector machines,SVM）和随机森林（random forest,RF）模型共构建出20种耦合模型,同时构建无联接方法直接将原始数据作为输入变量的4种单独LR、BPNN、SVM和RF模型,预测出总计24种工况下的滑坡易发性;最后分别使用ROC曲线、均值、标准差和差异显著性等指标分析上述24种工况下易发性结果的不确定性.结果表明:（1）基于WOE的耦合模型预测滑坡易发性的平均精度最高且不确定性较低,基于PS的耦合模型预测精度最低且不确定性最高,基于FR、Ⅳ和IOE的耦合模型介于两者之间;（2）单独数据驱动模型易发性预测精度略低于耦合模型,且未能计算出环境因子各子区间对滑坡发育的影响规律,但其建模效率高于耦合模型;（3）RF模型预测精度最高且不确定性较低,其次分别为SVM、BPNN和LR模型.总之WOE是更优秀的联接法且RF模型预测性能最优,WOE-RF模型预测的滑坡易发性不确定性较低且更符合实际滑坡概率分布特征.      

Relevance vector machine applied to slope stability analysis

This paper examines the potential of relevance vector machine (RVM) in slope stability analysis. The nonlinear relationship between slope stability and its influence factors is presented by the relevance vector learning mechanism based on a kernel‐based Bayesian framework. The six input variables used for the RVM for the prediction of stability slope are density (γ), friction angle (C), friction coefficient (?), slope angle (?_r), slope height (H), and pore water pressure (r_u). Comparison of RVM with some other methods is also presented. RVM has been used to compute the error bar. The results presented in this paper clearly highlight that the RVM is a robust tool for the prediction of slope stability. The experimental results show the effectiveness of the proposed approach. Copyright © 2011 John Wiley & Sons, Ltd.     

Active and online prediction of BOD5 in river systems using reduced-order support vector machine

Due to the limitations of hardware sensors for online measurement of the water quality parameters such as 5-day biochemical oxygen demand (BOD₅), the recent research efforts have focused on the software sensors for the rapid prediction of such parameters. The main objective in this research is to develop a reduced-order support vector machine (ROSVM) model based on the proper orthogonal decomposition to solve the time-consuming problem of the BOD₅ measurements. The performance of the newly developed methodology is tested on the Sefidrood River Basin, Iran. Subsequently, the predicted values of BOD₅, resulted from the selected developed ROSVM model, are compared with the results of support vector machine (SVM) model. According to the obtained results, selected ROSVM model seems to be more accurate, showing Person correlation coefficient (R) and root mean square error (RMSE) equal to 0.97 and 6.94, respectively. Further, the investigations based on developed discrepancy ratio (DDR) statistic for selection of the optimum model between the best accurate ROSVM and SVM models are carried out. Results of DDR statistic indicated superior performance of the selected ROSVM model comparing to the SVM technique for online prediction of BOD₅ in the Sefidrood River.     

基于SAPSO-ELM的瓦斯涌出量分源预测及应用

为了提高瓦斯涌出量预测精度,针对瓦斯涌出量影响因素的多重相关性、复杂性等问题,结合主成分分析法和分源预测理论,对开采层、邻近层、采空区的瓦斯涌出量数据分别进行主成分分析降维,得到预测指标。针对极限学习机(ELM)存在的输入权值矩阵与隐含层阈值随机生成的问题,利用模拟退火粒子群算法(SAPSO)对极限学习机的参数寻优,将新疆某煤矿回采工作面瓦斯涌出量及影响因素作为SAPSO-ELM模型的输入进行训练,再利用训练好的SAPSO-ELM模型对陕西某煤矿回采工作面的瓦斯涌出量进行验证预测,并对比原始ELM模型的预测结果。结果表明,SAPSO-ELM模型的平均相对误差为3.45%,ELM模型的平均相对误差为8.81%,与ELM模型相比,SAPSO-ELM模型预测精度及效率均优于原始ELM模型。分源预测理论和主成分分析法的结合有效解决了多因素间的多重相关性并降低了预测模型的复杂度,SAPSO-ELM预测模型实现了瓦斯涌出量的快速精准预测,对预防瓦斯事故发生和保障煤矿安全高效开采具有较好的指导作用。      

Modelling of shallow landslides with machine learning algorithms

This paper introduces three machine learning(ML)algorithms,the‘ensemble'Random Forest(RF),the‘ensemble'Gradient Boosted Regression Tree(GBRT)and the Multi Layer Perceptron neural network(MLP)and applies them to the spatial modelling of shallow landslides near Kvam in Norway.In the development of the ML models,a total of 11 significant landslide controlling factors were selected.The controlling factors relate to the geomorphology,geology,geo-environment and anthropogenic effects:slope angle,aspect,plan curvature,profile curvature,flow accumulation,flow direction,distance to rivers,water content,saturation,rainfall and distance to roads.It is observed that slope angle was the most significant controlling factor in the ML analyses.The performance of the three ML models was evaluated quantitatively based on the Receiver Operating Characteristic(ROC)analysis.The results show that the‘ensemble'GBRT machine learning model yielded the most promising results for the spatial prediction of shallow landslides,with a 95%probability of landslide detection and 87%prediction efficiency.     

WA联合ELM与OS-ELM的滑坡位移预测模型

滑坡累积位移监测曲线往往呈现出复杂的非线性增长特性,对此建立了不少相关的预测模型,而以往的预测模型存在着许多不足。本文基于小波函数（Wavelet Analysis,WA）,ELM与OS-ELM,提出一种名为WA联合ELM、OS-ELM的预测方法。首先,该方法基于小波函数,将滑坡累积位移分解成受内部地质条件影响的趋势项和受外部影响因子影响的周期项;然后,基于ELM与OS-ELM分别对趋势项和周期项进行预测;最后将趋势项和周期项的预测值叠加得到累积位移的预测值。结果表明,小波函数得到的趋势项展现出良好的趋势性,而周期项也展现出良好的周期性;以Sigmoid方程为核函数,隐含层神经元个数为33的ELM模型能准确高效对趋势项进行预测,而以RBF方程为核函数,隐含层神经元个数为100的OS-ELM模型能准确高效对周期项进行预测;累积位移预测数据的RMSE分别为0.1423和0.1315,预测结果相对较好,能够在滑坡位移预测领域发挥一定的作用。     

Application of statistical learning algorithms to ultimate bearing capacity of shallow foundation on cohesionless soil

This study employs two statistical learning algorithms (Support Vector Machine (SVM) and Relevance Vector Machine (RVM)) for the determination of ultimate bearing capacity (q_u) of shallow foundation on cohesionless soil. SVM is firmly based on the theory of statistical learning, uses regression technique by introducing varepsilon‐insensitive loss function. RVM is based on a Bayesian formulation of a linear model with an appropriate prior that results in a sparse representation. It also gives variance of predicted data. The inputs of models are width of footing (B), depth of footing (D), footing geometry (L/B), unit weight of sand (γ) and angle of shearing resistance (?). Equations have been developed for the determination of q_u of shallow foundation on cohesionless soil based on the SVM and RVM models. Sensitivity analysis has also been carried out to determine the effect of each input parameter. This study shows that the developed SVM and RVM are robust models for the prediction of q_u of shallow foundation on cohesionless soil. Copyright © 2010 John Wiley & Sons, Ltd.     

Machine learning techniques applied to prediction of residual strength of clay

Stability with first time or reactivated landslides depends upon the residual shear strength of soil. This paper describes prediction of the residual strength of soil based on index properties using two machine learning techniques. Different Artificial Neural Network (ANN) models and Support Vector Machine (SVM) techniques have been used. SVM aims at minimizing a bound on the generalization error of a model rather than at minimizing the error on the training data only. The ANN models along with their generalizations capabilities are presented here for comparisons. This study also highlights the capability of SVM model over ANN models for the prediction of the residual strength of soil. Based on different statistical parameters, the SVM model is found to be better than the developed ANN models. A model equation has been developed for prediction of the residual strength based on the SVM for practicing geotechnical engineers. Sensitivity analyses have been also performed to investigate the effects of different index properties on the residual strength of soil.     

基于邻域粗糙集和支持向量机的固结系数预测

采用邻域粗糙集和支持向量机建立滹沱河某地区软土固结系数的预测模型。基于自行改装的渗透固结仪,利用公式法确定不同压力下的固结系数。通过室内试验确定土体的指标参数,采用邻域粗糙集对该指标参数进行属性约简,将约简后的指标参数作为影响因素,分别建立支持向量机和神经网络的固结系数预测模型,预测未知样本的固结系数,并与实测值进行对比。结果表明:公式法可以准确客观地确定固结系数;支持向量机和BP神经网络建立的该地区软土固结系数预测模型均可以预测区域内未知点的固结系数,且支持向量机方法的预测精度比神经网络方法的预测精度提高了约10%。本文提出的方法直接从实验数据出发,通过易获取的影响因素建立特定地区固结系数预测模型,并可预测该区域其余未知点的固结系数。     

Displacement prediction of step-like landslide by applying a novel kernel extreme learning machine method

Landslide displacement prediction is an essential component for developing landslide early warning systems. In the Three Gorges Reservoir area (TGRA), landslides experience step-like deformations (i.e., periods of stability interrupted by abrupt accelerations) generally from April to September due to the influence of precipitation and reservoir scheduled level variations. With respect to many traditional machine learning techniques, two issues exist relative to displacement prediction, namely the random fluctuation of prediction results and inaccurate prediction when step-like deformations take place. In this study, a novel and original prediction method was proposed by combining the wavelet transform (WT) and particle swarm optimization-kernel extreme learning machine (PSO-KELM) methods, and by considering the landslide causal factors. A typical landslide with a step-like behavior, the Baishuihe landslide in TGRA, was taken as a case study. The cumulated total displacement was decomposed into trend displacement, periodic displacement (controlled by internal geological conditions and external triggering factors respectively), and noise. The displacement items were predicted separately by multi-factor PSO-KELM considering various causal factors, and the total displacement was obtained by summing them up. An accurate prediction was achieved by the proposed method, including the step-like deformation period. The performance of the proposed method was compared with that of the multi-factor extreme learning machine (ELM), support vector regression (SVR), backward propagation neural network (BPNN), and single-factor PSO-KELM. Results show that the PSO-KELM outperforms the other models, and the prediction accuracy can be improved by considering causal factors.     

A study of soft computing models for prediction of longitudinal wave velocity

Genetic algorithm (GA) and support vector machine (SVM) optimization techniques are applied widely in the area of geophysics, civil, biology, mining, and geo-mechanics. Due to its versatility, it is being applied widely in almost every field of engineering. In this paper, the important features of GA and SVM are discussed as well as prediction of longitudinal wave velocity and its advantages over other conventional prediction methods. Longitudinal wave measurement is an indicator of peak particle velocity (PPV) during blasting and is an important parameter to be determined to minimize the damage caused by ground vibrations. The dynamic wave velocity and physico-mechanical properties of rock significantly affect the fracture propagation in rock. GA and SVM models are designed to predict the longitudinal wave velocity induced by ground vibrations. Chaos optimization algorithm has been used in SVM to find the optimal parameters of the model to increase the learning and prediction efficiency. GA model also has been developed and has used an objective function to be minimized. A parametric study for selecting the optimized parameters of GA model was done to select the best value. The mean absolute percentage error for the predicted wave velocity (V) value has been found to be the least (0.258 %) for GA as compared to values obtained by multivariate regression analysis (MVRA), artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), and SVM.     

Spatial landslide susceptibility assessment using machine learning techniques assisted by additional data created with generative adversarial networks

In recent years,landslide susceptibility mapping has substantially improved with advances in machine learning.However,there are still challenges remain in landslide mapping due to the availability of limited inventory data.In this paper,a novel method that improves the performance of machine learning techniques is presented.The proposed method creates synthetic inventory data using Generative Adversarial Networks(GANs)for improving the prediction of landslides.In this research,landslide inventory data of 156 landslide locations were identified in Cameron Highlands,Malaysia,taken from previous projects the authors worked on.Elevation,slope,aspect,plan curvature,profile curvature,total curvature,lithology,land use and land cover(LULC),distance to the road,distance to the river,stream power index(SPI),sediment transport index(STI),terrain roughness index(TRI),topographic wetness index(TWI)and vegetation density are geo-environmental factors considered in this study based on suggestions from previous works on Cameron Highlands.To show the capability of GANs in improving landslide prediction models,this study tests the proposed GAN model with benchmark models namely Artificial Neural Network(ANN),Support Vector Machine(SVM),Decision Trees(DT),Random Forest(RF)and Bagging ensemble models with ANN and SVM models.These models were validated using the area under the receiver operating characteristic curve(AUROC).The DT,RF,SVM,ANN and Bagging ensemble could achieve the AUROC values of(0.90,0.94,0.86,0.69 and 0.82)for the training;and the AUROC of(0.76,0.81,0.85,0.72 and 0.75)for the test,subsequently.When using additional samples,the same models achieved the AUROC values of(0.92,0.94,0.88,0.75 and 0.84)for the training and(0.78,0.82,0.82,0.78 and 0.80)for the test,respectively.Using the additional samples improved the test accuracy of all the models except SVM.As a result,in data-scarce environments,this research showed that utilizing GANs to generate supplementary samples is promising because it can improve the predictive capability of common landslide prediction models.     

Prediction of water quality parameters of Karoon River (Iran) by artificial intelligence-based models

This paper describes the application of multi-layer perceptron (MLP), radial basis network and adaptive neuro-fuzzy inference system (ANFIS) models for computing dissolved oxygen (DO), biochemical oxygen demand (BOD) and chemical oxygen demand (COD) levels in the Karoon River (Iran). Nine input water quality variables including EC, PH, Ca, Mg, Na, Turbidity, PO₄, NO₃ and NO₂, which were measured in the river water, were employed for the models. The performance of these models was assessed by the coefficient of determination R ², root mean square error and mean absolute error. The results showed that the computed values of DO, BOD and COD using both the artificial neural network and ANFIS models were in close agreement with their respective measured values in the river water. MLP was also better than other models in predicting water quality variables. Finally, the sensitive analysis was done to determine the relative importance and contribution of the input variables. The results showed that the phosphate was the most effective parameters on DO, BOD and COD.     

基于时间序列分解和多变量混沌模型的滑坡阶跃式位移预测

三峡库区某些库岸滑坡在强降雨、库水位涨落等诱发因素影响下,其位移时间序列表现出阶跃式变化特征且可能存在混沌特性.但目前常用于滑坡位移预测的混沌模型,均建立在单变量混沌理论的基础之上.且已有的考虑了诱发因素的常规多变量模型,大都采用经验性的方法来选取输入变量;常规多变量模型对滑坡位移序列的非线性特征,及其与诱发因素间的动态响应关系缺乏数学理论上的深入分析.因此,提出一种基于指数平滑法、多变量混沌模型和极限学习机（extreme learing machine,ELM）的滑坡位移组合预测模型.指数平滑多变量混沌ELM模型首先对滑坡累积位移序列的混沌特性进行识别;然后用指数平滑法对累积位移进行预测,得到趋势项位移,并用累积位移减去趋势项位移得到剩余的波动项位移;之后对波动项位移及降雨量、库水位变化量这3个因子进行多变量相空间重构,并用ELM模型对多变量重构后的波动项位移进行预测;最后将预测得到的趋势项和波动项位移值相加,得到最终的累积位移预测值.以三峡库区白水河滑坡ZG93监测点的累积位移作为实例进行分析,并将模型与指数平滑多变量混沌粒子群-支持向量机（PSO-SVM）模型、指数平滑单变量混沌ELM模型作对比.结果表明滑坡位移序列存在混沌特性,模型能有效预测滑坡位移,其预测效果优于对比模型.且本文模型从混沌理论的角度将波动项位移与降雨量、库水位变化量的动态响应关系进行综合分析,更能反映滑坡位移系统演化的物理本质.