Showing papers in "Bulletin of Engineering Geology and the Environment in 2019"

A novel hybrid intelligent model of support vector machines and the MultiBoost ensemble for landslide susceptibility modeling

Abstract: The main aim of this study is to propose a novel hybrid intelligent model named MBSVM which is an integration of the MultiBoost ensemble and a support vector machine (SVM) for modeling of susceptibility of landslides in the Uttarakhand State, Northern India. Firstly, a geospatial database for the study area was prepared, which includes 391 historical landslides and 16 landslide-affecting factors. Then, the sensitivity of different combinations of these factors for modeling was validated using the forward elimination technique. The MBSVM landslide model was built using the datasets generated from the best selected factors and validated utilizing the area under the receiver operating characteristic (ROC) curve (AUC), statistical indexes, and the Wilcoxon signed-rank test. Results show that this novel hybrid model has good performance both in terms of goodness of fit with the training dataset (AUC = 0.972) and the capability to predict landslides with the testing dataset (AUC = 0.966). The efficiency of the proposed model was then validated by comparison with logistic regression (LR), a single SVM, and another hybrid model of the AdaBoost ensemble and an SVM (ABSVM). Comparison results show that the MBSVM outperforms the LR, single SVM, and hybrid ABSVM models. Thus, the proposed model is a promising and good alternative tool for landslide hazard assessment in landslide-prone areas.

Novel hybrid artificial intelligence approach of bivariate statistical-methods-based kernel logistic regression classifier for landslide susceptibility modeling

Abstract: Globally, and in China, landslides constitute one of the most important and frequently encountered natural hazard events. In the present study, landslide susceptibility evaluation was undertaken using novel ensembles of bivariate statistical-methods-based (evidential belief function (EBF), statistical index (SI), and weights of evidence (WoE)) kernel logistic regression machine learning classifiers. A landslide inventory comprising 222 landslides and 15 conditioning factors (slope angle, slope aspect, altitude, plan curvature, profile curvature, stream power index, sediment transport index, topographic wetness index, distance to rivers, distance to roads, distance to faults, NDVI, land use, lithology, and rainfall) was prepared as the spatial database. Correlation analysis and selection of conditioning factors were conducted using multicollinearity analysis and classifier attribute evaluation methods, respectively. The receiver operating characteristic curve method was used to validate the models. The areas under the success rate (AUC_T) and prediction rate (AUC_P) curves and landslide density analysis were also used to assess the prediction capability of the landslide susceptibility maps. Results showed that the EBF-KLR hybrid model had the highest predictive capability in landslide susceptibility assessment (AUC values of 0.814 and 0.753 for the training and validation datasets, respectively; AUC_T value of 0.8511 and AUC_P value of 0.7615), followed in descending order by the SI-KLR and WoE-KLR hybrid models. These findings indicate that hybrid models could improve the predictive capability of bivariate models, and that the EBF-KLR is a promising hybrid model for the spatial prediction of landslides in susceptible areas.

Spatial prediction of landslide susceptibility using data mining-based kernel logistic regression, naive Bayes and RBFNetwork models for the Long County area (China)

Abstract: The main goal of this study is to assess and compare three advanced machine learning techniques, namely, kernel logistic regression (KLR), naive Bayes (NB), and radial basis function network (RBFNetwork) models for landslide susceptibility modeling in Long County, China. First, a total of 171 landslide locations were identified within the study area using historical reports, aerial photographs, and extensive field surveys. All the landslides were randomly separated into two parts with a ratio of 70/30 for training and validation purposes. Second, 12 landslide conditioning factors were prepared for landslide susceptibility modeling, including slope aspect, slope angle, plan curvature, profile curvature, elevation, distance to faults, distance to rivers, distance to roads, lithology, NDVI (normalized difference vegetation index), land use, and rainfall. Third, the correlations between the conditioning factors and the occurrence of landslides were analyzed using normalized frequency ratios. A multicollinearity analysis of the landslide conditioning factors was carried out using tolerances and variance inflation factor (VIF) methods. Feature selection was performed using the chi-squared statistic with a 10-fold cross-validation technique to assess the predictive capabilities of the landslide conditioning factors. Then, the landslide conditioning factors with null predictive ability were excluded in order to optimize the landslide models. Finally, the trained KLR, NB, and RBFNetwork models were used to construct landslide susceptibility maps. The receiver operating characteristics (ROC) curve, the area under the curve (AUC), and several statistical measures, such as accuracy (ACC), F-measure, mean absolute error (MAE), and root mean squared error (RMSE), were used for the assessment, validation, and comparison of the resulting models in order to choose the best model in this study. The validation results show that all three models exhibit reasonably good performance, and the KLR model exhibits the most stable and best performance. The KLR model, which has a success rate of 0.847 and a prediction rate of 0.749, is a promising technique for landslide susceptibility mapping. Given the outcomes of the study, all three models could be used efficiently for landslide susceptibility analysis.

Predicting tunnel boring machine performance through a new model based on the group method of data handling

Abstract: The tunnel boring machine (TBM), developed within the past few decades, is designed to make the process of tunnel excavation safer and more economical. The use of TBMs in civil and mining construction projects is controlled by several factors including economic considerations and schedule deadlines. Hence, improved methods for estimating TBM performance are important for future projects. This paper presents a new model based on the group method of data handling (GMDH) for predicting the penetration rate (PR) of a TBM. In order to achieve this aim, after investigation of the most effective parameters of PR, rock quality designation, uniaxial compressive strength, rock mass rating, Brazilian tensile strength, weathering zone, thrust force per cutter and revolutions per minute were selected and measured to estimate TBM PR. A database composed of 209 datasets was prepared according to the mentioned model inputs and output. Then, based on the most influential factors of GMDH, a series of parametric investigations were carried out on the established database. In the following, five different datasets with different sets of training and testing were selected and used to construct GMDH models. Aside from that, five multiple regression (MR) models/equations were also proposed to predict TBM PR for comparison purposes. After that, a ranking system was used in order to evaluate the obtained results. As a result, performance prediction results of [i.e. coefficient of determination (R2) = 0.946 and 0.924, root mean square error (RMSE) = 0.141 and 0.169 for training and testing datasets, respectively] demonstrated a high accuracy level of GMDH model in estimating TBM PR. Although both methods are applicable for estimation of PR, GMDH is able to provide a higher degree of accuracy and can be introduced as a new model in this field.

A neuro-genetic predictive model to approximate overbreak induced by drilling and blasting operation in tunnels

Abstract: Overbreak in tunnel construction creates additional costs, and it could put the safety conditions at potential risk. This paper is aimed to predict overbreak in order to control it before drilling and blasting operations through two intelligence systems, namely, an artificial neural network (ANN) and a hybrid genetic algorithm (GA)-ANN. To achieve this aim, a database comprising of 406 datasets were prepared in the Gardaneh Rokh tunnel, Iran. In these datasets, rock mass rating (RMR), spacing, burden, special drilling, number of delays, powder factor and advance length were considered as inputs while overbreak is set as output system. Many intelligence models were created to achieve higher levels of accuracy in accordance with several performance indices, i.e., root mean square error (RMSE), variance account for (VAF) and coefficient of determination (R2). After selection of the best models, GA-ANN model results (VAF = 90.134 and 88.030, R2 = 0.903 and 0.881 and RMSE = 0.058 and 0.074 for training and testing, respectively) were better compared to ANN model results (VAF = 70.319 and 68.731, R2 = 0.703 and 0.693 and RMSE = 0.103 and 0.108 for training and testing, respectively). As a result, the GA-ANN predictive approach can be used for overbreak prediction with high performance capacity. Moreover, results of sensitivity analysis showed that overbreak is mainly influenced by the RMR parameter compared to other inputs.

Application of deep neural networks in predicting the penetration rate of tunnel boring machines

Abstract: Performance prediction in mechanized tunnel projects utilizing a tunnel boring machine (TBM) is a prerequisite to accurate and reliable cost estimation and project scheduling. A wide variety of artificial intelligence methods have been utilized in the prediction of the penetration rate of TBMs. This study focuses on developing a model based on deep neural networks (DNNs), which is an advanced version of artificial neural networks (ANNs), for prediction of the TBM penetration rate based on the data obtained from the Pahang–Selangor raw water transfer tunnel in Malaysia. To evaluate and document the success and reliability of the new DNN model, an ANN model based on five different data categories from the established database was developed and compared with the DNN model. Based on the results obtained of the coefficient of determination and root mean square error (RMSE), a significant increase in the performance prediction of the penetration rate is achieved by developing a DNN predictive model. The DNN model demonstrated better performance for penetration rate estimation compared with the ANN model and it can be introduced as a newly developed model in the field of TBM performance assessment.

Prioritization of effective factors in the occurrence of land subsidence and its susceptibility mapping using an SVM model and their different kernel functions

Abstract: This study attempted to map land subsidence susceptibility using a support vector machine (SVM) model and their different kernel functions in Kerman province, Iran. Initially, land subsidence locations were recognized using extensive field surveys and Google Earth images and, subsequently, a land subsidence distribution map was created in a GIS environment. Then, different effective factors in the occurrence of land subsidence in the study area including percentage slope, slope aspect, altitude, profile curvature, plan curvature, topographic wetness index (TWI), distance from river, lithological units, piezometric changes, land use and normalized difference vegetation index (NDVI) were selected as independent variables for the modeling process. Land subsidence susceptibility maps in the study area were produced using an SVM model and different kernel functions related to it such as linear, polynomial, sigmoid and radial basis functions. The results of model validation using 30% of the unused locations in the modeling process and receiver operating characteristic (ROC) showed that the maps of land subsidence susceptibility obtained from the SVM technique and kernel functions had the highest accuracy with AUC values of 0.894 to 0.857. According to the results of prioritization of effective factors, piezometric data (utilization of groundwater), NDVI and altitude were the most significant factors affecting the occurrence of land subsidence in Kerman province. Therefore, the results of spatial modeling of land subsidence and their susceptibility maps have a key role in the planning of land allocation and water resource management in the study area.

Risk assessment of water inrush in karst tunnels excavation based on normal cloud model

Abstract: Water inrush in karst tunnels is a dynamic process in which internal and external factors are involved. The evaluation of this process is fuzzy, complex, and uncertain. In the current research, few articles give full consideration to the fuzziness and randomness of the water inrush evaluation with useful dynamic feedback. A new assessment method has been proposed for the water inrush evaluation based on a combination of the weighting method and normal cloud model. Specifically, an evaluation index system is forged and each index is quantitatively classified into four grades. A synthetic weighted algorithm combining the analytic hierarchy process, entropy method, and statistical methods is proposed to assign the index weight rationally. Based on the cloud generator algorithm, three numerical characteristics are calculated and a sufficient number of cloud droplets are generated. The membership degree of each index belonging to each grade is constructed and the integrated certain grades are determined. In this paper, the multi-factor normal cloud assessment method is applied to the risk assessment of the Qiyueshan tunnel. The assessment result of the risk grade is accurate, that is, the water inrush risk of different samples at the same risk grade can be reflected in figures. The results not only show high consistency with other assessment methods but are also in good agreement with the excavation results. The proposed cloud model method demonstrates good practical reference for risk assessment of tunnel construction in karst areas and can be applied to tunneling, mining, and other engineering practices in the future.

Effects of thermal treatment on physico-morphological properties of Indian fine-grained sandstone

Abstract: The physical, mechanical and morphological properties of a rock undergo substantial change when exposed to the extreme temperatures that are encountered in processes such as nuclear waste disposal, underground coal gasification (UCG) and building fires. An attempt has been made in this article to study the different physical and morphological changes that occur within Indian sandstone due to thermal treatment. Tests were performed on a thermally treated air-cooled and non-cooled set of samples in order to observe the change in the physico-morphological properties. Heating has a profound effect on the physical properties such as density, porosity and compression wave velocity (VP), which have been further explained by thin-section, X-ray diffraction (XRD) and scanning electron microscope (SEM) studies. Thermal analyses such as thermogravimetric analysis (TGA) and differential thermal analysis (DTA) were performed to observe the chemical changes occurring in the specimen. Since Dholpur sandstone is a quartz-dominant rock, the thermally induced chemical degradation is minimal in nature. DTA studies revealed the quartz inversion to occur at 579.19 °C. Structural changes that are caused due to the random alignment and the thermal anisotropic behaviour of different minerals lead to microcracking, thereby affecting the physical properties. This study will provide an understanding of the thermal behaviour of rocks and the relationship of the thermal behaviour with physico-mechanical behaviour. The study can prove useful while designing structures in processes such as UCG, nuclear waste disposal, deep mining and geothermal energy; the study can also enable the formation of a protocol to restore the structural integrity and aesthetic value of fire-damaged buildings.

Study on the failure mechanism and stability control measures in a large-cutting-height coal mining face with a deep-buried seam

Abstract: Coal face spalling is a major issue affecting the safety of a large-cutting-height mining face, especially in deep mining. In order to analyze failure mechanisms and propose corresponding stability control measures in a large-cutting-height longwall face, panel 1303, with a mining depth of 860 m, which is arranged and advanced distances of 300 m and over 1000 m along the dip and strike directions of a coal seam, respectively, was selected as the engineering background. In addition to uniaxial compressive strength (UCS) tests, triaxial compression tests under different confining pressures and loading methods were carried out to investigate the deformation characteristics of the coal specimens. A mechanical model, the “coal face support roof”, was established to illustrate the factors affecting the stability of the coal face. Combined with numerical simulation, the dominant factor was obtained, and the stress distribution around the coal face at different advance distances was revealed. Based on the coal face failure mechanism, the pertinent in situ measures of “manila + grouting” reinforcement technology for controlling coal face spalling were proposed. The results showed that the coal face spalling depended mainly on vertical cyclic loading and horizontal unloading in both initial and periodic weighting. In terms of deep mining, the surrounding stress distribution played a vital role in coal face failure and instability. Specifically, two dimensions of loading conditions were found in the front 3 m of the coal face, and the principal stress σxx of the coal body was significantly less than the other two principal stresses in the front 8 m of the coal face. In addition, the horizontal principal stress σyy was greater than the vertical principal stress σzz. Therefore, the horizontal principal stress and strength of the coal body were the prominent influencing factors in the large-cutting-height coal face. The mining height and support system working resistance were also of great importance with respect to the stability of the coal face to some degree. Lastly, “manila + grouting” reinforcement technology proposed in this study resulted in 70–80% reduced potential for the occurrence of coal face spalling and in the degree of failure of the coal face, as well as grouting cost could be saved of 30–40% compared with pure grouting measures.

Attribute recognition model for risk assessment of water inrush

Abstract: An attribute recognition model of water inrush risk evaluation is established based on attribute mathematic theory and software is developed for risk assessment in a tunnel. In our model, the entropy weight method is applied to analyze the weights of evaluation indexes. Considering karst hydrologic and engineering geological conditions of a tunnel under construction, eight major influencing factors of water inrush (formation lithology, unfavorable geology, groundwater level, attitude of rocks, contact zone of dissolvable and insoluble rocks, layer and interlayer fissures, catchment ability and surrounding rock mass classification) are selected as the evaluation indexes, and an index system of water inrush risk assessment is constituted. The tunnel is divided into 26 sections, and 340 evaluation objects are selected from these 26 sections in order to construct a judgment matrix. The water inrush risk of the whole tunnel is evaluated by using the proposed software. The results indicate that the attribute recognition model of water inrush risk evaluation is scientific and reasonable and that the software is convenient for use in calculations and is easy to master.

A novel hybrid model of Bagging-based Naïve Bayes Trees for landslide susceptibility assessment

Abstract: Landslide susceptibility assessment was performed using the novel hybrid model Bagging-based Naive Bayes Trees (BAGNBT) at Mu Cang Chai district, located in northern Viet Nam. The model was validated using the Chi-square test, statistical indexes, and area under the receiver operating characteristic curve (AUC). In addition, other models, namely the Rotation Forest-based Naive Bayes Trees (RFNBT), single Naive Bayes Trees (NBT), and Support Vector Machines (SVM), were selected for the comparison. Results show that the novel hybrid model (AUC = 0.834) outperformed the RFNBT (0.830), SVM (0.805), and NBT (0.800). This indicates that the BAGNBT is a promising and better alternative method for landslide susceptibility modeling and mapping.

Deformation failure characteristics of coal–rock combined body under uniaxial compression: experimental and numerical investigations

Abstract: The deformation and failure behaviour of coal–rock combined body under uniaxial compression were investigated experimentally and numerically. The mechanical parameters, including the uniaxial compressive strength (UCS), elastic modulus and full-scale stress–strain curves, were obtained. A detailed analysis of the evolution of the internal cracks based on X-ray computed tomography (CT) observations and acoustic emission (AE) locations is presented. The experimental results show that the mechanical properties and deformation failure characteristics of the coal–rock combined body were governed mainly by the coal. The UCS and elastic modulus of the coal–rock combined body were slightly larger than those of the coal and most of the cracks occurring in the coal were a result of the uniaxial compression. Furthermore, a numerical simulation was conducted to validate the experimental evidence. Finally, based on this understanding, a constitutive relationship was proposed using the natural strain described in Hooke’s law for accurate modelling of the deformation of the coal–rock body. A good agreement was obtained between the numerical results and experimental data during the pre-peak regime.

The stability and roof-support optimization of roadways passing through unfavorable geological bodies using advanced detection and monitoring methods, among others, in the Sanmenxia Bauxite Mine in China’s Henan Province

Abstract: The safety and stability of roadways is greatly influenced by the complex geological conditions present in the Sanmenxia Bauxite Mine, Henan Province, China. In this study, based on data from field survey, advanced detection methods, numerical studies, and monitoring studies, we have adopted the method of steel fiber-reinforced shotcrete to improve the excavation rate of roadways and guarantee the safety of the tunnel when it passes through unfavorable geological bodies, such as shale rocks and broken argillaceous limestone. Field surveys showed that the stability of roof rocks is the major problem faced by engineers; however, tunnel construction using cast-in-situ concrete, which is the method currently applied, costs too much time, resulting in an excavation rate that is too slow to meet the requirements of the Sanmenxia Bauxite Mine. Here, we propose an optimized scheme which, when combined with numerical simulations and data from advanced detection techniques and field monitoring surveys, can improve the efficiency of roadway roof support. During the implementation of the new scheme, the geological anomalies ahead of the working face were detected in advance. It is assumed that the supporting effect of the steel fiber-reinforced shotcrete is equivalent to that of the cast-in-situ concrete as long as a certain thickness is reached. Moreover, the steel fiber-reinforced shotcrete has better mechanical properties than cast-in-situ concrete and achieves a better combination effect with surrounding rock masses. Based on geological conditions and numerical results, the shotcrete should be thickest in the middle area along the roadway axis passing through the unfavorable geological bodies, and gradually become less thick from the middle to both ends. Field tests were carried out to verify the effectiveness of the scheme. The monitoring results show that the roadway passing through broken argillaceous limestone was stable after being supported by shotcrete (at least 80 mm); its thickness should reach at least about 120 mm when passing through shale rock mass. The results indicate that the use of steel fiber-reinforced shotcrete can considerably shorten the construction time compared with cast-in-situ concrete support. The scheme has proved to be a feasible, economical, and time-saving method for underground excavation in the Sanmenxia Bauxite Mine.

A comparative assessment of information value, frequency ratio and analytical hierarchy process models for landslide susceptibility mapping of a Himalayan watershed, India

Abstract: This research work presents a comparative performance of geographic information system (GIS)-based statistical models for landslide susceptibility mapping (LSM) of the Himalayan watershed in India. A total of 190 landslide locations covering an area of 14.63 km2 were identified in the watershed, using high-resolution linear imaging self-scanning (LISS IV) data. The causative factors used for LSM of the study area are slope, aspect, lithology, curvature, lineament density, land cover and drainage buffer. The spatial database has been prepared using remote sensing data along with ancillary data like geological maps. LSMs were prepared using information value (InV), frequency ratio (FR) and analytical hierarchy process (AHP) models. The validation results using the prediction rate curve technique show 89.61%, 87.12% and 88.26% area under curve values for FR, AHP and InV models, respectively. Therefore, the frequency ratio (FR) model could be used for LSM in other parts of this hilly terrain.

A review on land subsidence caused by groundwater withdrawal in Xi’an, China

Abstract: This paper presents a review on the land subsidence caused by groundwater withdrawal in Xi’an, China. With the increasing demands of people’s livelihood and economy during the urbanization process, Xi’an has suffered severe hazards due to land subsidence caused by the excessive exploitation of groundwater since the 1960s. According to past records, the development of land subsidence in Xi’an can be divided into three stages; i) preliminary stage (1959 to 1971), ii) rapid development stage (1972 to 1990), and iii) slow development stage (1991 to present). In the 1990s, the annual groundwater withdrawal volume reached the maximum value of about 1388 × 106 m3/year, and the annual land subsidence also reached the maximum value (about 130 mm/year). The policy for controlling groundwater withdrawal was announced by the Xi’an Municipal Government in 1996, and then the land subsidence rate showed a significant descending tendency. Many researchers developed a series of approaches to yield the prediction of land subsidence caused by groundwater withdrawal, which can be divided into three categories: i) mathematical statistics approaches; ii) numerical approaches; iii) artificial intelligence approaches. Not only are the approaches’ advantages and disadvantages analyzed in this paper, but three emerging investigations on land subsidence in Xi’an are also discussed. The three emerging investigations aim to: i) analyze the relationship between land subsidence and ground fissures, ii) search the monitoring techniques available for obtaining more accurate data, and iii) investigate the effect of sand particle crushing under high stress level on the development of land subsidence.

Landslide susceptibility mapping at Ovacık-Karabük (Turkey) using different artificial neural network models: comparison of training algorithms

Abstract: This study aims to investigate the performances of different training algorithms used for an artificial neural network (ANN) method to produce landslide susceptibility maps. For this purpose, Ovacik region (southeast of Karabuk Province), located in the Western Black Sea Region (Turkey), was selected as the study area. A total of 196 landslides were mapped, and a landslide database was prepared. Topographical elevation, slope angle, aspect, wetness index, lithology, and vegetation index parameters were taken into account for the landslide susceptibility analyses. Two different ANN structures, which were composed of single and double hidden layers, were applied to compare the effects of the ANN. Four different training algorithms, namely batch back-propagation, quick propagation, conjugate gradient descent (CGD), and Levenberg–Marquardt, were used for the training stage of the ANN models. Thus, eight different landslide susceptibility maps were produced for the study area using different ANN structures and algorithms. In order to assess the effects and spatial performances of the considered training algorithms on the ANN models, the relative operating characteristics (ROC) and relation value (rij) approaches were used. The susceptibility map produced by CGD1 has the highest AUC (0.817) and rij values (0.972). Comparison of the susceptibility maps indicated that CGD training algorithm is the slowest one among the other algorithms, but this algorithm showed the highest performance on the results.

Numerical investigation on a grouting mechanism with slurry-rock coupling and shear displacement in a single rough fracture

Abstract: A slurry diffusion model of a single random-roughness fissure is established to consider the slurry and geological fracture coupling based on the Navier-Stokes equation. The slurry flow characteristics and coupling response in a rough fracture are investigated. The influencing mechanism of roughness on the slurry flow was revealed. The calculation model of effective aperture is determined in a rough fracture. The shear displacement effects on the slurry flow are studied. The results show that this slurry diffusion model can more accurately reflect the grouting. A rougher fracture has a great effect on the flow because of the larger low-speed domains. The pressure gradient and maximum diffusion velocity increase parabolically with changes in the relative roughness. The conventional flat-panel model can cause an increasing deviation rate for determination of grouting parameters. The coupling degree distribution is temporal with spatial variations, and increases with the time-dependent viscosity, roughness (decreasing effective aperture), and a shortening path. The viscosity is the key controlling factor in grouting pressure. The roughness response after shear displacement is more significant, further revealing that on-site grout splitting often occurs at the narrow undulating tip with high viscosity and shear on a rough flow surface. The rough fracture model considering fluid-solid coupling is more consistent with the grouting phenomenon in engineering. And the roughness and shear are the key geological factors of grout splitting.

V shaped gully method for controlling rockfall on high-steep slopes in China

Abstract: Large, deep open-pit mining projects lead to rockfall, which becomes an engineering hazard second only to slope stability. The V shaped gully (V-gully) method introduced here was shown to have a significant effect on preventing rockfall through reduction of kinetic energy and rainfall discharge. The method, where a gravel cushion was laid on the V-gully surface, was applied in a typical open-pit mine. A drain dyke should be set at the bottom of the V-gully to discharge rainfall effectively. The slope stability analysis under different drainage rates was calculated by Modified Sarma (MSARMA) software. Slope stability, engineering activities and weathering condition of slope were taken into account to identify the rockfall prone areas. RocFall numerical software was used to calculate the effect of various angle V-gullies on the movements of various rockfall. A striking reduction in kinetic energy of rockfall was achieved, and the effects of V-gully protective measures were studied. Then, a system for controlling rockfall effectively and economically was designed for a particular slope according to the field situation. The effectiveness and reasonableness of design were evaluated through numerical simulations and field testing. This paper provides a theoretical and practical basis for a novel rockfall prevention technique on the high-steep slopes in similar engineering projects.

Adaptive neuro-fuzzy inference system (ANFIS) modeling for landslide susceptibility assessment in a Mediterranean hilly area

Abstract: In this paper, an adaptive neuro-fuzzy modeling (ANFIS) is applied in order to map landslide susceptibility for a Mediterranean catchment (Peloponnese, Greece). The relationship between landslides and factors influencing their occurrence is investigated in GIS environment. Seven conditioning factors, including elevation, slope angle, profile curvature, stream density, distance to main roads, geology, and vegetation were considered in the analysis. Six ANFIS models with different membership functions were developed to generate the corresponding landslide susceptibility maps. The outputs, representing the probability level of landslide occurrence, were grouped into five classes. They were then evaluated using an independent dataset of landslide events in two different validation methods: receiver operating characteristics (ROC) analysis and success and prediction rates. The majority of the calculated area under the curve values for the two validation methods was in the range 0.70–0.90 indicating between fair and very good prediction accuracy for the six models. These values also showed that the prediction accuracy depends on the membership functions examined in the ANFIS modeling. Among these functions, the difference of two sigmoidally shaped (Dsigmf) and product of two sigmoidally shaped (Psigmf) presented the highest prediction accuracy.

Micro/macro physical and mechanical variation of red sandstone subjected to cyclic heating and cooling: an experimental study

Abstract: Since rock mass in many fields of rock engineering usually undergoes a cyclic heating and cooling process, it is very meaningful to investigate the variations of rock micro/macro physical and mechanical properties subjected to cyclic heating and cooling. However, due to the complex and invisible characteristics of rock microstructure, the effect of cyclic heating and cooling on rock macro/micro physical and mechanical properties still requires further investigation. In this study, to explore the microscopic mechanism underlying the variations of rock macroscopic properties during cyclic heating and cooling, uniaxial compression tests and P-wave velocity tests were conducted to obtain the macromechanical properties of red sandstone specimens subjected to varying numbers of heating and cooling cycles. Acoustic emission (AE) tests were also carried out to capture the variations in the microscopic damage process during each heating and cooling cycle. A scanning electron microscope and differential thermal analysis–thermal gravimetric analyzer were used to analyze the development of the microcracks and variations in the micrograin mass after each cyclic heating and cooling. The test results demonstrate that after being heated to 500 °C, some of the minerals (microcline, albite and calcite) in the sandstone decomposed and microcrocks developed due to the uneven thermal stress. When the sandstone was cooled in water, more microdefects were induced to the microstructure due to mineral transformations and uneven contractions, causing further deterioration to the integrity and compactness of the rock matrix. This phenomenon was also reflected in the uniaxial compression tests and AE tests. Due to the damage caused by each heating and cooling cycle, the AE hit rate decreased and the macrophysical and mechanical properties of rock deteriorated as the number of cycles increases. Comparison of the test results obtained from samples subjected to different heating and cooling cycles reveals that although each cycle could cause damage to the rock sample, the first and fifth cycles induced more severe damage, as indicated by the sharp decrease in the physical and mechanical properties of the rock after the first and fifth cycles.

Geographical information systems (GIS) and Multi-Criteria Decision Analysis (MCDA) integration for sustainable landfill site selection considering dynamic data source

Abstract: Landfill site selection is complex task in which many different factors need to be considered. The aim of this study is to identify suitable landfill sites for a 35-year period using a Geographical Information Systems (GIS) with Multi-Criteria Decision Analysis (MCDA). If an analysis is carried out to resolve a problem as it exists today, then the same problem will recur in the years to come due to changes in the input parameters. Future prospections should always be considered when using GIS with MCDA for the creation of mid- and/or long-term solutions. To assess the proposed approach, the city of Antalya was chosen for study, being the city with the highest population growth in Turkey. Twelve available parameters (i.e., digital elevation model, aspect, slope, temperature, precipitation, earthquake zones, distance to road, visibility from roads, distance to population density, geology, landslide density, and distance to fault lines), prepared through a series of GIS analyses, were used to carry out a landfill site selection. The weights of the parameters were obtained from a constructed Analytical Hierarchy Process (AHP) matrix, and the consistency index and the consistency ratio were recorded as 0.091 and 0.062, respectively. For the study, protection zones were omitted. The analysis revealed a number of potential landfill sites. Furthermore, the volume of solid waste for the next 35-year period was calculated using dynamic population data, and possible candidate sites for landfill were generated. The proposed approach can serve as a guide for future works.

An artificial neural network based model to predict spatial soil type distribution using piezocone penetration test data (CPTu)

Abstract: Soil types mapping and the spatial variation of soil classes are essential concerns in both geotechnical and geoenvironmental engineering. Because conventional soil mapping systems are time-consumi ...

Supporting mechanism and mechanical behavior of a double primary support method for tunnels in broken phyllite under high geo-stress: a case study

Abstract: Large squeezing deformation has always been a critical concern in the construction of deep-buried tunnels in soft-weak rock masses. This paper describes a case study on the large deformation mechanism and supporting method of the Maoxian tunnel in Sichuan Province, China, which is located in the core area influenced by the 2008 Wenchuan earthquake and suffered severe large deformation in broken phyllite under high geo-stress. Through a survey on the geological features, the deformation mechanism of surrounding rock and the failure characteristics of supporting structures of the Maoxian tunnel in F1 fault zone were studied. It was found that the occurrence of large deformation was due to the combined action of the high geo-stress and poor self-stability of carbonaceous phyllite. In order to control the squeezing deformation, single and double primary support methods were adopted in succession. A comparative field test was conducted to study their supporting mechanism and mechanical behavior in terms of surrounding rock pressure, internal stress of the steel arch, and axial force and bending moment of the secondary lining. The results revealed that the single primary support method cannot ensure the long-term safety of the tunnel, since many cracks in concrete occurred after about 180 days. The double primary support method, however, was able to control the large deformation and rheological effects of broken phyllite under high geo-stress effectively.

Automatic calculation of rainfall thresholds for landslide occurrence in Chukha Dzongkhag, Bhutan

Abstract: Bhutan is highly prone to landslides, particularly during the monsoon season. Several landslides often occur along the Phuentsholing–Thimphu highway, a very important infrastructure for the country. Worldwide, empirical rainfall thresholds represent a widely used tool to predict the occurrence of rainfall-induced landslides. Nevertheless, no thresholds are currently designed and proposed for any region in Bhutan. In this work, we define empirical cumulated event rainfall–rainfall duration thresholds for the possible initiation of landslides using information on 269 landslides that occurred between 1998 and 2015 along the 90-km highway stretch between the towns of Phuentsholing and Chukha, in southwestern Bhutan, and daily rainfall measurements obtained from three rain gauges. For this purpose, we apply a consolidated frequentist method and use an automatic tool that identifies the rainfall conditions responsible for the failures and calculates thresholds at different exceedance probabilities and the uncertainties associated with them. Analyzing rainfall and landslide data, we exclude from the analysis all the landslides for which the rainfall trigger is not certain, so we reduce the number of landslides from 269 to 43. The calculated thresholds are useful to identify the triggering conditions of rainfall-induced landslides and to predict the occurrence of the failures in the study area, which is, to date, poorly studied. These rainfall thresholds might be implemented in an early warning system, in order to reduce the risk posed by these phenomena to the population living and traveling along the investigated road stretch.

Influence of the spatial variability of the root cohesion on a slope-scale stability model: a case study of residual soil slope in Thailand

Abstract: The bioengineering method using vegetation is an ecological approach for slope stabilisation. However, due to a large variability of vegetation root patterns, a precise quantification of root reinforcement is relatively difficult, leading to a reluctance to use such a technique in practice. This paper presents a probabilistic framework for slope stability analysis considering the spatial variability of root reinforcement. A residual soil slope under a heavy rainfall event was used to model the seepage and stability analysis. The effect of root reinforcement was considered through an additional soil shear strength or root cohesion. Typical characteristics of the root reinforcement of vetiver grass (Chrysopogon zizanioides) in Thailand were assumed in the analysis. A probabilistic analysis was performed considering both stationary and non-stationary random fields of root cohesion. The results indicated that the failure of the vegetated slope could occur when the variance coefficient of the root cohesion was more than a critical value (a critical cov = 0.45 for the uniformly distributed root cohesion case and a critical cov = 0.32 for the case of linear decrease of root cohesion in this particular slope). In practice, the efficiency of the bioengineering method can be improved by controlling the variation of root cohesion within such limits.

Failure characteristics of rock-like materials with single flaws under uniaxial compression

Abstract: A uniaxial compression test was conducted with a servo loading apparatus to study the failure of a rock-like specimen with a pre-existing single flaw. The evolution of cracks was monitored with digital image correlation technology and simulated with the expanded distinct element method based on the strain strength criterion. The concentration and evolution of the principal strain field were found to be consistent with the initiation, propagation, and coalescence of cracks. As the inclination angle increased, the position of the maximum principal strain concentration changed from within the flaw to the flaw tips, and the distribution of the horizontal displacement field changed from symmetric to antisymmetric. The initiation stress and peak strength were affected by the inclination angle; they were minimum when the inclination angle was 60°. As the inclination angle increased, the failure mode of the specimens transformed from mostly tensile failure to mostly shear failure.

Enhancing the accuracy of rainfall-induced landslide prediction along mountain roads with a GIS-based random forest classifier

Abstract: Along mountain roads, rainfall-triggered landslides are typical disasters that cause significant human casualties. Thus, to establish effective mitigation measures, it would be very useful were government agencies and practicing land-use planners to have the capability to make an accurate landslide evaluation. Here, we propose a machine learning methodology for the spatial prediction of rainfall-induced landslides along mountain roads which is based on a random forest classifier (RFC) and a GIS-based dataset. The RFC is used as a supervised learning technique to generalize the classification boundary that separates the input information of ten landslide conditioning factors (slope, aspect, relief amplitude, toposhape, topographic wetness index, distance to roads, distance to rivers, lithology, distance to faults, and rainfall) into two distinctive class labels: ‘landslide’ and ‘non-landslide’. Experimental results with a cross validation process and sensitivity analysis on the RFC model parameters reveal that the proposed model achieves a superior prediction accuracy with an area under the curve of 0.92. The RFC significantly outperforms other benchmarking methods, including discriminant analysis, logistic regression, artificial neural networks, relevance vector machines, and support vector machines. Based on our experimental outcome and comparative analysis, we strongly recommend the RFC as a very capable tool for spatial modeling of rainfall-induced landslides.

Consistency analysis of Hoek–Brown and equivalent Mohr–coulomb parameters in calculating slope safety factor

Abstract: In slope stability analysis, the Mohr–Coulomb (MC) criterion is usually adopted, however, it has some limitations with regard to the rock mass description. Then the Hoek–Brown (HB) criterion was created and then widely used, and the equivalent Mohr–Coulomb (MC) parameters were proposed. However, the consistency of the HB and equivalent MC parameters in calculating slope safety factor remains to be further explored and confirmed. In this study, the gravity increase method and HB parameters were combined to analyze slope stability, and the calculation results of safety factor were compared with those of equivalent MC parameters in the same conditions. This approach was conducted to explore the parameter equivalence by analyzing the consistency of the two safety factors. Then, the HB parameters and slope height (H), which can cause the safety factor distinction (∆Fs), were studied, and the expression of ∆Fs to influencing factors was tested to examine their relationship. The results indicate the following: (1) The HB and equivalent MC parameters are not completely consistent in calculating the slope safety factor, unless the safety factor is relatively small. (2) The sensitivity of the HB parameters to ∆Fs is GSI > D > mi >σci. Moreover, when H is 20 m or 30 m, each HB parameter is linearly related with ∆Fs, in which GSI, mi, and σci are positively correlated with ∆Fs, while D is negatively correlated. (3) The H is found to be in an inversely proportional relationship with ∆Fs. The ∆Fs decreased in a slow downward trend as the H increased. Simultaneously, the linear relationship between HB parameters and ∆Fs is gradually destroyed by the increase of H.

The use of tunnelling parameters and spoil characteristics to assess soil types: a case study from alluvial deposits at a pipejacking project site

Abstract: Generally, when there are only a few boreholes present along a tunnel design alignment, geological understanding of the worksite may not be adequate and the ability to optimise the tunnelling parameters is limited. This lack of boreholes will cause an increased potential of geo-hazards during tunnelling works. This study proposes an alternative method to determine the major and other components of ground under such circumstances. Five factors, cutter wheel torque, sieve residue, flow rate of feedline, pressure in the feed and discharge lines and density of bentonite slurry, are adopted for determining the major and other ground components. Comparisons of the soil types based upon the results of grading and Atterberg limits tests on the spoil and soil samples, respectively, and those resulting from the proposed method indicate good consistency. The proposed method provides an opportunity for establishing a more comprehensive geological structure for refining the tunnelling parameters, reducing the potential of geo-hazards associated with the inappropriate tunnelling parameters.