The main objective of the present study was to produce a novel ensemble data mining technique that involves an adaptive neuro-fuzzy inference system (ANFIS) optimized by Shuffled Frog Leaping ...Algorithm (SFLA) and Particle Swarm Optimization (PSO) for spatial modeling of landslide susceptibility. Step-wise Assessment Ratio Analysis (SWARA) was utilized for the evaluation of the relation between landslides and landslide-related factors providing ANFIS with the necessary weighting values. The developed methods were applied in Langao County, Shaanxi Province, China. Eighteen factors were selected based on the experience gained from studying landslide phenomena, the local geo-environmental conditions as well as the availability of data, namely; elevation, slope aspect, slope angle, profile curvature, plan curvature, sediment transport index, stream power index, topographic wetness index, land use, normalized difference vegetation index, rainfall, lithology, distance to faults, fault density, distance to roads, road density, distance to rivers and river density. A total of 288 landslides were identified after analyzing previous technical surveys, airborne imagery and conducting field surveys. Also, 288 non-landslide areas were identified with the usage of Google Earth imagery and the analysis of a digital elevation model. The two datasets were merged and later divided into two subsets, training and testing, based on a random selection scheme. The produced landslide susceptibility maps were evaluated by the receiving operating characteristic and the area under the success and predictive rate curves (AUC). The results showed that AUC based on the training and testing dataset was similar and equal to 0.89. However, the processing time during the training and implementation phase was considerable different. SWARA-ANFIS-PSO appeared six times faster in respect to the processing time achieved by SWARA-ANFIS-SFLA. The proposed novel approach, which combines expert knowledge, neuro-fuzzy inference systems and evolutionary algorithms, can be applied for land use planning and spatial modeling of landslide susceptibility.
Display omitted
•Novel ensemble data mining technique for landslide susceptibility assessments•Step-wise Assessment Ratio Analysis for weighting landslide related variables•Adaptive neuro-fuzzy inference system (ANFIS) optimized by evolutionary algorithms•Integrate Shuffled Frog Leaping Algorithm, Particle Swarm Optimization and ANFIS
This paper proposes a process monitoring and fault diagnosis method based on a regular vine (R vine) and Bayesian network. The R vine model structure is determined by searching for the maximum sum of ...combinations of correlations among variables, which makes the model more robust and able to describe data more flexibly. A double-space strategy based on the R vine is used to detect the process fault, which can improve the ability to detect weak faults. Furthermore, a Bayesian network is built according to the first tree of the R vine model to diagnose the detected fault and find the root cause. The causality between the nodes of the Bayesian network is determined via the Granger test. The effectiveness of the proposed method is verified by numerical examples and industrial examples.
Taibai County is a mountainous area in China, where rainfall-induced landslides occur frequently. The purpose of this study is to assess landslide susceptibility using the integrated Random Forest ...(RF) with bivariate Statistical Index (SI), the Certainty Factor (CF), and Index of Entropy (IOE). For this purpose, a total of 212 landslides for the study area were identified and collected. Of these landslides, 70% (148) were selected randomly for building the models and the other landslides (64) were used for validating the models. Accordingly, 12 landslide conditioning factors were considered that involve altitude, slope angle, plan curvature, profile curvature, slope aspect, distance to roads, distance to faults, distance to rivers, rainfall, NDVI, land use, and lithology. Then, the spatial correlation between conditioning factors and landslides was analysed using the RF method to quantify the predictive ability of these factors. In the next step, three landslide models, the RF-SI, RF-CF and RF-IOE, were constructed using the training dataset. Finally, the receiver operating characteristic (ROC) and statistical measures such as the kappa index, positive predictive rates, negative predictive rates, sensitivity, specificity, and accuracy were employed to validate and compare the predictive capability of the three models. Of the models, the RF-CF model has the highest positive predictive rate, specificity, accuracy, kappa index and AUC values of 0.838, 0.824, 0.865, 0.730 and 0.925 for the training data, and the highest positive predictive rate, negative predictive rate, sensitivity, specificity, accuracy, kappa index and AUC values of 0.896, 0.934, 0.938, 0.891, 0.914, 0.828, and 0.946 for the validation data, respectively. In general, the RF-CF model produced an optimized balance in terms of AUC values and statistical measures.
•A novel hybrid integration model of RF with bivariate statistical (SI, CF, and IOE) methods•RF model was used to select conditioning factors.•The ROC curve, kappa index, PPR, NPR, sensitivity, specificity, and ACC were used to assess the models.•RF-CF model shows better results than RF-SI and RF-IOE models.
When adsorbing gas is injected into coal, the gas fills in the fractures quickly and a pressure difference between matrix and fractures is created. Because of this difference, there is a pressure ...gradient within the matrix. The gradient evolves with time from the initial equilibrium (zero gradient) to the final equilibrium (zero gradient), so does the adsorption–swelling induced matrix deformation. Previous studies have not taken this effect into consideration. In this study, we hypothesize that the pressure gradient affects the expansion of the gas-invaded area/volume with the matrix and the propagation of the expansion front creates a non-uniform deformation within the matrix. Under this hypothesis, a relation between coal permeability and the expansion of the gas-invaded area with the matrix can be established. When the gas-invaded area is localized in the vicinity of the fracture wall, the expansion of the matrix within this area narrows the fracture opening. We define this as local swelling/shrinking. This local swelling/shrinking is controlled primarily by the coal internal structure. When the gas-invaded area is further spread over the matrix, the expansion of the whole matrix may narrow or widen the fracture opening depending on the external boundary conditions. This global swelling/shrinking is controlled primarily by the external boundary conditions. These conceptual understandings are defined through strain rate-based coal permeability models for both the matrix and the fractures. This strain rate based time-dependent permeability model was verified against experimental observations, that couples coal deformation, the gas flow in the matrix system and gas flow in the fracture system.
Unconventional natural gas, including coalbed methane and shale gas, has become important natural gas resources. Coal and shale reservoirs are characterised by low porosity and low permeability and ...difficult for gas production. These reservoirs are also considered as fractured reservoirs, i.e. the natural fracture/cleat system in coals and bedding direction microfractures in shales. Permeabilities of these reservoirs are sensitive to stress change. During gas production, the pressure drawdown significantly increases effective stress, and thus decreases the absolute permeability. The relationship between permeability and stress is characterised by fracture compressibility, which is difficult and costly to be obtained from the field, but can be acquired easily from laboratory measurement. In this review article, the laboratory methods to obtain fracture compressibility were reviewed. Literature data on fracture compressibility for coals and shales were collated and the relationships between fracture compressibility and pressure, stress and rock properties were discussed. It is found that fracture compressibility is higher for coals than for shales, and the fracture compressibility for proppant supported fracture is even lower than that for the same shale or coal. Moreover, fracture compressibility is variable depending on gas type, gas pressure, and stress. Fracture compressibility has no correlation with absolute permeability in general, but has a weak positive correlation for the same sample.
•Fracture compressibility is found higher for coals than for shales in general.•Proppant supported fracture compressibility is typically smaller than that for natural fractures or original samples.•Fracture compressibility is variable and dependent on gas type, gas pressure, and stress.•Fracture compressibility has no correlation with absolute permeability.
Rockbursts occurred frequently during the excavation of several parallel tunnels in the Jinping II hydropower station under a maximum overburden of 2525m over an average length of 17.5km. In order to ...investigate the nucleation and evolution mechanism of rockbursts, a comprehensive monitoring campaign consisting of a digital borehole camera, cross-hole acoustic apparatus, and sliding micrometer was undertaken for in situ measurements in two specially excavated test tunnels B and F. This paper presents the comprehensive monitoring methods applied, and results of numerical analysis applied to a typical rockburst that fortuitously occurred during the testing period. Precursory characteristics preceding rockbursts are: (a) abundant crack initiation, propagation and coalescence, (b) deformation of surrounding rock mass involving an accelerated deformation stage, quiescence stage and reaccelerated deformation stage, and (c) decrease of the characteristic elastic wave velocity of the rock mass. The nucleation and evolution of rockbursts discussed consist of four stages: a) stress adjustment, b) energy accumulation, c) crack initiation, propagation and coalescence, and d) fractured rock collapse and ejection. The results provide a direct case history to assist the prediction and support of rockburst disasters, and contribute to field excavation of deeply buried tunnels.
► Field direct precursors of rockburst were investigated in special deep test tunnels. ► Digital borehole camera, sliding micrometer and acoustic apparatus were adopted. ► Abundant crack initiation, propagation and coalescence before rockburst were observed. ► Variable deformation stages and decrease of elastic velocity of rock mass were found. ► Changes of stress and energy by unloading effect lead to rock fracture and ejection.
The Bayu Tunnel is the most problematic tunnel project in the Lhasa-Nyingchi section of the Sichuan-Tibet Railway. Rockbursts have occurred frequently during the construction process, which has ...seriously threatened the safety of construction workers and equipment. Aiming at the rockburst hazard prediction problem of the Bayu Tunnel, in this paper an integrating method of combination weighting and matter-element extension theory is proposed. This method introduces the rock brittleness evaluation index BICSS, Russenes criterion and rock mass integrity coefficient as the main rockburst evaluation indexes, which characterized the control factors of rock physical and mechanical property, surrounding rock stress and rock mass integrity on rockburst tendency, respectively. Then the rockburst tendency evaluation index system is constructed. The weight of each rockburst control factor is determined by the combination weighting algorithm of triangular fuzzy number analytic hierarchy process (TFN-AHP), criteria importance though intercriteria correlation (CRITIC), and determination of comprehensive weight (MDI). The correlation degree between the rockburst sample and rockburst grade can be calculated using the matter-element extension theory, after which the prediction of rockburst grade is carried out. Finally, the applicability, superiority and the necessity of selected rockburst control factors in this method are verified and discussed further. The research results provide a new analysis method and technical means for rockburst tendency assessment of deep tunnel.
•A novel integrating method for rockburst tendency prediction is proposed•Matter-element extension model and combination weighting method are integrated•The rock brittleness evaluation index BICSS is involved in rockburst tendency prediction model•Rock properties, complex geological structures and geostress are considered
Rockburst is a type of dynamic failure and often causes considerable damage during the construction of underground engineering. Geological structures have been regarded as a crucial factor ...influencing rockburst distribution, and more related research is urgently needed. This paper presents microseismic monitoring information, characteristics of 151 rockbursts, and various structural planes collected in detail from a deep tunnel located in Southwest China. The results show that microseismic activity tends to increase when joints with large scale begin to appear and vanish. Rockburst risk always increases between the spandrels and around the intersection of two joint sets. The joint dip angle changes the rockburst risk. In the presence of one or two joint sets with a 70– 80° dip angle, rockbursts are more prone to occur. The conclusions can be interpreted with the discrete element method. When more joints are encountered, the total number of rockbursts may decrease, but the rockbursts are more likely to be strong or induce collapse, causing more damage. This research can improve the understanding of rockburst mechanisms and guide the choice of construction and support methods.
•Abundant rockburst cases and joint information are analyzed statistically.•Rockbursts are more prone to occur when the joint sets have a 70–80° dip angle.•The rockburst is more likely to be strong or induce collapse with more joints.
Traversing the Qinghai–Tibetan Plateau, the Sichuan–Tibet Railway is by far the most difficult railway project in the world. The Qinghai–Tibetan Plateau features the most active crustal dynamics on ...earth, the strongest coupling effects of endogenic and exogenic dynamics, and the environment most sensitive to global climate change. The project area is characterized by extremely cold climate, high elevation and relief, high seismic intensity, high geothermal activity, and high tectonic stress. Consequently, the threat of various disaster risks is ever-present at different stages of the entire life cycle of the Sichuan–Tibet Railway. There is urgent need to systematically study these problems at various levels from the fundamental science to the development of key technologies. This article investigates the different disaster risks recognized during the various stages of construction of the Sichuan–Tibet Railway project, and summarizes the scientific challenges and technical problems faced in relation to disaster risk prevention and control. This work also introduces the scientific deployment and relevant research progress of the Sichuan–Tibet Fund special project initiated by the National Natural Science Foundation of China. Here, we also aim to solve the major fundamental scientific challenges in terms of long-term risk prevention and control during the construction of the Sichuan–Tibet Railway, and lay a theoretical foundation to promote breakthroughs in the bottleneck of key technologies. The scientific challenges addressed in the study of disaster risk associated with the Sichuan–Tibet Railway include the following: The quantitative assessment of the activity of deep-large faults and strong earthquake prediction, the evolution of physical fields in areas of strong tectonic activity, the development mechanisms of tunnel hazards, the slope evolution processes under coupled endogenic and exogenic dynamics in alpine gorges, the impact of climate change on the formation and evolution of surface hazards, and the evolution of extreme wind fields in deep-cut canyons. The technical problems faced in disaster risk prevention and mitigation in relation to the Sichuan–Tibet Railway are as follows: Advanced identification, monitoring, and early warning of geological disasters in mountainous areas with steep and complex terrain; risk analysis, prevention, and control of railway engineering disasters based on their dynamic processes; tunnel engineering hazard monitoring, early warning, risk analysis, prevention, and control technologies; key technologies for emergency response; and the green and resilient railway system and lifecycle risk management. The Sichuan–Tibet Fund special project will include five key research topics: (1) the interior geological structure and dynamic evolution of the eastern plateau; (2) the hazard-inducing mechanisms of coupled internal and external forces in canyons and gullies within plateaus; (3) the cataclysm mechanics of deeply buried long-distance tunnel engineering; (4) risk identification and projection of major disasters affecting the railway; and (5) the integrated management of scientific innovations and super large-scale railway construction. Systematic research is expected to reveal the evolution of earth surface movements and coupled engineering-disturbance related disasters. It will also enable the formation of a comprehensive risk analysis method for major engineering disasters, and promote the development of green, safe, efficient, and resilient engineering disaster risk reduction technologies that will support the disaster risk management during the entire lifecycle of the Sichuan–Tibet Railway.
•Investigated the natural and anthropic disaster risks in different construction stages of the Sichuan–Tibet Railway.•Identified the scientific challenges and technical gaps faced in disaster risk reduction of the Sichuan–Tibet Railway.•Proposed fundamental research and technologies development directions for a safe and resilient Sichuan–Tibet Railway.