## 2018 Vol. 26, No. 5

2018, 26(5): 1113-1120.

In loess area, both artificial rainfall field test and in-situ observation under natural rainfall conditions suggest that the influence of surface water is limited to the top several meters, below which the variation of soil water content is little. Therefore, it is a controversy to fact that the surface water infiltration can induce loess landslides with deep seated failure surfaces. In this study, an in-situ observation test is undertaken in a thick loess layer for a period of one year from June 1

^{st}, 2015 to May 31^{st}, 2016. Results show that water infiltration in the loess can be divided into three zones, namely, active, stable, and saturated zones. The active zone is influenced significantly by precipitation and evaporation events, where water flow is a transient behavior. The stable zone is characterized with a stable water content and steady seepage flow. The saturated zone is where groundwater exits, which can be drained in the forms of spring. In the long run, the hydraulic condition reaches an equilibrium state with a stable water content in the stable zone and stable ground water table level. However, this equilibrium state can be disturbed with a higher surface water infiltration, for example irrigation water and leakage from pipelines, which contributes to a new equilibrium state with higher stable water content in stable zone and higher GWT level. Loess landslides with deep seated failure surfaces may occur during the transitional process from the initial equilibrium state to the new equilibrium state. Yanlian deep loess landslide is taken as an example to illustrate the mechanism of surface water infiltration induced deep landslides. The hydraulic response to long-term surface drop water and its influence to slope stability are analyzed. Results highlight the close relationship between deep loess landslides and long-term surface water infiltration, which is easy to be neglected due to its slow flow rate. In loess area, both artificial rainfall field test and in-situ observation under natural rainfall conditions suggest that the influence of surface water is limited to the top several meters, below which the variation of soil water content is little. Therefore, it is a controversy to fact that the surface water infiltration can induce loess landslides with deep seated failure surfaces. In this study, an in-situ observation test is undertaken in a thick loess layer for a period of one year from June 1^{st}, 2015 to May 31^{st}, 2016. Results show that water infiltration in the loess can be divided into three zones, namely, active, stable, and saturated zones. The active zone is influenced significantly by precipitation and evaporation events, where water flow is a transient behavior. The stable zone is characterized with a stable water content and steady seepage flow. The saturated zone is where groundwater exits, which can be drained in the forms of spring. In the long run, the hydraulic condition reaches an equilibrium state with a stable water content in the stable zone and stable ground water table level. However, this equilibrium state can be disturbed with a higher surface water infiltration, for example irrigation water and leakage from pipelines, which contributes to a new equilibrium state with higher stable water content in stable zone and higher GWT level. Loess landslides with deep seated failure surfaces may occur during the transitional process from the initial equilibrium state to the new equilibrium state. Yanlian deep loess landslide is taken as an example to illustrate the mechanism of surface water infiltration induced deep landslides. The hydraulic response to long-term surface drop water and its influence to slope stability are analyzed. Results highlight the close relationship between deep loess landslides and long-term surface water infiltration, which is easy to be neglected due to its slow flow rate.
2018, 26(5): 1121-1130.

Geological hazards lead to great threatens in China. The vulnerability of geo-hazards bearing capacity directly determines the severe of geological disasters. Its corresponding quantitative assessment of vulnerability is very importance. This paper presents the index system of the vulnerability for geo-hazards bearing capacity. The system includes four indexes and nineteen sub-indexes. The four indexes are life index, material index, ecological environment index and social economy index. A case study is performed in Qingchuan County, China. The township units and administrative village units are used as the basic unit in GIS. Both TOPSIS model and Weighted Sum model are adopted to evaluate the vulnerability of geo-hazards bearing capacity in this area. Five vulnerability levels are classified and include high vulnerability level, relatively high vulnerability level, moderate vulnerability level, relatively low vulnerability level, and low vulnerability level. Comparative study of these two models indicates that the results derived from the two models are consistent as a whole. 73 percent of the area has the same vulnerability level. The rest 27 percent area differs with a gap of one vulnerability level. In the areas of high, low vulnerability levels derived by the two models are quite consistent with each other. The agreements between two models in relatively high and medium level are 86% and 73%respectively, while in relatively low vulnerability level was slightly worse(54%). An evaluated comparison of township unit and administrative village unit presents the good consistency at the whole. Approximately above half level of vulnerability of two units are exact the same or general same. The evaluation results of administrative villages are apparently more accurate. The reason is that there are obvious differences among various administrative villages in their index values, while the township standard eliminates the differences among administrative villages.
Geological hazards lead to great threatens in China. The vulnerability of geo-hazards bearing capacity directly determines the severe of geological disasters. Its corresponding quantitative assessment of vulnerability is very importance. This paper presents the index system of the vulnerability for geo-hazards bearing capacity. The system includes four indexes and nineteen sub-indexes. The four indexes are life index, material index, ecological environment index and social economy index. A case study is performed in Qingchuan County, China. The township units and administrative village units are used as the basic unit in GIS. Both TOPSIS model and Weighted Sum model are adopted to evaluate the vulnerability of geo-hazards bearing capacity in this area. Five vulnerability levels are classified and include high vulnerability level, relatively high vulnerability level, moderate vulnerability level, relatively low vulnerability level, and low vulnerability level. Comparative study of these two models indicates that the results derived from the two models are consistent as a whole. 73 percent of the area has the same vulnerability level. The rest 27 percent area differs with a gap of one vulnerability level. In the areas of high, low vulnerability levels derived by the two models are quite consistent with each other. The agreements between two models in relatively high and medium level are 86% and 73%respectively, while in relatively low vulnerability level was slightly worse(54%). An evaluated comparison of township unit and administrative village unit presents the good consistency at the whole. Approximately above half level of vulnerability of two units are exact the same or general same. The evaluation results of administrative villages are apparently more accurate. The reason is that there are obvious differences among various administrative villages in their index values, while the township standard eliminates the differences among administrative villages.

2018, 26(5): 1131-1138.

This paper studies the effect of long-term dry and wet circulation on the strength of ground fissure zone and its nearby loess. It takes the typical loess samples of ground fracture site in Xi'an as the research object. It studies the shear strength parameters and the strength attenuation characteristics of the loess specimens with different moisture content under dry and wet circulation with laboratory test method(mainly for direct shear test). It puts forward a ground fissured loess strength attenuation calculation method. Research results show the follows. (1)Shear strength of ground fissured loess is decreased gradually with the cycle number and water content. One dry-wet cycle or greater than 18%moisture content have affected on the shear strength parameters significantly. (2)The intensity attenuation of the 1 drying and wetting cycles is most obvious. It increases with the increase of dry and wet cycles. (3)Under the action of 1 dry and wet cycle, the attenuation of its strength is most obvious with the moisture content. It gradually decreases with the increase of water content. Its strength attenuates higher when the water content is below 12%, and the strength attenuation is smaller when the water content is greater than 12%.
This paper studies the effect of long-term dry and wet circulation on the strength of ground fissure zone and its nearby loess. It takes the typical loess samples of ground fracture site in Xi'an as the research object. It studies the shear strength parameters and the strength attenuation characteristics of the loess specimens with different moisture content under dry and wet circulation with laboratory test method(mainly for direct shear test). It puts forward a ground fissured loess strength attenuation calculation method. Research results show the follows. (1)Shear strength of ground fissured loess is decreased gradually with the cycle number and water content. One dry-wet cycle or greater than 18%moisture content have affected on the shear strength parameters significantly. (2)The intensity attenuation of the 1 drying and wetting cycles is most obvious. It increases with the increase of dry and wet cycles. (3)Under the action of 1 dry and wet cycle, the attenuation of its strength is most obvious with the moisture content. It gradually decreases with the increase of water content. Its strength attenuates higher when the water content is below 12%, and the strength attenuation is smaller when the water content is greater than 12%.

2018, 26(5): 1139-1154.

In mountainous areas, casualties caused by earthquake-induced landslides often account for more than half of the total earthquakes. Especially in the loess area, due to the far-distance loess landslide caused by the earthquake, the disastrousness is even more serious. This paper attempts to solve the problems of earthquake-induced loess slope instability such as peak acceleration, critical displacement and moving distance. It uses discrete element(PFC)method, calibrates the stress-strain curve of the indoor three-axis experiment, and transforms high-precision aerial three-dimensional topographic data. It takes the loess slope in Haikou Village, Ningxia Province as the research object to carry out numerical simulation of slope unstable failure and movement process under the three-dimensional earthquake. It monitors the stress components at different locations under the earthquake, compares the

*p*and*q*values of the monitoring points with the*p-q*failure lines obtained from the indoor three-axis experiments. It obtains the stress paths and peak accelerations in the process of slope failure. Combining with the maximum displacement of the monitored particles, the critical maximum displacement of slope failure is obtained. The peak acceleration of the slope failure is 0.135g and the critical displacement is 50 cm. At the same time, according to different ground environmental conditions, the damage range of the slope after sliding failure under different friction coefficients is predicted. The paper provides a new visualization method for the seismic design and earthquake prevention and mitigation of slope engineering in the loess area. In mountainous areas, casualties caused by earthquake-induced landslides often account for more than half of the total earthquakes. Especially in the loess area, due to the far-distance loess landslide caused by the earthquake, the disastrousness is even more serious. This paper attempts to solve the problems of earthquake-induced loess slope instability such as peak acceleration, critical displacement and moving distance. It uses discrete element(PFC)method, calibrates the stress-strain curve of the indoor three-axis experiment, and transforms high-precision aerial three-dimensional topographic data. It takes the loess slope in Haikou Village, Ningxia Province as the research object to carry out numerical simulation of slope unstable failure and movement process under the three-dimensional earthquake. It monitors the stress components at different locations under the earthquake, compares the*p*and*q*values of the monitoring points with the*p-q*failure lines obtained from the indoor three-axis experiments. It obtains the stress paths and peak accelerations in the process of slope failure. Combining with the maximum displacement of the monitored particles, the critical maximum displacement of slope failure is obtained. The peak acceleration of the slope failure is 0.135g and the critical displacement is 50 cm. At the same time, according to different ground environmental conditions, the damage range of the slope after sliding failure under different friction coefficients is predicted. The paper provides a new visualization method for the seismic design and earthquake prevention and mitigation of slope engineering in the loess area.
2018, 26(5): 1162-1169.

On the basis of remote sensing interpretation, drones pictures and geological field survey, we analyzes the correlation between the distribution of slope geological hazards and the slope factors including shape, topographic slope, dip angle, structure characteristics and the lithology in Shanyang-Shangnan Vanadium mining area. The results are as follows. (1)The distribution of geohazards in the mining area has a marked bedding-dipping slope effect. The geological disasters of such slopes are 75.86% of the total. (2)72.42% of the slope hazards are concentrated in the slopes with angles of 30° to 50°. (3)The slope hazards have a corresponding relationship with the stratum inclination. Most geological disasters are concentrated in the range of bedding dip angles from 40°to 60°. (4)The slope hazards tend to occur in the combination of structural planes with the intersection angles within the range of 70°to 90°. The nearly orthogonal combination of structural planes has significant amplification effect of collapses. (5)The study also shows that the slope hazards usually occur in the siliceous rocks or interbedding of dolomite and clay rocks, while collapse occurs in dolomite. At the same time, we propose a remote sensing identification method for slope disasters, combining with the comparative analysis of remote sensing images in the study area. Then, we find a set of early identification method for slope hazards in mining areas. This research can provide technical support for mining enterprises and residents to identify and prevent slope disasters.
On the basis of remote sensing interpretation, drones pictures and geological field survey, we analyzes the correlation between the distribution of slope geological hazards and the slope factors including shape, topographic slope, dip angle, structure characteristics and the lithology in Shanyang-Shangnan Vanadium mining area. The results are as follows. (1)The distribution of geohazards in the mining area has a marked bedding-dipping slope effect. The geological disasters of such slopes are 75.86% of the total. (2)72.42% of the slope hazards are concentrated in the slopes with angles of 30° to 50°. (3)The slope hazards have a corresponding relationship with the stratum inclination. Most geological disasters are concentrated in the range of bedding dip angles from 40°to 60°. (4)The slope hazards tend to occur in the combination of structural planes with the intersection angles within the range of 70°to 90°. The nearly orthogonal combination of structural planes has significant amplification effect of collapses. (5)The study also shows that the slope hazards usually occur in the siliceous rocks or interbedding of dolomite and clay rocks, while collapse occurs in dolomite. At the same time, we propose a remote sensing identification method for slope disasters, combining with the comparative analysis of remote sensing images in the study area. Then, we find a set of early identification method for slope hazards in mining areas. This research can provide technical support for mining enterprises and residents to identify and prevent slope disasters.

2018, 26(5): 1170-1177.

The instability of loess slope is one of the engineering problems that urgently need to be solved in geotechnical engineering. Firstly, the loess slope in Northern Shaanxi is selected as the research object. Four field simulated rainfall experiments under the condition of rain intensity are carried out. The boundary conditions of the slope are improved by digging the isolation groove on both sides of the slope. The changes of the depth of the water immersion and the soil moisture content of the soil under different rainfall intensity conditions are measured. Then, we analyze the rainfall infiltration process and slope stress variation characteristics under different rain intensity conditions, and compare the differences between different infiltration laws under different rain intensity conditions. The test results show that the rule is as follows. The deepest infiltration is at slope toe. The second is at slope crest. The last is at slope middle surface. The infiltration rate is the fastest at slope crest, followed by the slope toe, and the last at the middle slope. The rainfall infiltration capacity gradually weakens as the depth increases. The volumetric water content and earth pressure change amplitude of the measuring point at the same buried depth become larger with the increase of rain intensity. The time is shortened for the abrupt change of water content and soil pressure. The effect of the scour is more obvious. Finally, the seepage analysis based on Geo-studio software verifies the correctness of the field test results and clarifies the influence of rain intensity on rainfall infiltration of loess slope.
The instability of loess slope is one of the engineering problems that urgently need to be solved in geotechnical engineering. Firstly, the loess slope in Northern Shaanxi is selected as the research object. Four field simulated rainfall experiments under the condition of rain intensity are carried out. The boundary conditions of the slope are improved by digging the isolation groove on both sides of the slope. The changes of the depth of the water immersion and the soil moisture content of the soil under different rainfall intensity conditions are measured. Then, we analyze the rainfall infiltration process and slope stress variation characteristics under different rain intensity conditions, and compare the differences between different infiltration laws under different rain intensity conditions. The test results show that the rule is as follows. The deepest infiltration is at slope toe. The second is at slope crest. The last is at slope middle surface. The infiltration rate is the fastest at slope crest, followed by the slope toe, and the last at the middle slope. The rainfall infiltration capacity gradually weakens as the depth increases. The volumetric water content and earth pressure change amplitude of the measuring point at the same buried depth become larger with the increase of rain intensity. The time is shortened for the abrupt change of water content and soil pressure. The effect of the scour is more obvious. Finally, the seepage analysis based on Geo-studio software verifies the correctness of the field test results and clarifies the influence of rain intensity on rainfall infiltration of loess slope.

2018, 26(5): 1178-1187.

How to select the bench width of the stepped slope is still a topic of research. In this paper, centrifuge model tests and FEM-based strength reduction method are carried out to investigate the influence of benches on reinforced soil slopes with steps. The benches can be used to decompose the slope into several secondary slopes(steps). Centrifuge test results show that the damage of the slope is transferred to the steps of the slope, and the damage scale is reduced. Soil reinforcing can enhance the integrity of the slope as well as reduce the interaction between the steps. Benches with appropriate width can make the failure occur in the secondary slopes, which is helpful for the overall safety of the whole slope. The FEM numerical results show that the effect of cohesion can weaken when the slope is high enough. The benches divide the slope into several secondary slopes with lower height. As a result, the cohesion strength of the soil can be fully utilized in the stability of the slope. Since the effect of reinforcement is mainly to increase the apparent cohesion of the soil, the reinforced slope can be more stable with appropriate height of steps. For reinforced high soil slopes, the modulus and extensibility of reinforcement are key factors. The joint reinforcing effect cannot be mobilized if some reinforcements snap very early before others take effect.
How to select the bench width of the stepped slope is still a topic of research. In this paper, centrifuge model tests and FEM-based strength reduction method are carried out to investigate the influence of benches on reinforced soil slopes with steps. The benches can be used to decompose the slope into several secondary slopes(steps). Centrifuge test results show that the damage of the slope is transferred to the steps of the slope, and the damage scale is reduced. Soil reinforcing can enhance the integrity of the slope as well as reduce the interaction between the steps. Benches with appropriate width can make the failure occur in the secondary slopes, which is helpful for the overall safety of the whole slope. The FEM numerical results show that the effect of cohesion can weaken when the slope is high enough. The benches divide the slope into several secondary slopes with lower height. As a result, the cohesion strength of the soil can be fully utilized in the stability of the slope. Since the effect of reinforcement is mainly to increase the apparent cohesion of the soil, the reinforced slope can be more stable with appropriate height of steps. For reinforced high soil slopes, the modulus and extensibility of reinforcement are key factors. The joint reinforcing effect cannot be mobilized if some reinforcements snap very early before others take effect.

2018, 26(5): 1188-1195.

The Haiyuan

*M*_{S}8.5 earthquake on December 16, 1920, occurred at the loess plateau. It triggered a large number of landslides. The landslides directly caused a large number of casualties. In recent years, there have been some special studies on landslides triggered by the Haiyuan earthquake. However, most of these studies have limitations on their local study areas or individual landslides. There are few detailed and comprehensive results about landslides triggered by the earthquake. This situation has become an obstacle to understand the overall incidence and severity, spatial distribution law of the Haiyuan earthquake-triggered landslides. In this study, we carry out the visual interpretation of landslides in the high seismic intensity(Ⅸ-Ⅺ)areas of Haiyuan earthquake based on Google Earth platform, as well as analyze the relationship between landslides and several influence factors. Results show that the earthquake triggered at least 5384 landslides, which have a total landslide area of 218.78 km^{2}. The highest density of landslides is located at the NW part of seismic intensity Ⅸ circle. Through analyzing the relationship between these landslides and topography, earthquake and geological factors, it is found that elevation 1700~2000 m registers the largest landslide number and density. Most landslides are concentrated in the slope angle range of 15°~25°whereas the number density of landslides increases with the increase of slope angle. The lower the slope position is, i.e., the closer to rivers, the greater the density of the landslides. The areas with Cenozoic strata underlying, especially the Quaternary loess covered areas, register most of the landslides and the high density of landslides. This study provides a scientific reference for exploring the law of the occurrence of seismic landslides and reduction and mitigation of earthquake-triggered landslides in loess area. The Haiyuan*M*_{S}8.5 earthquake on December 16, 1920, occurred at the loess plateau. It triggered a large number of landslides. The landslides directly caused a large number of casualties. In recent years, there have been some special studies on landslides triggered by the Haiyuan earthquake. However, most of these studies have limitations on their local study areas or individual landslides. There are few detailed and comprehensive results about landslides triggered by the earthquake. This situation has become an obstacle to understand the overall incidence and severity, spatial distribution law of the Haiyuan earthquake-triggered landslides. In this study, we carry out the visual interpretation of landslides in the high seismic intensity(Ⅸ-Ⅺ)areas of Haiyuan earthquake based on Google Earth platform, as well as analyze the relationship between landslides and several influence factors. Results show that the earthquake triggered at least 5384 landslides, which have a total landslide area of 218.78 km^{2}. The highest density of landslides is located at the NW part of seismic intensity Ⅸ circle. Through analyzing the relationship between these landslides and topography, earthquake and geological factors, it is found that elevation 1700~2000 m registers the largest landslide number and density. Most landslides are concentrated in the slope angle range of 15°~25°whereas the number density of landslides increases with the increase of slope angle. The lower the slope position is, i.e., the closer to rivers, the greater the density of the landslides. The areas with Cenozoic strata underlying, especially the Quaternary loess covered areas, register most of the landslides and the high density of landslides. This study provides a scientific reference for exploring the law of the occurrence of seismic landslides and reduction and mitigation of earthquake-triggered landslides in loess area.
2018, 26(5): 1196-1202.

The calculation result of stability of anisotropic rock slope is directly dependent on the value of shear strength parameters of the rock mass. In order to take the influence of rock anisotropy on slope stability into account, we use CSMR instead of RMR in the H-B strength criterion, and modify the parameters

*m*_{b}and*s*. Then, the equivalent M-C strength parameters of anisotropic rock mass can be obtained. They are combined with the calculation of the slope engineering at the exit of the diversion tunnel of a hydropower station in Jinsha River. The safety factors of the slope stability before and after the correction of the strength parameters under the same bedrock conditions and different excavation design schemes are calculated and compared. The results indicate that the stability of the slope is greatly affected by the anisotropy of rock mass. Without considering the anisotropy of rock mass, the stability of slope cannot be accurately reflected in the calculation result of slope stability. After anisotropy correction, the calculation results of slope safety factors of the two design schemes are 1.11 and 1.70 respectively, which are basically consistent with the qualitative and semi-quantitative evaluation. In summary, the method for correcting the parameters of anisotropic rock mass of slope is feasible. The calculation result of stability of anisotropic rock slope is directly dependent on the value of shear strength parameters of the rock mass. In order to take the influence of rock anisotropy on slope stability into account, we use CSMR instead of RMR in the H-B strength criterion, and modify the parameters*m*_{b}and*s*. Then, the equivalent M-C strength parameters of anisotropic rock mass can be obtained. They are combined with the calculation of the slope engineering at the exit of the diversion tunnel of a hydropower station in Jinsha River. The safety factors of the slope stability before and after the correction of the strength parameters under the same bedrock conditions and different excavation design schemes are calculated and compared. The results indicate that the stability of the slope is greatly affected by the anisotropy of rock mass. Without considering the anisotropy of rock mass, the stability of slope cannot be accurately reflected in the calculation result of slope stability. After anisotropy correction, the calculation results of slope safety factors of the two design schemes are 1.11 and 1.70 respectively, which are basically consistent with the qualitative and semi-quantitative evaluation. In summary, the method for correcting the parameters of anisotropic rock mass of slope is feasible.
2018, 26(5): 1203-1210.

This paper takes the Kunming road underground utility tunnel crossing through f3 ground fissure in Xi'an city, China as research object. The deformation and force characteristics of underground segmented utility tunnel is analysed with finite element numerical simulationmethod. The results show that the vertical settlement deformation at the roof of underground utility tunnel exhibits reverse-S shape as a whole, and increases with the settlement of the hanging wall of ground fissure increasing. The longitudinal deformation of underground utility tunnel can be divided into three deformation sections:warping deformation section at the hanging wall of ground fissure, differential settlement section, and integral deformation section at the footwall of ground fissure. Under the action of ground fissure with a vertical dislocation 50 cm in the design life of 100 years, the horizontal displacement peak value of utility tunnel roof appears in the position of ground fissure zone. The maximum values are 4.1 cm on roof and 3.2 cm on the floor of utility tunnel. The joints between utility segmented tunnels is easy to pull apart and be damaged and should be reinforced. Nearby the ground fissure zone, the contact pressure between the utility tunnel floor and soil stratum decreases to 0 at the hanging wall with the increase of vertical dislocation of ground fissure and underground utility tunnel floor appears disengaging phenomenon. Grouting holes should be reserved convenient to grouting treatment, But the contact pressure between underground utility tunnel roof and soil stratum has an increase trend. The strain of the underground utility tunnel roof and floor exceeds the limit of the concrete strain when settlement of the hanging wall of ground fissure reaches 20 cm. Therefore, the deformation and failure mode of the underground utility tunnel is mainly tensile failure under the action of ground fissure. The research can provide a scientific basis for engineering design of underground utility tunnel crossing ground fissure zone in Xi'an city and other developed areas of ground fissure.
This paper takes the Kunming road underground utility tunnel crossing through f3 ground fissure in Xi'an city, China as research object. The deformation and force characteristics of underground segmented utility tunnel is analysed with finite element numerical simulationmethod. The results show that the vertical settlement deformation at the roof of underground utility tunnel exhibits reverse-S shape as a whole, and increases with the settlement of the hanging wall of ground fissure increasing. The longitudinal deformation of underground utility tunnel can be divided into three deformation sections:warping deformation section at the hanging wall of ground fissure, differential settlement section, and integral deformation section at the footwall of ground fissure. Under the action of ground fissure with a vertical dislocation 50 cm in the design life of 100 years, the horizontal displacement peak value of utility tunnel roof appears in the position of ground fissure zone. The maximum values are 4.1 cm on roof and 3.2 cm on the floor of utility tunnel. The joints between utility segmented tunnels is easy to pull apart and be damaged and should be reinforced. Nearby the ground fissure zone, the contact pressure between the utility tunnel floor and soil stratum decreases to 0 at the hanging wall with the increase of vertical dislocation of ground fissure and underground utility tunnel floor appears disengaging phenomenon. Grouting holes should be reserved convenient to grouting treatment, But the contact pressure between underground utility tunnel roof and soil stratum has an increase trend. The strain of the underground utility tunnel roof and floor exceeds the limit of the concrete strain when settlement of the hanging wall of ground fissure reaches 20 cm. Therefore, the deformation and failure mode of the underground utility tunnel is mainly tensile failure under the action of ground fissure. The research can provide a scientific basis for engineering design of underground utility tunnel crossing ground fissure zone in Xi'an city and other developed areas of ground fissure.

2018, 26(5): 1211-1218.

As an objective system, the evaluation model of weights of evidence can calculate the weight values of different factors, assign the weight process objectively, and distinguish the impact of different factors on the sensitivity of collapse. In order to understand the sensitivity of different factors to collapse, this paper uses the information of the collapse disasters that have already occurred and used the example of the Yan Duhe Town in Badong County, Hubei Province as an example. 106 collapse points are selected, and the weight method of evidence is adopted. The sensitivity analysis of six basic factors is carried out by ArcGIS software. The six factors include slope, slope direction, distance to fault, distance to water system, distance to highway, and formation lithology. 78 randomly selected collapse points are used as a training set for modeling, and the other 28 are used as a test set to test the evaluation result. The analysis results show that the sensitivity of the distance from the collapse point to the highway is the largest. The area with high sensitivity to collapse is linearly distributed near the road and the river. The extremely sensitive area and the high sensitivity area account for 15.27% of the whole research area. There are 22 collapse points in the test set distributed in the area. This is to say that 78.58% of the collapse points are correctly evaluated, which is consistent with the on-site investigation results. The collapse sensitivity map obtained has a guiding role in disaster prevention and reduction work.
As an objective system, the evaluation model of weights of evidence can calculate the weight values of different factors, assign the weight process objectively, and distinguish the impact of different factors on the sensitivity of collapse. In order to understand the sensitivity of different factors to collapse, this paper uses the information of the collapse disasters that have already occurred and used the example of the Yan Duhe Town in Badong County, Hubei Province as an example. 106 collapse points are selected, and the weight method of evidence is adopted. The sensitivity analysis of six basic factors is carried out by ArcGIS software. The six factors include slope, slope direction, distance to fault, distance to water system, distance to highway, and formation lithology. 78 randomly selected collapse points are used as a training set for modeling, and the other 28 are used as a test set to test the evaluation result. The analysis results show that the sensitivity of the distance from the collapse point to the highway is the largest. The area with high sensitivity to collapse is linearly distributed near the road and the river. The extremely sensitive area and the high sensitivity area account for 15.27% of the whole research area. There are 22 collapse points in the test set distributed in the area. This is to say that 78.58% of the collapse points are correctly evaluated, which is consistent with the on-site investigation results. The collapse sensitivity map obtained has a guiding role in disaster prevention and reduction work.

2018, 26(5): 1227-1236.

Geo-materials usually possess the characteristics of limit strain, which can be obtained by laboratory or numerical method. Shear failure is one of the main failure mode of the materials in slope, thus the limit strain of slope material can be used as a failure criteria. In fact, during the failure process of slope, the strength parameters of geo-materials in slope do not degrade simultaneously. Instead, they usually progressively damage until global failure. In this study, regarding slope failure we proposed a dynamic strength reduction method based on the limit shear strain criterion. In this method, we only reduce the strength parameters of elements whose shear strain is beyond the limit value. The reduction process will continue until the global slope failure associated with the formation of whole sliding surface whose limit shear strains beyond the limit value. A numerical modeling of an actual slope using this method is conducted. Results show that the simulated slope failure pattern and deformation are in good agreement with the in-situ observed failure phenomenon and monitoring data, respectively. This suggests that the proposed numerical method has a promising application prospect in studying the progressive failure and stability of slopes.
Geo-materials usually possess the characteristics of limit strain, which can be obtained by laboratory or numerical method. Shear failure is one of the main failure mode of the materials in slope, thus the limit strain of slope material can be used as a failure criteria. In fact, during the failure process of slope, the strength parameters of geo-materials in slope do not degrade simultaneously. Instead, they usually progressively damage until global failure. In this study, regarding slope failure we proposed a dynamic strength reduction method based on the limit shear strain criterion. In this method, we only reduce the strength parameters of elements whose shear strain is beyond the limit value. The reduction process will continue until the global slope failure associated with the formation of whole sliding surface whose limit shear strains beyond the limit value. A numerical modeling of an actual slope using this method is conducted. Results show that the simulated slope failure pattern and deformation are in good agreement with the in-situ observed failure phenomenon and monitoring data, respectively. This suggests that the proposed numerical method has a promising application prospect in studying the progressive failure and stability of slopes.

2018, 26(5): 1237-1242.

The performance of the bentonite buffer in nuclear waste repository concept relies to a great extent on the buffer surrounding the canister having sufficient dry density. Loss of buffer material caused by erosion remains as the most significant process reducing the density of the buffer. Yield stress of bentonite colloids is a key parameter to express the erosion process of bentonite aqueous solution. A model for the yield stress of aggregates is presented. It incorporates fractal dimension taking into account the solid volume fraction and the aggregate diameter. The model shows that the yield stress(

*σ*_{y}) of aggregates increases with the solid volume fraction (*φ*) as a power law, and is given by_{s}*σ*_{y}=*σ*_{y0}*φ*_{s}^{m}, where the exponent(*m*) is related to fractal dimension(*D*), and*σ*_{y0}is a referenced parameter. The relationship between exponent(*m*) and fractal dimension is validated by published data of aggregates and represents the measured data very well, over a wide range of the solid volume fractions. The referenced parameter(*σ*_{y0}) is calibrated from experiments of yield stress using power law fittings. The agreement between theory and experiments supports the idea that yielding is ultimately caused by the rupture of a few interparticle bonds within aggregates. The performance of the bentonite buffer in nuclear waste repository concept relies to a great extent on the buffer surrounding the canister having sufficient dry density. Loss of buffer material caused by erosion remains as the most significant process reducing the density of the buffer. Yield stress of bentonite colloids is a key parameter to express the erosion process of bentonite aqueous solution. A model for the yield stress of aggregates is presented. It incorporates fractal dimension taking into account the solid volume fraction and the aggregate diameter. The model shows that the yield stress(*σ*_{y}) of aggregates increases with the solid volume fraction (*φ*) as a power law, and is given by_{s}*σ*_{y}=*σ*_{y0}*φ*_{s}^{m}, where the exponent(*m*) is related to fractal dimension(*D*), and*σ*_{y0}is a referenced parameter. The relationship between exponent(*m*) and fractal dimension is validated by published data of aggregates and represents the measured data very well, over a wide range of the solid volume fractions. The referenced parameter(*σ*_{y0}) is calibrated from experiments of yield stress using power law fittings. The agreement between theory and experiments supports the idea that yielding is ultimately caused by the rupture of a few interparticle bonds within aggregates.
2018, 26(5): 1243-1249.

This paper takes the cement clay of different cement contents as the studied object. The samples of cement clay soil are used for nuclear magnetic resonance test to study the effect of cement content on the distribution of micro-pore. The samples are used for the variable water head penetration test to study the effect of cement content on permeability. On this basis, the paper investigates the microscopic mechanism of the effect of cement content on permeability. The study results show the follow. The permeability coefficient of cement-soil decreases as the cement content increases. It decreases intensely at cement content ranging from 4%to 12%, and decreases slowly at cement content ranging from 15%to 25%. There are three peaks in the

*T*_{2}distribution curves of cement-soil, which correspond to small pores, medium pores and large pores, respectively. As the cement content increases, the total area of*T*_{2}distribution curve tends to decrease. At the stage of low cement content, the area of the second and third peaks diminishes obviously. But at the stage of high cement content, the area of the first peak diminishes obviously. Thus it can be seen that in the range of low cement content, the hydration of cement-soil is prior to block up large pores and medium pores, which makes the permeability coefficient decreases significantly with the increase of cement content. While in the range of high cement content, the area of small pores mainly declines with the increment of cement content. But that of large pores and medium pores has little difference, so the increase of cement content has little effect on the decrease of permeability coefficient of cement-soil. This paper takes the cement clay of different cement contents as the studied object. The samples of cement clay soil are used for nuclear magnetic resonance test to study the effect of cement content on the distribution of micro-pore. The samples are used for the variable water head penetration test to study the effect of cement content on permeability. On this basis, the paper investigates the microscopic mechanism of the effect of cement content on permeability. The study results show the follow. The permeability coefficient of cement-soil decreases as the cement content increases. It decreases intensely at cement content ranging from 4%to 12%, and decreases slowly at cement content ranging from 15%to 25%. There are three peaks in the*T*_{2}distribution curves of cement-soil, which correspond to small pores, medium pores and large pores, respectively. As the cement content increases, the total area of*T*_{2}distribution curve tends to decrease. At the stage of low cement content, the area of the second and third peaks diminishes obviously. But at the stage of high cement content, the area of the first peak diminishes obviously. Thus it can be seen that in the range of low cement content, the hydration of cement-soil is prior to block up large pores and medium pores, which makes the permeability coefficient decreases significantly with the increase of cement content. While in the range of high cement content, the area of small pores mainly declines with the increment of cement content. But that of large pores and medium pores has little difference, so the increase of cement content has little effect on the decrease of permeability coefficient of cement-soil.
2018, 26(5): 1250-1256.

As a typical porous medium, loess has its porosity as one of its most notable characteristics. Porosity not only influences the physical and chemical characteristics of loess, but also seriously affects the hydraulic properties of loess, especially the permeability of loess. The capillary model and the classic model of porous media seepage theory have been widely used in the fields of oil and gas field mining, coal-formed gas mining and geotechnical engineering. The porous media seepage theory model is less used in the porous structure of loess. Therefore, this paper takes the permeability of loess as an example. This paper summarizes and analyzes the relatively successful porous media capillary seepage models. It selects the seepage model suitable for quantitative characterization of the permeability of loess. It discusses the capillary seepage model of porous medium. The discussions are combined with the comparison of the results of calculation and the results of the permeation test. The applicability of this method to describe loess permeability is examined. The "sausage" structure of pore(or "lotus root" structure) is proposed. The study provides a new theoretical basis for the study of loess permeability.
As a typical porous medium, loess has its porosity as one of its most notable characteristics. Porosity not only influences the physical and chemical characteristics of loess, but also seriously affects the hydraulic properties of loess, especially the permeability of loess. The capillary model and the classic model of porous media seepage theory have been widely used in the fields of oil and gas field mining, coal-formed gas mining and geotechnical engineering. The porous media seepage theory model is less used in the porous structure of loess. Therefore, this paper takes the permeability of loess as an example. This paper summarizes and analyzes the relatively successful porous media capillary seepage models. It selects the seepage model suitable for quantitative characterization of the permeability of loess. It discusses the capillary seepage model of porous medium. The discussions are combined with the comparison of the results of calculation and the results of the permeation test. The applicability of this method to describe loess permeability is examined. The "sausage" structure of pore(or "lotus root" structure) is proposed. The study provides a new theoretical basis for the study of loess permeability.

2018, 26(5): 1257-1264.

This paper uses similar materials and makes rock specimens with unanchored and/or anchored prefabricated fissures. Bolts and cracks at 45ånd 90åre made into the anchor rods in the rock specimens. It discusses the regulation of propagation rules, anchorage effects and failure modes of rock samples through uniaxial compression tests after 60 cycles of freeze-thaw cycles with strain monitoring. The test results show the follows. (1)Two types of anchorages have equal effect in cracks supporting before 30 cycles of freeze-thaw cycles. The higher angle between bolts and cracks would have the larger power of anchorage to the cracks. (2)There is a regular form including six stages, cold shrinkage, frost heaving, frost heaving stability, melt-thaw, and rebound stability under natural conditions. Two supporting conditions do not show obvious stages of contraction and rebound. The 90°supporting condition is more effective than the 45°supporting condition. (3)The cement slurry is applied to fractured rock mass by hydraulic grout. It can repair the whole anchoring system. (4)The anchor rod enhances the ability of the fractured rock mass to resist the expansion of the fracture, reduces the suddenness of the fractured rock mass in the fractured rock mass. The failure mode is transformed from tensile damage to tensile shear complex damage.
This paper uses similar materials and makes rock specimens with unanchored and/or anchored prefabricated fissures. Bolts and cracks at 45ånd 90åre made into the anchor rods in the rock specimens. It discusses the regulation of propagation rules, anchorage effects and failure modes of rock samples through uniaxial compression tests after 60 cycles of freeze-thaw cycles with strain monitoring. The test results show the follows. (1)Two types of anchorages have equal effect in cracks supporting before 30 cycles of freeze-thaw cycles. The higher angle between bolts and cracks would have the larger power of anchorage to the cracks. (2)There is a regular form including six stages, cold shrinkage, frost heaving, frost heaving stability, melt-thaw, and rebound stability under natural conditions. Two supporting conditions do not show obvious stages of contraction and rebound. The 90°supporting condition is more effective than the 45°supporting condition. (3)The cement slurry is applied to fractured rock mass by hydraulic grout. It can repair the whole anchoring system. (4)The anchor rod enhances the ability of the fractured rock mass to resist the expansion of the fracture, reduces the suddenness of the fractured rock mass in the fractured rock mass. The failure mode is transformed from tensile damage to tensile shear complex damage.

2018, 26(5): 1265-1271.

The moisture content of loess have a significant influence on the strength of the soil. It changes with the cyclic change of moisture content, which leads to a series of diseases. In order to study the diffusion and migration law of water vapor and temperature in unsaturated loess, we analyze the water vapor and temperature diffusion migration in unsaturated remolded loess through the indoor filling model test. We inject the high temperature and high pressure water vapor into the remolded loess. The experimental results show that in unsaturated remolded loess, the range of water vapor diffusion is approximately ellipsoid. The water vapor diffusion rate decreases gradually along the radial direction. The vapor diffusion rate, diffusion range and humidification degree can increase with the vapor pressure increase. Vapor pressure accelerates water diffusion and also the temperature migration. The heating rate of soil is about 0.75 ℃ ·min

^{-1}when the pressure is 50 kPa and 1.12 ℃ ·min^{-1}when the pressure is 200 kPa. The greater the soil particle hinders the water vapor diffusion when the density of the soil is more larger. The hindrance is more obvious and the larger the water content decreases with the increase of the radial diffusion distance of the water vapor. The moisture content of loess have a significant influence on the strength of the soil. It changes with the cyclic change of moisture content, which leads to a series of diseases. In order to study the diffusion and migration law of water vapor and temperature in unsaturated loess, we analyze the water vapor and temperature diffusion migration in unsaturated remolded loess through the indoor filling model test. We inject the high temperature and high pressure water vapor into the remolded loess. The experimental results show that in unsaturated remolded loess, the range of water vapor diffusion is approximately ellipsoid. The water vapor diffusion rate decreases gradually along the radial direction. The vapor diffusion rate, diffusion range and humidification degree can increase with the vapor pressure increase. Vapor pressure accelerates water diffusion and also the temperature migration. The heating rate of soil is about 0.75 ℃ ·min^{-1}when the pressure is 50 kPa and 1.12 ℃ ·min^{-1}when the pressure is 200 kPa. The greater the soil particle hinders the water vapor diffusion when the density of the soil is more larger. The hindrance is more obvious and the larger the water content decreases with the increase of the radial diffusion distance of the water vapor.
2018, 26(5): 1279-1284.

The Smoothed Particle Hydrodynamics(SPH)method has been widely used in the simulation of geohazards with large deformation. However, as a particle method, SPH analysis is usually time consuming for the case with a lot of particles. Thus it is essential to conduct the study on the efficiency improvement of SPH method. Here, we propose a 3D SPH model for the analysis of soil flowing based on the principles of SPH method, non Newtonian fluid theory and the concept of equivalent viscosity. Then, the parallel implementation of this model is conducted using the OpenMP framework. For a model test of flowing granular soil, the 2D and 3D simulations are conducted with the proposed SPH model. The comparisons of flowing distance and impact force indicate that both the boundary effect and viscosity effect in the 3D cases are larger than those in the 2D cases, which can cause the decreasing of flowing distance and impact force. Meanwhile, by comparing the speed up ratios under different number of threads, we found that the efficiency has been greatly improved. The proposed SPH model can achieve a good simulation of flowing soil and make the best use of multi-core technology of modern computers. The paper can provide a potential reference to other numerical methods in engineering geology.
The Smoothed Particle Hydrodynamics(SPH)method has been widely used in the simulation of geohazards with large deformation. However, as a particle method, SPH analysis is usually time consuming for the case with a lot of particles. Thus it is essential to conduct the study on the efficiency improvement of SPH method. Here, we propose a 3D SPH model for the analysis of soil flowing based on the principles of SPH method, non Newtonian fluid theory and the concept of equivalent viscosity. Then, the parallel implementation of this model is conducted using the OpenMP framework. For a model test of flowing granular soil, the 2D and 3D simulations are conducted with the proposed SPH model. The comparisons of flowing distance and impact force indicate that both the boundary effect and viscosity effect in the 3D cases are larger than those in the 2D cases, which can cause the decreasing of flowing distance and impact force. Meanwhile, by comparing the speed up ratios under different number of threads, we found that the efficiency has been greatly improved. The proposed SPH model can achieve a good simulation of flowing soil and make the best use of multi-core technology of modern computers. The paper can provide a potential reference to other numerical methods in engineering geology.

2018, 26(5): 1285-1291.

The typical silt in Wuhan is solidified with Ionic Soil Stabilizer(ISS). The plasticity index of soil with different mix proportions is obtained through the test of Atterberg limits. The compressive properties, microstructure, specific surface area, chemical composition, energy spectrum analysis and cation exchange test are carried out before and after the solidification of silt with ISS. The macroscopic and microscopic mechanism of the solidification of silt with ISS are studied. The test results show that the compressive strength of silt treated with ISS decreases and porosity decreases. Ionic Soil Stabilizer promotes aggregation and condensation of soil particles, aggregates obviously increase, specific surface area decreases, specific surface energy decreases, and soil structure is denser. There is no change in the chemical composition of the silt before and after solidification. Cation exchange reaction occurs between the ISS and the counter ion layer on the surface of the silt colloid particles. The K

^{+}and Na^{+}ions in the diluents of the curing agent are captured and adsorbed by the silt. The Ca^{2+}and Mg^{2+}ions on the surface of clay particles are resolved, which makes the thickness of the diffusion layer thinner, the*ξ*potential decreased, and the clay colloid particles coalesced. ISS improves the strength and stability of the soil particles, so as to achieve the purpose of soil modification and solidification. The typical silt in Wuhan is solidified with Ionic Soil Stabilizer(ISS). The plasticity index of soil with different mix proportions is obtained through the test of Atterberg limits. The compressive properties, microstructure, specific surface area, chemical composition, energy spectrum analysis and cation exchange test are carried out before and after the solidification of silt with ISS. The macroscopic and microscopic mechanism of the solidification of silt with ISS are studied. The test results show that the compressive strength of silt treated with ISS decreases and porosity decreases. Ionic Soil Stabilizer promotes aggregation and condensation of soil particles, aggregates obviously increase, specific surface area decreases, specific surface energy decreases, and soil structure is denser. There is no change in the chemical composition of the silt before and after solidification. Cation exchange reaction occurs between the ISS and the counter ion layer on the surface of the silt colloid particles. The K^{+}and Na^{+}ions in the diluents of the curing agent are captured and adsorbed by the silt. The Ca^{2+}and Mg^{2+}ions on the surface of clay particles are resolved, which makes the thickness of the diffusion layer thinner, the*ξ*potential decreased, and the clay colloid particles coalesced. ISS improves the strength and stability of the soil particles, so as to achieve the purpose of soil modification and solidification.
2018, 26(5): 1292-1299.

Buffer materials filled in the nuclear waste deep geology disposal should not only own to excellent capacity of isolation and adsorption, but also process high thermal conductivity. Hence, choosing natural graphite powders as additives can improve the thermal conductivity of Na-bentonite. These mixtures have both isolated and heat-conducted properties. The graphite powders, whose maximum grain sizes are 50, 100, 200 and 325 mesh, respectively, are added into the bentonite at fixed graphite adding ratio(20%, Wt.). The effect of graphite particle size on the hydro-mechanical behaviors of graphite-bentonite mixtures is studied by conducting some experiments including free swelling ratio, constant volume swelling pressure and permeability. It can be found that the maximum particle size 100 to 200 mesh of graphite would be the best choice to obtain the lowest permeability and highest swelling pressure for buffer material at the same graphite content. It can be ascribed to the contacted mode between graphite and bentonite. The graphite particles are flat, especially for the large size ones, and are easier to form macro-pores at the edge of them, even if after being compacted. As for smaller size graphite particles, there are existing larger contacted areas between graphite and bentonite. Plus the hydrophobicity of graphite, more leakage passage would be formed at the graphite-bentonite interfaces. These conclusions provide some scientific references for preparing buffer material used in the nuclear waste deep geology disposal.
Buffer materials filled in the nuclear waste deep geology disposal should not only own to excellent capacity of isolation and adsorption, but also process high thermal conductivity. Hence, choosing natural graphite powders as additives can improve the thermal conductivity of Na-bentonite. These mixtures have both isolated and heat-conducted properties. The graphite powders, whose maximum grain sizes are 50, 100, 200 and 325 mesh, respectively, are added into the bentonite at fixed graphite adding ratio(20%, Wt.). The effect of graphite particle size on the hydro-mechanical behaviors of graphite-bentonite mixtures is studied by conducting some experiments including free swelling ratio, constant volume swelling pressure and permeability. It can be found that the maximum particle size 100 to 200 mesh of graphite would be the best choice to obtain the lowest permeability and highest swelling pressure for buffer material at the same graphite content. It can be ascribed to the contacted mode between graphite and bentonite. The graphite particles are flat, especially for the large size ones, and are easier to form macro-pores at the edge of them, even if after being compacted. As for smaller size graphite particles, there are existing larger contacted areas between graphite and bentonite. Plus the hydrophobicity of graphite, more leakage passage would be formed at the graphite-bentonite interfaces. These conclusions provide some scientific references for preparing buffer material used in the nuclear waste deep geology disposal.

2018, 26(5): 1300-1310.

The vulnerability of the foundation of a factory building in Guizhou is due to the Alkali pollution, which originates the decrease of void ratio, specific weight, and plasticity index of the land. To study the transformation of shear strength of red clay, the authors design laboratory experiments combined with the pollution of the alkali concentration and the sources of the production process of Alumina. The facts of maintenance time and concentration of alkali liquor are also considered into the experiments. Within the help of technologies like scanning electron microscope(SEM), the results are concluded and supported by the raw data. Starting with the microstructure of the fracture surface of the red clay contaminated by alkali when the shear displacement is 4 mm, the fracture mode of the micro-pore and particle is analyzed qualitatively and quantitatively, and the fracture mode during the shear failure is explored. The achieved results suggest that in high NaOH solution shear strength are increased, most of the pores are inert pores, the direction of the pores is good, and the particles become longer. In the sources of the production process of Alumina, shear strength are the best. All most of the pores are inert pores, direction of the pores is the best, the particles become the longest. In low NaOH solution, shear strength are decreased, there are many macropores in the soil, the direction of pores is poor, and the particles are still in good roundness. It should be paid attention to in the project, which can provide some reference for the improvement of red clay polluted by alkali.
The vulnerability of the foundation of a factory building in Guizhou is due to the Alkali pollution, which originates the decrease of void ratio, specific weight, and plasticity index of the land. To study the transformation of shear strength of red clay, the authors design laboratory experiments combined with the pollution of the alkali concentration and the sources of the production process of Alumina. The facts of maintenance time and concentration of alkali liquor are also considered into the experiments. Within the help of technologies like scanning electron microscope(SEM), the results are concluded and supported by the raw data. Starting with the microstructure of the fracture surface of the red clay contaminated by alkali when the shear displacement is 4 mm, the fracture mode of the micro-pore and particle is analyzed qualitatively and quantitatively, and the fracture mode during the shear failure is explored. The achieved results suggest that in high NaOH solution shear strength are increased, most of the pores are inert pores, the direction of the pores is good, and the particles become longer. In the sources of the production process of Alumina, shear strength are the best. All most of the pores are inert pores, direction of the pores is the best, the particles become the longest. In low NaOH solution, shear strength are decreased, there are many macropores in the soil, the direction of pores is poor, and the particles are still in good roundness. It should be paid attention to in the project, which can provide some reference for the improvement of red clay polluted by alkali.

2018, 26(5): 1311-1317.

Due to the unique structural property of the structural clay of Zhanjiang group, the bearing capacity of pile is complex in the clay. If the traditional pile foundation design calculation theory and method are applied to calculate the bearing capacity of the pile foundation, the deviation between the theoretical calculation value and the measured value of the pile is more or less. Based on direct shear test and unconfined compression strength test of Zhanjiang group structural clay, as well as the shear displacement method, the model of softening of pile-clay interface is put forward. When the pile-soil interface is in the elastic stage, the side friction of pile increases linearly with displacement. In the stage of softening, the side friction of pile decreases linearly with the displacement. The reduction rate is related to the soil sensitivity and normal pressure of the pile-soil interface. In the stage of sliding, the side friction of pile can not change with the displacement. Therefore, the displacement and axial force analysis of single pile under the vertical load in Zhanjiang group structural clay are derived according to the proposed model. Furthermore, the calculation and test results are compared and the feasibility of the proposed method is verified.
Due to the unique structural property of the structural clay of Zhanjiang group, the bearing capacity of pile is complex in the clay. If the traditional pile foundation design calculation theory and method are applied to calculate the bearing capacity of the pile foundation, the deviation between the theoretical calculation value and the measured value of the pile is more or less. Based on direct shear test and unconfined compression strength test of Zhanjiang group structural clay, as well as the shear displacement method, the model of softening of pile-clay interface is put forward. When the pile-soil interface is in the elastic stage, the side friction of pile increases linearly with displacement. In the stage of softening, the side friction of pile decreases linearly with the displacement. The reduction rate is related to the soil sensitivity and normal pressure of the pile-soil interface. In the stage of sliding, the side friction of pile can not change with the displacement. Therefore, the displacement and axial force analysis of single pile under the vertical load in Zhanjiang group structural clay are derived according to the proposed model. Furthermore, the calculation and test results are compared and the feasibility of the proposed method is verified.

2018, 26(5): 1318-1325.

The soft marine clay, deposited under the marine environment in North of Jiangsu plain, is rich of the montmorillonite and saline water. During the post-sedimentary period, the soil suffers the invasion of the surface and underground freshwater, leading to the alteration and reduction of pore water salinity. Presently, the secondary consolidation behavior is generally based on the in-situ soil behavior in a current environment. Hence, this non-consideration on the pore water salinity alteration may result in the insufficient redundancy and potential risk. To clarify the evolution of the soils' secondary consolidation behavior due to the alteration of pore water salinity, the impact of pore water salinity and clay mineral on the soil behavior is necessary to be investigated. To simplify the mineralogy influence, the commercial kaolin and bentonite are selected. The consolidation behavior of artificial clays changing with pore water salinities(NaCl solution) are investigated using oedometer tests. The results showed that the compression index

*C*_{c}and the primary consolidation process of bentonite-kaolinite mixture decrease with water salinity increasing. At a certain applied stress and void ratio, the secondary consolidation coefficient*C*_{α}of montmorillonitic clay also decreases with water salinity increasing. Meanwhile, the ratio of*C*_{α}to*C*_{c}is not a constant and deviated from empirical data. The main mechanism is that the compression of diffused double layer. However, the water salinity does not significantly influence secondary consolidation behavior of kaolin. Therefore, it is important to take pore water salinity into consideration for secondary consolidation prediction of montmorillonite-rich clays. The soft marine clay, deposited under the marine environment in North of Jiangsu plain, is rich of the montmorillonite and saline water. During the post-sedimentary period, the soil suffers the invasion of the surface and underground freshwater, leading to the alteration and reduction of pore water salinity. Presently, the secondary consolidation behavior is generally based on the in-situ soil behavior in a current environment. Hence, this non-consideration on the pore water salinity alteration may result in the insufficient redundancy and potential risk. To clarify the evolution of the soils' secondary consolidation behavior due to the alteration of pore water salinity, the impact of pore water salinity and clay mineral on the soil behavior is necessary to be investigated. To simplify the mineralogy influence, the commercial kaolin and bentonite are selected. The consolidation behavior of artificial clays changing with pore water salinities(NaCl solution) are investigated using oedometer tests. The results showed that the compression index*C*_{c}and the primary consolidation process of bentonite-kaolinite mixture decrease with water salinity increasing. At a certain applied stress and void ratio, the secondary consolidation coefficient*C*_{α}of montmorillonitic clay also decreases with water salinity increasing. Meanwhile, the ratio of*C*_{α}to*C*_{c}is not a constant and deviated from empirical data. The main mechanism is that the compression of diffused double layer. However, the water salinity does not significantly influence secondary consolidation behavior of kaolin. Therefore, it is important to take pore water salinity into consideration for secondary consolidation prediction of montmorillonite-rich clays.
2018, 26(5): 1326-1335.

In order to reveal the influence of single layer thickness on crack propagation path of flaggy rock, we used ABAQUS for numerical simulation. We carried out three-point bending tests of layered rock with specimens' length and width constant(512 mm×128 mm). There were 5 layer thicknesses. They were respectively 64 mm, 32 mm, 16 mm, 8 mm, and 4 mm. Information of acoustic emission and main crack path were extracted. We summarized the regulation of peak load and acoustic emission data. The results of the study show that:(1)The strength of flaggy rock is relevant to single layer's thickness. The larger the monolayer thickness is, the smaller the bending strength of rock mass is, meanwhile this change is nonlinear. (2)Although with different thickness, the bedding surface consistently shows the effect of preventing crack propagating upward. (3)With monolayer's thickness decreasing, the total crack length of rock mass first increases and then decreases. (4)When the cracks expand along the weak surface, the fracture type is a combination of slip and tensile fracture. In contrast, when they expand inside the rock mass, tensile fractures dominate. (5)Acoustic emission event will not occur instantly. In the contrary, they increase sharply just when the peak intensity approaching. Accompanied by the stress plunging, the number of acoustic emission events reaches the maximum. The research results will provide a basis for predicting the fracture process of layered rocks.
In order to reveal the influence of single layer thickness on crack propagation path of flaggy rock, we used ABAQUS for numerical simulation. We carried out three-point bending tests of layered rock with specimens' length and width constant(512 mm×128 mm). There were 5 layer thicknesses. They were respectively 64 mm, 32 mm, 16 mm, 8 mm, and 4 mm. Information of acoustic emission and main crack path were extracted. We summarized the regulation of peak load and acoustic emission data. The results of the study show that:(1)The strength of flaggy rock is relevant to single layer's thickness. The larger the monolayer thickness is, the smaller the bending strength of rock mass is, meanwhile this change is nonlinear. (2)Although with different thickness, the bedding surface consistently shows the effect of preventing crack propagating upward. (3)With monolayer's thickness decreasing, the total crack length of rock mass first increases and then decreases. (4)When the cracks expand along the weak surface, the fracture type is a combination of slip and tensile fracture. In contrast, when they expand inside the rock mass, tensile fractures dominate. (5)Acoustic emission event will not occur instantly. In the contrary, they increase sharply just when the peak intensity approaching. Accompanied by the stress plunging, the number of acoustic emission events reaches the maximum. The research results will provide a basis for predicting the fracture process of layered rocks.

2018, 26(5): 1336-1341.

The roughness of rock structure plane affects its shear strength. The joint roughness coefficient(JRC)has been widely used as a roughness parameter. However, there are sampling precision problems existing in the acquisition of

*JRC*values. In order to obtain reasonable*JRC*values, this paper conducts a study on the precision of*JRC*profile sampling. 10 structure planes of sandstone, diorite and granite with a size of 10 cm ·10 cm are selected as study objects, respectively. A hand-held 3D laser scanner is used to scan the structure plane surface to obtain its geometric features. Digital processing is carried out with the help of 3D morphological image post-processing software. On the basis of this, fractal theory is selected to calculate*JRC*values under different sampling point spacing conditions. Then sampling point interval effects on*JRC*value is analyzed and its influencing factors are studied. Results show that:*JRC*values are affected mainly by the sample point interval on profile line. However, when the sampling interval is less than a certain critical value, the*JRC*basically remains unchanged. The critical interval of sampling point is related to the roughness of structural plane and has a negative exponential relation with the*JRC*values. In addition, the negative exponential relation is influenced by rock type. The smaller the rock particles are, the better the imitative effect is. When the three types of rock are considered comprehensively, a negative exponential function*y*=1.8314*e*^{-0.067x}is obtained between the critical interval of sampling point and*JRC*value. So, to remove the interval effects on the JRC evaluation, the critical interval of sampling points can be quantified according to the function. This study can provide a scientific method for the collection of roughness information of structural plane. And the precision and efficiency of structural plane information collection can be ensured by selecting a reasonable number of sampling points. The roughness of rock structure plane affects its shear strength. The joint roughness coefficient(JRC)has been widely used as a roughness parameter. However, there are sampling precision problems existing in the acquisition of*JRC*values. In order to obtain reasonable*JRC*values, this paper conducts a study on the precision of*JRC*profile sampling. 10 structure planes of sandstone, diorite and granite with a size of 10 cm ·10 cm are selected as study objects, respectively. A hand-held 3D laser scanner is used to scan the structure plane surface to obtain its geometric features. Digital processing is carried out with the help of 3D morphological image post-processing software. On the basis of this, fractal theory is selected to calculate*JRC*values under different sampling point spacing conditions. Then sampling point interval effects on*JRC*value is analyzed and its influencing factors are studied. Results show that:*JRC*values are affected mainly by the sample point interval on profile line. However, when the sampling interval is less than a certain critical value, the*JRC*basically remains unchanged. The critical interval of sampling point is related to the roughness of structural plane and has a negative exponential relation with the*JRC*values. In addition, the negative exponential relation is influenced by rock type. The smaller the rock particles are, the better the imitative effect is. When the three types of rock are considered comprehensively, a negative exponential function*y*=1.8314*e*^{-0.067x}is obtained between the critical interval of sampling point and*JRC*value. So, to remove the interval effects on the JRC evaluation, the critical interval of sampling points can be quantified according to the function. This study can provide a scientific method for the collection of roughness information of structural plane. And the precision and efficiency of structural plane information collection can be ensured by selecting a reasonable number of sampling points.
2018, 26(5): 1342-1350.

The Roughness Joint Model(RJM)considering JRC is established in Matlab program. Simplified sinusoidal curves are used to represent roughness joints. Then it is introduced into the Particle Flow Code to conduct the uniaxial compression test. We compare the stress-strain curves of intact rock mass, linear fractured rock mass and RJM rock mass, and change the fracture inclination(angle with horizontal direction)

*α*, rock bridge angle*β*, fracture density*γ*. Then numerical models of fractured rock mass with different fracture distributions are established to conduct a series of numerical simulation tests. By monitoring the initiation and propagation of secondary fractures, the damage process of jointed rock is analyzed. Findings can be drawn as follows. (1)The presence of fractures significantly reduces the strength of rock mass. The peak strength and peak strain of RJM model are higher than those of linear cracks. (2)The compression strength of rock mass increases with the increase of joint inclination and decreases with the increase of fracture density, but changes in a nonlinear law with the fracture bridge angle. The peak strength at rock bridge inclination angle of 45°is the lowest, peak strain is the minimum. (3)The distribution of fractures has influence on failure mode, The expansion of the micro-fissures reflects the anisotropy of the mechanical properties of the rock mass. (4)Increasing the fracture density at different dip angles can reduce the strength of the rock mass in different level. When the dip angle is 75°, the density has the least influence on the strength, but at 30° and 60° the influence is the greatest. The Roughness Joint Model(RJM)considering JRC is established in Matlab program. Simplified sinusoidal curves are used to represent roughness joints. Then it is introduced into the Particle Flow Code to conduct the uniaxial compression test. We compare the stress-strain curves of intact rock mass, linear fractured rock mass and RJM rock mass, and change the fracture inclination(angle with horizontal direction)*α*, rock bridge angle*β*, fracture density*γ*. Then numerical models of fractured rock mass with different fracture distributions are established to conduct a series of numerical simulation tests. By monitoring the initiation and propagation of secondary fractures, the damage process of jointed rock is analyzed. Findings can be drawn as follows. (1)The presence of fractures significantly reduces the strength of rock mass. The peak strength and peak strain of RJM model are higher than those of linear cracks. (2)The compression strength of rock mass increases with the increase of joint inclination and decreases with the increase of fracture density, but changes in a nonlinear law with the fracture bridge angle. The peak strength at rock bridge inclination angle of 45°is the lowest, peak strain is the minimum. (3)The distribution of fractures has influence on failure mode, The expansion of the micro-fissures reflects the anisotropy of the mechanical properties of the rock mass. (4)Increasing the fracture density at different dip angles can reduce the strength of the rock mass in different level. When the dip angle is 75°, the density has the least influence on the strength, but at 30° and 60° the influence is the greatest.
2018, 26(5): 1360-1365.

A tunnel-stratum dynamic response model which crossing ground fissure is designed set in Xi'an subway project. The dynamic model test of the interaction between tunnel and stratum under the subway vibration load is carried out. The dynamic response law of stratum with ground fissure is revealed. The results show the follows. When the ground fissure is not active, the acceleration response of the arch vault is relatively smaller than that of arch bottom. It indicates that the acceleration of arch bottom caused by subway vibration is attenuated significantly when it transfers to the overlying soil through lining. When the hanging wall falls down, the vibration response of the tunnel vault and bottom is obviously stronger than that when the ground fissure is not active. It indicates that, the activity of ground fissure obviously influences the vibration performance of tunnel. Meanwhile, the acceleration amplitude between the hanging and heading wall is obviously different due to the train running position variation and the fall of hanging wall. This effect is unfavorable to the vibration performance of the tunnel, and appropriate measures should be taken to prevent the local damage or destruction of the tunnel lining.
A tunnel-stratum dynamic response model which crossing ground fissure is designed set in Xi'an subway project. The dynamic model test of the interaction between tunnel and stratum under the subway vibration load is carried out. The dynamic response law of stratum with ground fissure is revealed. The results show the follows. When the ground fissure is not active, the acceleration response of the arch vault is relatively smaller than that of arch bottom. It indicates that the acceleration of arch bottom caused by subway vibration is attenuated significantly when it transfers to the overlying soil through lining. When the hanging wall falls down, the vibration response of the tunnel vault and bottom is obviously stronger than that when the ground fissure is not active. It indicates that, the activity of ground fissure obviously influences the vibration performance of tunnel. Meanwhile, the acceleration amplitude between the hanging and heading wall is obviously different due to the train running position variation and the fall of hanging wall. This effect is unfavorable to the vibration performance of the tunnel, and appropriate measures should be taken to prevent the local damage or destruction of the tunnel lining.

2018, 26(5): 1366-1374.

Seawater intrusion is an important factor affecting the safety of construction and operation of underground water-sealed oil caverns. In the island environment, massive excavation of underground caverns can significantly change the natural groundwater seepage field of the island, causing seawater intrusion. On the one hand, seawater intrusion can increase the risk of water inrush. On the other hand, the chloride ion in the seawater can cause erosion of the structure and facilities of underground caverns, which can affect the service life of the underground oil storage cavern. Therefore, consideration is necessary about the water-sealed reliability and the stability of rock masses in the construction of groundwater oil storage cavern in island environment. Attention must be paid on seawater intrusion. Scientifically deployed water curtain systems can effectively reduce the risk of seawater intrusion. The study is based on the related theory of pollution transport, and uses finite element numerical simulation method. A simulation of seawater intrusion is carried out about the proposed large scale underground oil storage cavern of an island in Zhejiang province. Research results show the follows. Without fresh water curtain system, excavation of underground caverns can cause seawater intrusion. Horizontal water curtain system can weaken seawater intrusion to a certain extent on the basis of satisfying the reliability of water-sealed. Vertical water curtain system can obviously inhibit the occurrence of seawater intrusion. The research results provide a theoretical basis for the study of seawater intrusion in underground oil storage cavern in island environment.
Seawater intrusion is an important factor affecting the safety of construction and operation of underground water-sealed oil caverns. In the island environment, massive excavation of underground caverns can significantly change the natural groundwater seepage field of the island, causing seawater intrusion. On the one hand, seawater intrusion can increase the risk of water inrush. On the other hand, the chloride ion in the seawater can cause erosion of the structure and facilities of underground caverns, which can affect the service life of the underground oil storage cavern. Therefore, consideration is necessary about the water-sealed reliability and the stability of rock masses in the construction of groundwater oil storage cavern in island environment. Attention must be paid on seawater intrusion. Scientifically deployed water curtain systems can effectively reduce the risk of seawater intrusion. The study is based on the related theory of pollution transport, and uses finite element numerical simulation method. A simulation of seawater intrusion is carried out about the proposed large scale underground oil storage cavern of an island in Zhejiang province. Research results show the follows. Without fresh water curtain system, excavation of underground caverns can cause seawater intrusion. Horizontal water curtain system can weaken seawater intrusion to a certain extent on the basis of satisfying the reliability of water-sealed. Vertical water curtain system can obviously inhibit the occurrence of seawater intrusion. The research results provide a theoretical basis for the study of seawater intrusion in underground oil storage cavern in island environment.

2018, 26(5): 1375-1383.

As a part of Shanghai international shipping center, Yangshan Deepwater Port fourth dock is located in the northwest of the first to third docks. It is formed by sea embankment around Particle Island, Zhushan Island, Turtle Island and Fish Reef. It is the world's largest dock built relying on the island. The dock is the largest and most advanced intellectualized in the world. The new technology is adopted on the basis of the foundation. It requires high exploration results and difficult technology. Many problems faced by the offshore area are not standardized. Investigation works include completely independent intellectual property rights and comprehensive set of key technology exploration platform series, novel sampling, soft soil test new technology, and sea mud as a whole. They reveal the port consolidation of dredger fill load deformation, and put forward appropriate and effective foundation treatment scheme. Optimization design for cost reduction provides a strong data support, and achieves good social and economic benefits. These innovations symbolize the breakthroughs in China's geotechnical investigation in China's offshore geotechnical engineering, and provide useful references for similar super engineering surveys in the future.
As a part of Shanghai international shipping center, Yangshan Deepwater Port fourth dock is located in the northwest of the first to third docks. It is formed by sea embankment around Particle Island, Zhushan Island, Turtle Island and Fish Reef. It is the world's largest dock built relying on the island. The dock is the largest and most advanced intellectualized in the world. The new technology is adopted on the basis of the foundation. It requires high exploration results and difficult technology. Many problems faced by the offshore area are not standardized. Investigation works include completely independent intellectual property rights and comprehensive set of key technology exploration platform series, novel sampling, soft soil test new technology, and sea mud as a whole. They reveal the port consolidation of dredger fill load deformation, and put forward appropriate and effective foundation treatment scheme. Optimization design for cost reduction provides a strong data support, and achieves good social and economic benefits. These innovations symbolize the breakthroughs in China's geotechnical investigation in China's offshore geotechnical engineering, and provide useful references for similar super engineering surveys in the future.

2018, 26(5): 1384-1389.

Aiming at different surface movement and deformation can be produced by the inclined tunnel excavation. The formulas of ground surface movement and deformation due to semicircle-tunnel excavation have been derived by using stochastic medium theory. By calculating the ground surface movement and deformation at different sections of inclined tunnel, the surface of maximum subsidence, horizontal displacement, the maximum value of tilt and maximum curvature are obtained. The influence of surface movement and deformation is analyzed at different positions of inclined tunnel. Result may provide reference for the planning of the electric power plant.
Aiming at different surface movement and deformation can be produced by the inclined tunnel excavation. The formulas of ground surface movement and deformation due to semicircle-tunnel excavation have been derived by using stochastic medium theory. By calculating the ground surface movement and deformation at different sections of inclined tunnel, the surface of maximum subsidence, horizontal displacement, the maximum value of tilt and maximum curvature are obtained. The influence of surface movement and deformation is analyzed at different positions of inclined tunnel. Result may provide reference for the planning of the electric power plant.

2018, 26(5): 1390-1396.

For the stability analysis of underground cavern, initial stress regression analysis is a very important first step. The existing initial stress regression methods, such as function method, displacement back analysis method and boundary method, are not easy to operate. Their accuracy is not high. This paper takes the Huangdaowater sealed underground oil cavern as an example. The relationship between the principal stress and the vertical self-weight stress is obtained, according to the measured initial in-situ stress. On the basis of the numerical simulation calculated results of self-weight stress, the horizontal principal stress convert to the normal stress and shear stress of model coordinates by coordinate transformation. Finally, based on the FISH language of FLAC

^{3D}, read the normal stress and shear stress obtained in the previous step, and calculate to convergence. Then the initial stress field is generated directly. The calculated results are compared with the measured results in the field. The relative error is less than 10%. This method is simple and effective. For the stability analysis of underground cavern, initial stress regression analysis is a very important first step. The existing initial stress regression methods, such as function method, displacement back analysis method and boundary method, are not easy to operate. Their accuracy is not high. This paper takes the Huangdaowater sealed underground oil cavern as an example. The relationship between the principal stress and the vertical self-weight stress is obtained, according to the measured initial in-situ stress. On the basis of the numerical simulation calculated results of self-weight stress, the horizontal principal stress convert to the normal stress and shear stress of model coordinates by coordinate transformation. Finally, based on the FISH language of FLAC^{3D}, read the normal stress and shear stress obtained in the previous step, and calculate to convergence. Then the initial stress field is generated directly. The calculated results are compared with the measured results in the field. The relative error is less than 10%. This method is simple and effective.
2018, 26(5): 1397-1407.

In recent years, the development of global geological engineering has made a series of progress. Bibliometric methods could quantitatively characterize the development of global scientific production and research trends in geological engineering. Publications of the last decade(2008~2017) about geological engineering were retrieved from the Web of Science(WoS) Core Collection database on May 07, 2018.27025 records were analysed, using VOS viewer as a tool. The results shows:(1) The number of publications increased noticeably in the past decade. China is the most active country in geological engineering research and has more collaborations in performing the studies with each other. (2) The high frequency terms in titles and abstracts show the research scope of geological engineering remain steady, but the research fields are broaden. (3) Five Chinese research institutions are in the top 10 productive institutions and have the most collaborations with the rest institutions of the world. Seven Chinese authors are among the top 13 productive authors. Ng C W W from China and Rowe R K from Canada are the most active researchers in the field of geological engineering, but h-index of Chinese institutions and authors are relatively lower. (4) Among the top 10 preferred journals, Engineering Geology published the most articles, with 1860, and Rock Mechanics and Rock Engineering has the highest impact factor, with 2.905. This is the first bibliometric approach to examine research related to geological engineering. It can provide reference for editors to the development of geological engineering.
In recent years, the development of global geological engineering has made a series of progress. Bibliometric methods could quantitatively characterize the development of global scientific production and research trends in geological engineering. Publications of the last decade(2008~2017) about geological engineering were retrieved from the Web of Science(WoS) Core Collection database on May 07, 2018.27025 records were analysed, using VOS viewer as a tool. The results shows:(1) The number of publications increased noticeably in the past decade. China is the most active country in geological engineering research and has more collaborations in performing the studies with each other. (2) The high frequency terms in titles and abstracts show the research scope of geological engineering remain steady, but the research fields are broaden. (3) Five Chinese research institutions are in the top 10 productive institutions and have the most collaborations with the rest institutions of the world. Seven Chinese authors are among the top 13 productive authors. Ng C W W from China and Rowe R K from Canada are the most active researchers in the field of geological engineering, but h-index of Chinese institutions and authors are relatively lower. (4) Among the top 10 preferred journals, Engineering Geology published the most articles, with 1860, and Rock Mechanics and Rock Engineering has the highest impact factor, with 2.905. This is the first bibliometric approach to examine research related to geological engineering. It can provide reference for editors to the development of geological engineering.

2018, 26(5): 1155-1161.

Strong earthquake can affect the engineering quality of the rock mass forming the shallow strata of the high and steep rock slope. Moreover, engineering rock mass quality of the slope damaged by strong earthquake is crucial to making construction management plan and assessing the safety of slope. Considering actual characteristics of the slope caused by seismic disturbance, this paper applies three-dimensional laser scanning technology, based on the basic quality(BQ)system, to recognize rock mass discontinuity and joint cracks. Meanwhile, the paper combines with rock mechanical test, puts forward a method for engineering slope rock mass quality assessment. The presented method is used to assess the engineering rock mass quality of the Hongshiyan slope(zone Ⅱ)in Ludian, Yunnan Province. The results show that the shallow rock mass is awfully broken, a bit hard, and is classified as Ⅳ-class. Furthermore, the results are consistent with the practical situation. The present method can identify the structural surface information on the premise of not arriving at measurement site and is particularly appropriate for broken rock mass quality assessment of high and steep slope where surveyors cannot arrive. The paper can provide a reference for other similar projects.
Strong earthquake can affect the engineering quality of the rock mass forming the shallow strata of the high and steep rock slope. Moreover, engineering rock mass quality of the slope damaged by strong earthquake is crucial to making construction management plan and assessing the safety of slope. Considering actual characteristics of the slope caused by seismic disturbance, this paper applies three-dimensional laser scanning technology, based on the basic quality(BQ)system, to recognize rock mass discontinuity and joint cracks. Meanwhile, the paper combines with rock mechanical test, puts forward a method for engineering slope rock mass quality assessment. The presented method is used to assess the engineering rock mass quality of the Hongshiyan slope(zone Ⅱ)in Ludian, Yunnan Province. The results show that the shallow rock mass is awfully broken, a bit hard, and is classified as Ⅳ-class. Furthermore, the results are consistent with the practical situation. The present method can identify the structural surface information on the premise of not arriving at measurement site and is particularly appropriate for broken rock mass quality assessment of high and steep slope where surveyors cannot arrive. The paper can provide a reference for other similar projects.

2018, 26(5): 1219-1226.

This paper aims at the problem of low angle and long distance fluidized loess landslide triggered by earthquake. It takes the Dangjiacha landslide as a typical example. It was triggered by Haiyuan Earthquake in 1920. A detailed field investigation and indoor research are carried out for the Dangjiacha landslide. After comprehensive analysis, it reveals the genetic mechanism and movement characteristic of Dangjiacha landslide. In the earthquake process, the original structure of Ma Lan loess was destroyed destructively. The upper unsaturated loess was damaged in tension with the seismic shaking. Edge cracked, excess pore water pressure was accumulated in saturated zone. The saturated loess was liquefied and plastic deformation had accumulated. The shear surface gradually transfixed. Under the seismic horizontal ejection, the slope collapsed in high-speed. Three sliders slid to the middle groove, along the direction of 295 degree, 300 degree and 80 degree to the north respectively. Then they turned down the ditch with collision. Liquefied layer flowed forward like mudslides, showing a typical "the shear-thinning fluid", with extremely low steady-state strength and very low apparent dynamic viscosity coefficient. The upper loess floated with low angle and long distance movement, finally rushed out the valley, then turning along Bao Yu Gou to downstream for the mountain barrier. The displacement was up to 3 km. The results have important reference value or guiding significance for further understanding the formation mechanism of fluidized loess landslide under strong earthquake conditions and preventing geological disasters triggered by earthquake liquefaction in loess area.
This paper aims at the problem of low angle and long distance fluidized loess landslide triggered by earthquake. It takes the Dangjiacha landslide as a typical example. It was triggered by Haiyuan Earthquake in 1920. A detailed field investigation and indoor research are carried out for the Dangjiacha landslide. After comprehensive analysis, it reveals the genetic mechanism and movement characteristic of Dangjiacha landslide. In the earthquake process, the original structure of Ma Lan loess was destroyed destructively. The upper unsaturated loess was damaged in tension with the seismic shaking. Edge cracked, excess pore water pressure was accumulated in saturated zone. The saturated loess was liquefied and plastic deformation had accumulated. The shear surface gradually transfixed. Under the seismic horizontal ejection, the slope collapsed in high-speed. Three sliders slid to the middle groove, along the direction of 295 degree, 300 degree and 80 degree to the north respectively. Then they turned down the ditch with collision. Liquefied layer flowed forward like mudslides, showing a typical "the shear-thinning fluid", with extremely low steady-state strength and very low apparent dynamic viscosity coefficient. The upper loess floated with low angle and long distance movement, finally rushed out the valley, then turning along Bao Yu Gou to downstream for the mountain barrier. The displacement was up to 3 km. The results have important reference value or guiding significance for further understanding the formation mechanism of fluidized loess landslide under strong earthquake conditions and preventing geological disasters triggered by earthquake liquefaction in loess area.

2018, 26(5): 1272-1278.

This paper is based on the boundary surface plastic constitutive model developed with finite element platform OpenSees. It established a dynamic simple shear test model and a three-dimensional finite element model of saturated sand. The target was to explore the effect of vertical component of pulse-like ground motion on the site liquefaction. Therefore, 10 sets of seismic waves with pulse-like characteristics were selected from the Chi-Chi earthquake in Taiwan as inputting load. The simulated vertical displacement, cyclic stress ratio(CSR), pore pressure ratio and equivalent cycle number of the soil column were compared and analyzed under the ground motion with or without the vertical component. Hence the differences clearly defined the effect of vertical acceleration of pulse-like ground motion on the liquefaction of sand. The results show that the vertical acceleration component of tri-directional pulse-like seismic wave has little influence on the final site settlement but significantly increased the duration. As the CSR of soil column is less affected by vertical ground motion, the tri-directional ground motion can be simplified as horizontal bidirectional ground motion when analyzing the shearing properties of the site. The divergence of pore pressure ratio caused by the tri-directional pulse-like seismic wave is larger, and meanwhile, it spends more time to dissipate the pore pressure. Besides, tri-directional pulse-like seismic waves generate a larger equivalent cycle number and increase the duration of ground motion. As a result, the vertical acceleration have more tendency to induce site liquefaction.
This paper is based on the boundary surface plastic constitutive model developed with finite element platform OpenSees. It established a dynamic simple shear test model and a three-dimensional finite element model of saturated sand. The target was to explore the effect of vertical component of pulse-like ground motion on the site liquefaction. Therefore, 10 sets of seismic waves with pulse-like characteristics were selected from the Chi-Chi earthquake in Taiwan as inputting load. The simulated vertical displacement, cyclic stress ratio(CSR), pore pressure ratio and equivalent cycle number of the soil column were compared and analyzed under the ground motion with or without the vertical component. Hence the differences clearly defined the effect of vertical acceleration of pulse-like ground motion on the liquefaction of sand. The results show that the vertical acceleration component of tri-directional pulse-like seismic wave has little influence on the final site settlement but significantly increased the duration. As the CSR of soil column is less affected by vertical ground motion, the tri-directional ground motion can be simplified as horizontal bidirectional ground motion when analyzing the shearing properties of the site. The divergence of pore pressure ratio caused by the tri-directional pulse-like seismic wave is larger, and meanwhile, it spends more time to dissipate the pore pressure. Besides, tri-directional pulse-like seismic waves generate a larger equivalent cycle number and increase the duration of ground motion. As a result, the vertical acceleration have more tendency to induce site liquefaction.

2018, 26(5): 1351-1359.

Theoretical research method is the basic way to analyze mechanical problem. The obvious slip plane is the precondition to backfill overall stability analysis using the theoretical research method. The predecessor research results show that the interface of backfill and surrounding rock is the kind of slip plane, but the data that used to analyze the backfill stability is lack. Based on the predecessor research results, this study obtains the vertical stress differential expression in the backfill combining the inclined slip plane and ground deformation of No.2 Zone of Jinchuan Mine. Considering the sliced backfill of No. 2 Mine zone, the simply supported beam model is used to calculate the local stability of backfill. We use the numerical simulative software to analyze the maximum and minimum principal stresses of the backfill footwall, and provide the judgment of the backfill failure. The results show that the overall failure can't occur, but the local failure could occur. The upper void of two sub-levels mining leads to the deformation and failure concentrated on the corners of both sides. For the bottom void, the deformation range extends and also concentrates at corner of both sides. We should make some support in time and prevent the backfill local failure.
Theoretical research method is the basic way to analyze mechanical problem. The obvious slip plane is the precondition to backfill overall stability analysis using the theoretical research method. The predecessor research results show that the interface of backfill and surrounding rock is the kind of slip plane, but the data that used to analyze the backfill stability is lack. Based on the predecessor research results, this study obtains the vertical stress differential expression in the backfill combining the inclined slip plane and ground deformation of No.2 Zone of Jinchuan Mine. Considering the sliced backfill of No. 2 Mine zone, the simply supported beam model is used to calculate the local stability of backfill. We use the numerical simulative software to analyze the maximum and minimum principal stresses of the backfill footwall, and provide the judgment of the backfill failure. The results show that the overall failure can't occur, but the local failure could occur. The upper void of two sub-levels mining leads to the deformation and failure concentrated on the corners of both sides. For the bottom void, the deformation range extends and also concentrates at corner of both sides. We should make some support in time and prevent the backfill local failure.