Abstract The Jurong Series of rocks in Singapore primarily comprise weakly metamorphosed sedimentary rocks that have been folded and faulted. The Series occupy a large proportion of the western side of Singapore where many major civil engineering works underground are underway or are planned. Although many geotechnical investigations have been conducted and several projects have been completed published data on the properties of these rocks is sparse. Geotechnical investigations have been carried out on project specific basis and there has not been a collation of material or mass properties in the public domain. For example, interpretative reports have described the rocks qualitatively as water-bearing with highly conductive features. In practice claims have arisen regarding expectations of strength and conductivity and these are difficult to resolve in the absence of an adequate data base. The Series includes diverse rock materials such as conglomerates, sandstone, siltstone, claystone, limestone and tuff amongst others. These rocks have a wide range of properties. The authors have collected basic properties of strength of rock material for different types of rock and for various grades of weathering and in situ conductivity from packer tests with the intention of providing basic data illustrating the broad range of properties of these rocks to which others may add subsequently. 1. Introduction The Jurong Series of rocks (Jurong Rocks) are found extensively over the western side of Singapore. They primarily comprise sedimentary rocks that have been subjected to a low grade of metamorphism and subjected to weathering in situ. The rocks of the Jurong Series are complicated because they comprise several formations with diverse lithology. They are extensively and intensively folded, sheared and faulted. Because of their complexity, determination of engineering properties can be very difficult (Zhao. 2001). In particular packer tests to determine conductivity yield widely ranging values within short distances and strength tests vary widely between adjacent specimens even when taken from the same run of core.

Abstract Shafts have a key role as access, transportation, and ventilation routes in mining, tunneling, and underground construction. In addition, the high-level radioactive waste disposal planned in Japan will require the shafts that are deeper than several hundred meters. In recent years the raise boring method is widely employed for shaft excavation in limestone quarries in Japan; however, this method possibly encounters serious troubles in fractured or weathered rock masses, such as collapse of shaft walls, stoppage of excavation, and breakage of tools. These troubles lead to an extension of construction period and an escalation of budget. Therefore, it is essential to understand the rock mass conditions around a shaft before or during excavation. In this study, an estimation method of rock strength from the excavation data of a tunnel boring machine (TBM) was applied to the shaft excavation with a raise boring machine (RBM) in a limestone quarry. Rock strengths were estimated from the thrust force and cutting depth and from the torque and cutting depth during shaft excavation with the RBM; the two strengths show a similar trend from the bottom of the shaft to the surface. In addition, the depth at the low rock strength was coincident with that at argillaceous or cracked shaft walls. The ratio of the two estimated strengths is probably an important index that alerts collapse of shaft walls and stoppage of excavation. This study validated the applicability of the estimation method of rock strength to the raise boring method. 1. Introduction Shafts have a key role as access, transportation, and ventilation routes in mining, tunneling, and underground construction. In addition, the high-level radioactive waste disposal planned in Japan will require the shafts that are deeper than several hundred meters. These vertical or inclined shafts are excavated with the drilling and blasting (D & B) method or the mechanical method, in a similar way to horizontal tunnel excavation. The D & B method is suitable for hard rock breakage and has been used for shaft excavation; however, the operation is non-continuous, and problems on safety, noise, and vibration possibly occur. The mechanical methods include the raise boring, the down reaming, the boxhole boring, and the shaft boring modified from the horizontal tunnel boring (Bilgin et al., 2014). As described in the next section, in recent years the raise boring method is widely employed in limestone quarries in Japan. This method requires no explosives and no rock supports under ideal rock mass conditions, and hence realizes safe and rapid shaft excavation. In contrast, excavation in fractured or weathered rock masses may cause serious troubles such as collapse of shaft walls, stoppage of excavation, and breakage of tools. Therefore, it is essential to understand the rock mass conditions around a shaft before or during excavation with the raise boring method. However, few reports have been published on the mechanisms and the excavation data of the raise boring method (Shaterpour-Mamaghani and Bilgin, 2016; Shaterpour-Mamaghani et al., 2016).

Abstract A vertical cut off wall for water barrier was constructed at the final landfill disposal site in Oita prefecture of Japan to prevent the inflow of groundwater from the surroundings of the landfill site and leakage of leachate (sewage) to the outside. Due to existence of hard rock formations existed in the target ground to construct the cut off wall for water barrier, a Trench cutting Re-mixing Deep wall (TRD) construction method using a pre-drilling by a double rock auger in combination was planned for excavation of cut off wall for water barrier, initially. However, if this plan was adopted, there should be tight schedule for the construction due to pre-drilling. The alternative with CSM (Cutter Soil Mixing) construction method which does not require pre-drilling even in hard rock formation was proposed and the excavation was carried out with CSM method against relatively less hard rock formations. Also, a new machine imported from overseas was used for CSM construction method. Since, in Japan, there is no previous experience for drilling of hard formation layer with the new CSM machine, a pre-test was carried out. In the pre-test, the quality of the soil mixing wall and its drilling capacity against the hard rock formation were checked. The pre-test result was included in the construction cycle for CSM method with the new machine. From this point of view, case study of construction with CSM method which was applied to vertical cut off wall for water barrier of waste landfill waste disposal site in Japan is described and the pre-test performance of rock drilling is summarized in this paper. 1. Introduction The final landfill disposal site of waste in Oita prefecture of Japan was developed for final landfill site of general disposals such as burning residual, non-inflammables and bulky garbage choice residual. Its landfill area and volume are 14,200 m 2 and 71,000 m 3 respectively, and its landfill has been completed already. However, this disposal site was designated as an inappropriate disposal site, and required to take a countermeasure in order to satisfy the national landfill closing criteria with the state aid business. Measures were mainly planned to construct a vertical water barrier for preventing the inflow of groundwater from the surroundings to the landfill site and the outflow of leached water (wastewater) to the downstream groundwater, and to perform final earth capping for controlling the penetration of rainwater into the landfill site. Initially, the cut off wall for water barrier was planned with a Trench cutting Re-mixing Deep wall (TRD) method using a pre-drilling by a double rock auger in combination due to existence of hard rock formations in the target ground. However, in the actual construction, both the TRD method which is the in-situ stirring and mixing method, and the CSM (Cutter Soil Mixing) method which does not use the supplementary method, were used in combination to shorten the construction schedule. This change of the construction method was attributed to the corroboration of the pre-test results of the CSM machine for the applicability to hard rock carried out when it was introduced from overseas, and the results of the pre-test at the actual construction site. In this report, based on these construction progress, we report a track record in which the CSM method was used for of the vertical cut off wall for water barrier of the final landfill disposal site of waste firstly in Japan, and a pre-test results of rock drilling performance, approximately 10 years ago.

Abstract Understanding of a post-closure geological environment around a large underground facility is important for the safety assessment of geological disposal of high-level radioactive waste. Japan Atomic Energy Agency (JAEA) has performed the GREET (Groundwater REcovery Experiment in Tunnel) at the Mizunami Underground Research Laboratory (MIU) to evaluate the environmental recovery process after closure. In the GREET, a mock up test drift (Closure test drift; CTD) was filled with in-situ groundwater as a simulation experiment of drift closure. The CTD is located in fractured crystalline rock at 500m depth below ground surface and has a floodable volume of approx. 900m 3 . Information of fracture distribution has been obtained by borehole investigation and mapping of gallery walls. The change of hydraulic pressure, hydrochemical condition and rock deformation have been monitored in and around the CTD during the tunnel excavation and water-filling. In parallel to the in-situ investigation, we perform a Hydro-Mechanical-Chemical (HMC) coupled simulation to enhance the understanding of the recovery process in fractured granite. This study presents the simulation results of excavation stage. The sensitivity analysis were conducted with homogeneous model based on the investigations before excavation of CTD for the rough setting of simulation conditions. In addition, we try to reproduce the detail of HMC process with heterogeneous models generated by discrete fracture network model. Comparison of simulated results with observed data leads to the conclusion that the range of change in inflow during excavation can be predicted. However, model update is necessary for prediction of groundwater chemistry and spatial distribution of abrupt change in the drawdown. 1. Introduction The Mizunami Underground Research Laboratory (MIU) is being operated by the Japan Atomic Energy Agency (JAEA), in the Cretaceous Toki Granite in the Tono area, Central Japan. The MIU project is a broad-based, multi-disciplinary study of the deep geological environment, providing a scientific basis for the research and development of technologies needed for geological disposal in crystalline rock. The MIU design consists of two shafts, and several horizontal research galleries (Figure 1). The geological settings around MIU site are summarized in e.g. Ishibashi et al., (2016).

Abstract Sequential excavation method (SEM) is commonly used for the underground rock cavern construction. One of the major focuses in the SEM process is the selection of the excavation sequence parameters including the subdivision of cavern cross-section and the round length. In this paper, the parameters of excavation design were going to be optimized by adopting the approximate excavation performance using the response surface generated by artificial neural network (ANN) model. Firstly, the training data was generated using numerical studies. Multi-staged 2D plain strain models were adopted to conduct the numerical simulations, and further associated with tunnel advance processes using the convergence and confinement method (CCM). The parameter studies were involving the studies of rock types, cavern sizes, excavation methods and cavern performance. Then, a 3-layer ANN model was used to mapping the relationship between the excavation design parameters and the tunnel performance. At last, by adopting the proposed ANN model with the optimizing function in EXCEL, a revised excavation chart was proposed to help the engineers to quickly find the optimized sequential excavation parameters. 1. Introduction The sequential excavation method (SEM) is widely used for the construction of rock caverns, shafts and other underground structures. It takes advantage of the capacity of the rock mass to support itself by deliberately controlling and adjusting the stress and deformation field which takes place in the surrounding rock mass during the excavation. Federal highway administration (2009) has proposed four essential processes in the SEM design, include: the classification of ground condition and excavation, the definition of excavation method and support classes, the instrumentation and monitoring, and the ground improvement prior to rock cavern excavation. One of the major focuses in the SEM process is the selection of the excavation sequence parameters including the subdivision of cavern cross-section, the round length (maximum unsupported excavation length) and the supports installation time. Subdividing the cavern cross-sections could heavily reduce the risk of the cavern instability during excavation (Graziani et al., 2005; Lunardi and Barla, 2014; Zhang and Goh, 2012). However, too many subdivisions will increase the required equipment and manpower and thus increase total construction costs. To optimize the excavation designs, it is important to approximate the performance of tunneling under specified SEM parameters. Response surface method (RSM) has been proposed as a useful method to predict the tunneling performance. It has been studied by researchers to present the performance in explicit form (Lü et al. 2017; Hamrouni et al, 2018). Artificial neural network (ANN) is one of the effective ways to approximate the response surface. It has been widely used in data analysis in civil engineering (Zhao and Ren, 2002; Zhao et al, 2008). Essentially, the network is trained by adapting the weights and biases using optimization methods to minimize the mean square error between the predicted and the target values. Some commercial codes such as MATLAB have provided for convenient use of ANN.

Abstract Rock related hazard identification and risk assessment within the confined underground mining space of a tabular mine can be a mastery skill which is sometime difficult to transfer. Failure to identify the lead indicators of ground instability, may unfortunately lead to a major loss, either in revenue, equipment or even a human life. Recent development in the virtual reality space as well as the use of games in education proved to have had a positive impact in the teaching and learning environment. The use of a computer simulated environment in the form of a virtual reality game where an underground worker is given the opportunity to identify the rock fall related hazards and associated risks. This then enables the worker to experience the cause-and-effect of the support design and implementation based on his/her decisions within the virtual game world. The environment is generated with a number of hazards which needs to be identified and taken cognizance of for doing a support design (cause). Given the scenario and available support elements, the candidate then determines the support methodology. The chosen elements are then marked by the trainee within the virtual game world, after which the approach is logically tested in the system. The results (effect) are then visually presented, positive or negative, based on the applied stability design and selected support elements. 1. Introduction New tunnel development in the South African mining industry peaked at levels as high as 800 km per annum in the hard rock industries, whereas the soft rock (coal) industry using a board and pillar layout far exceeded this. These excavations serves various purposes, from linking the stoping areas to the access excavations, to providing fresh air and travelling routes for man and material (including ore and waste). Industry statistics indicate that a high proportion of fatal injuries occur within 5 m from the working face. This is as expected, since the bulk of the workforce is predominately concentrated at the face.

Abstract An area consisting of UG2 Spit Reef provided a unique challenge to extract the reserves by means of a safe and feasible mining strategy. The proposed strategy was required to ameliorate the exposure of personnel to geotechnical risk whilst providing a feasible scenario from a cost perspective. The initial full channel mining strategy would extract the reserves at stoping widths in excess of 2.0 m at perceived low levels of risk and low grade. A paradigm shift was required to deviate from this historic method. The proposed mining strategy entailed undercutting the internal waste layer resulting in the extraction of the UG2 reserves at stoping widths of less than 1.2 m at quantified low risk levels and high grade. No successful attempts had previously been recorded where the UG2 Split Reef had been undercut within a highly discontinuous rock mass area such as the Spruitfontein Fault zone. This resistance to change was devolatilized by the introduction of a rigid areal support medium concomitant with several other support unit alterations. This was nevertheless received with negative criticism by a mining team intent on resisting change. The benefit though of extracting the bottom reef portion exclusively, even with inflated support costs resulted in a beneficiated grade (channel grade) to in excess of 6.0 grams per tonne processed versus the 2.6 grams per tonne from the full channel mining extraction strategy. The proposed mining method and support strategy was presented in March of 2015 and implemented in March of 2017. This paper intends to share the findings pertaining to geotechnical design, support performance and design back-analysis from monitoring results. 1. Introduction Lonmin Platinum Marikana operations are situated along the Western limb of the Bushveld Igneous Complex in the North West Province of South Africa. Karee 3 Shaft (K3) is the largest of the 12 shafts on the Lonmin Marikana lease area. The shaft is able to hoist on average 12 400 t of UG2 and Merensky ore per day (285 200 t per month). The shallower reserves of both Merensky and UG2 Reefs have nearly been depleted and what remains (Fig. 1) are the reserves along both the eastern and western boundaries of the shaft as well as the deeper reserves (825 m below surface) along the sub-decline. Both reef bodies dip from South to North on average 11° with the Merensky Reef overlaying the UG2 Reef (145 m separation). The Spruitfontein Fault can be seen along the western boundary of the shaft (Fig. 1). This fault consists of 2 major faults in conjunction with several sympathetic faults and associated jointing concentrated in an area colloquially referred to as the Spruitfontein Fault zone. The western UG2 Reef horizon is also affected by the presence of Iron Rich Ultramafic Pegmatiod (Fig. 1 - IRUP indicated in maroon which is mostly unmined) and Split Reef (indicated in blue). Pothole features (indicated in yellow) that displace the reefs are also prevalent over the entire K3 shaft property along both UG2 and Merensky horizons.

Abstract This paper presents construction method and monitoring program for the stability for the existing Ekeberg tunnels in connection with the construction of the Follobanen tunnel - a new high-speed railway tunnel. When crossing the Ekeberg tunnels, the Follobanen tunnels are just few meters below. The construction of the Follobanen is approved with three conditions that (1) No negative effect to the stability of the Ekeberg tunnels, (2) The traffic flow in the Ekeberg tunnel must be maintained at all time, and (3) Any risk of instability in the existing tunnel must be detected beforehand to make necessary precaution actions. In order to deal with the challenges, SINTEF has developed a comprehensive procedure, combining continuous rock stress measurements and displacement measurements with 2-D and 3-D numerical modelling. Continued rock stress measurements and rock stress change monitoring are actively used together with numerical model to monitor the stability situation in Ekeberg tunnel. This is to make sure that any risk of instability in the existing tunnel can be detected beforehand to make necessary precaution actions. This paper is extended version of the one presented in November 2017, in Norwegian conference - Rock Blasting Conference, Oslo, Norway. 1. Introduction This paper is an extended version of a paper published in Rock Blasting Conference in Oslo, Norway - a national conference (Trinh & Holmøy, 2017). Bane NOR (Norwegian National Rail Administration) has decided to construct the Follobanen Project - with new railway tunnels close to the existing Ekeberg road tunnels in Oslo area in Norway, connecting Oslo and Ski. The Ekeberg tunnels have diameter around 11.5 m, and the Follobanen tunnels about 9.5 m diameter. The Follobanen Project comprises a 22km long twin-tube-tunnel to be excavated mainly with tunnel boring machines (TBM), but also by drill & blast and drill & split. The drill & blast and drill & split tunnel section is in the first part of the Follobanen tunnels, near Oslo Central Station, and where the Follobanen tunnels go under the Ekeberg tunnels. The main construction phase commenced in 2015, and it is scheduled for completion in the end of 2021.

Abstract Numerical analysis of mechanical behavior of anisotropic rock mass was carried out using some practical data obtained before and during excavation of a large scale underground cavern, Bunsui 1 st power station. The large cavern was planned in a place where well-developed pelitic schist dominated and the sheet-like grains were inclined to open along the schistosity. To design the reasonable rock support system of the large cavern and ensure the stability during the excavation, the mechanical anisotropic properties should be quantitatively estimated by the appropriate analysis method. There are several analysis methods which can consider discontinuity of rocks. But no one other than the Multiple Yield Model (MYM) cannot consider both strength anisotropy and deformation anisotropy. In this article the applicability of the MYM to the pelitic schist was examined by laboratory experiments and in the actual excavation phase. The results of the step-by-step stability analysis using the accumulated data were reflected to optimize the rock support system. The construction has now successfully completed without any severe troubles but with large displacements due to the rock anisotropy. 1. Introduction To excavate a large cavern in well-developed anisotropic rock mass such as the pelitic schist, it is important to evaluate the mechanical properties of the anisotropic rock. And the appropriate numerical analysis method is also important to secure the mechanical stability during the excavation. Accurate prediction of the rock mass behavior is required using results of the geological investigation. There are two analytical methods which can deal with discontinuity of rocks. One is the equivalent continuum analysis method such as MBC. In the method the discontinuity of the rock is expressed using the constitutive law with the equivalent elasticity coefficients. The other ones are the discontinuous models such as DEM and DDA. They express the discontinuity as the geometric distribution of the different materials. However, these methods can not consider both the strength anisotropy and the deformation anisotropy efficiently. Authors considered the schistosity of the pelitic schist as potential discontinuity in the MYM which is one of the equivalent continuum analytical methods.

Abstract Limited information on the ground structure and properties, as well as simplifications in the design lead to uncertainties in assessing ground and systems behaviors. Considering all expected and unexpected conditions in the a-priori design is acceptable only in special cases. For safe and economical tunnel construction the residual risk needs to be managed by appropriate monitoring and a safety management system. The paper addresses requirements for the geotechnical safety management, as well as state-of the-art monitoring and data evaluation techniques in the context of the observational approach. Experience with safety management systems over the last two decades shows that the risk can be significantly reduced. Case histories are used to demonstrate the potential of proper monitoring and safety management. 1. Introduction Uncertainty is unavoidable in underground projects. It originates from the impossibility to completely investigate the geological and geotechnical conditions, even if considerable effort has been made during the preparation phase. Upscaling of small scale laboratory test results to representative rock mass volumes and simplifications in our models and analyses further decrease the accuracy of the prediction of the ground and system behaviors. To allow for safe and economical construction of underground structures, sound preparation, as well as accompanying measures during construction are essential. When applying the observational approach, following key issues shall be addressed in accordance with Eurocode EC 7 (ӦNORM EN 1997–1): Design concept for determining excavation and support Determination of safety relevant parameters, including definition of expected behavior and criteria for the assessment of the system stability on the basis of the expected ground conditions Monitoring concept, allowing a continuous comparison of expected and observed behavior, including all organizational and technical requirements Management concept for cases where ground conditions and/or system behavior deviate from that predicted, both for favorable and unfavorable deviations. During construction attention should be paid to following issues: Appropriate organization of the site, and employment of qualified personnel Technical provisions for high quality observation of the system behavior Efficient information flow and reporting

Abstract Portal sections of road and motorway tunnels are usually the most demanding parts of tunnel construction because they are usually executed under inclined surface and, in most cases, in areas with poor rock quality. Currently, there is a limited number of investigation on the stability of ground associated with the excavation of tunnel portal. This is because of the increased computational cost required to model the detailed three-dimensional excavation process. As such, the use of idealized two-dimensional simulation can be expected in order to draw selected preliminary results with respect to the impact of the tunnel portal excavation. The inclined surface of ground is simplified into a triangle shape and is modeled as a linear elastic material. Particular emphasis is placed on the initial stress state that the major principal stress ( σ 1 ) is dependent on the slope height and is not recognized as such simple two/three-dimensional state, i.e. lithostatic loading under horizontal ground surface, where σ 1 is vertical and equal to the overburden rock load. The principal stress orientates itself and is recognized as to be parallel to the surface near the slope surface. Under the stress state, the vertical component σ z of stress is calculated with the Mohr stress circle. A two-dimensional FEM mesh is assembled so that the mesh would include the tunnel cross section. Based on the stress state under the inclined surface, a series of model runs are performed focusing on the influence of the slope angle and the Poisson's ratio. The initial stress state has been assumed with linear elastic body, however, the tunnel cross section analysis will give a better understanding on the deformation and stability of tunnel portal construction. 1. Introduction Portal sections of road and motorway tunnels are usually the most demanding parts of tunnel construction because they are usually executed in ground under inclined ground surface (Fig. 1). The lining for tunnel must be carefully designed with considering earth pressure and strength of ground (JSCE, 2016). With the surface inclination, the major principal stress σ 1 is dependent on the angle of inclination and is not recognized as such simple two/three-dimensional state, i.e. lithostatic loading under horizontal ground surface,, where σ 1 is in vertical direction and equal to the overburden rock load. The principal stress orientates itself and is recognized as to be parallel to the inclination near the slope surface. Furthermore, in the shallow ground near the ground surface, geological conditions are poor with low or lower soil/rock quality, then the excavation of tunnel portal will easily lead to not only the slope instability but also the tunnel itself. Wang, et. al. (2017) have analyzed the deformation of ground around a tunnel portal section using a three-dimensional finite element code and have concluded that the measurement of surface settlement is very important to check the stability of ground. Currently, there is a limited number of investigation on the stability of the ground associated with the excavation of tunnel portal. This is because of the increased computational cost required to model the ground and the detailed three-dimensional excavation process. Two-dimensional numerical modeling will save the cost. A two-dimensional modeling for the tunnel portal excavation using a finite element code was proposed by Muhammad, et. al. (2017). They concluded that the selection of an appropriate excavation sequence to ensure the maximum safety within the project constraints and pre-defined tunnel support. However, the stress state under inclined ground surface as mentioned in the above cannot be explained with the two-dimensional modeling. As such, the use of idealized two-dimensional simulation can be expected in order to draw selected preliminary results with respect to the impact of the tunnel portal excavation. In this paper, inclined surface of ground is simplified into a triangle and is modeled as a linear elastic material. Particular emphasis is placed on the vertical component σ 7 of stress is calculated with the Mohr stress circle and evaluated with the overburden rock load γh, where γ is the unit weight and h is the depth from the surface. A two-dimensional mesh is assembled so that the mesh would include the tunnel cross section. Based on the stress state under the inclined surface, a series of model runs are performed focusing on the influence of the slope angle and the Poisson's ratio. In the tunnel cross section analysis, the horizontal stress in the horizontal plane, perpendicular to the tunnel axis, is estimated as a function of the Poisson's ratio. The research results provide suggestions for the engineering support measures of tunnel portal section.

Abstract The high-speed railway between Beijing and Zhangjiakou in China that is a very famous project all around the world is now under construction. Badaling station is at the middle of this railway line and is designed as an underground station. The transition zone between the running tunnel to Zhangjiakou direction and Badaling station has a large span cross section with a dimension of up to 30 meters. Meanwhile, this large cross section also goes through the fault fracture zone. As a result, the supporting scheme and stability of the surrounding rock as well as seismic safety are the main concern about this major project. In this paper a 3-D rock-tunnel dynamic interaction finite element modeling is carried out to analyze the construction stage and seismic performance of the large span tunnel cross section. Numerical results have demonstrated the rationality of support system and revealed the seismic performance of the large span cross section. 1. Introduction The new Badaling Tunnel is located between Changping Nankou Town and Badaling Town in Yanqing County. The world's deepest and largest high-speed train station (Badaling underground station) which will be an important part of the 12km long tunnel between Beijing and Zhangjiakou will be 102m deep with a floor area of about 36000 m 2 . Starting section of the transition section of the station in Beijing directions is DK67+653 and that in Zhangjiakou direction is DK68+285. This tunnel station comprises of three different sized cross sections namely: small distance spaced section, large-span and triple arch section. The transition zone towards Zhangjiakou direction spans through a fault fracture zone which makes it vulnerable to seismic activity and needs to be investigated. Figure 1 shows the plan of the station. Based on the prevailing geological conditions, seismic effects and station structure details form the comprehensive geological survey report, it is necessary to analyze the overall seismic performance of the Badaling underground station. The transition tunnel is of a maximum net width and height about 30.83m and 17.57m, respectively, with a height - span ratio of 0.57. The Zhangjiakou direction transition section passes through a fault fracture zone and is the focus of this investigation. The plan of this transition section is shown in figure 2. The surrounding rock is graded 3–5 and the Norwegian method is adopted for the excavation. Initial support system against the rock includes shotcrete, prestressed cables and prestressed anchors.

Abstract Round length in conventional tunneling can pose an economical effect. However, in Japan, the round length is generally determined through a ground class and support pattern according to the guidelines provided by owners. Round lengths, which are against the guidelines, are not easily adopted mainly owing to the lack of experiments. In this study, numerical analysis was conducted to grasp a rough outline of the influence of round length on the displacement and stress of support during excavation. Based on the results, field tests were conducted to examine the actual behavior of a tunnel when the round length was changed to 1.2–2.0 m from 1.0–1.5 m, which are originally provided by the guideline. Accordingly, the following major conclusions were drawn. According to the numerical analysis of continuous ground, the influence of the changing round length is small in terms of displacement and stress of support. The same trend was observed in the field tests of this study. However, the frequency of the fall of loose rocks increased in the case of discontinuity-dominant rock mass. Based on the results, an index for determination of round length is proposed in this paper. 1. Introduction The construction of tunnels in mountains requires low cost and a short construction period. This includes the extension of the round length as a method of rationalization. The installation interval of excavation support has expanded compared to that of the standard round length, and the reduction of support members and increase in the construction speed are considered possible. However, the area of unsupported ground has increased, raising some concerns about the lowering of safety at the time of construction and instability of support due to deformation or increase in stress on the support. To solve such problems, we performed a three-dimensional (3D) numerical analysis or field test by using the extended round length to determine the effect of the extended round length on ground and support. 2. Standard round length The standard round length of a mountain tunnel is decided according to the ground classification. This length is calculated between 1 and 2 m and is related to the thickness of the spray concrete, pitch and size of the steel-arched support, and pitch and size of the rock bolt. These combinations are termed as the support patterns and are classified into five categories, as listed in Table 1.

Abstract The present work takes one deep mine repairing project of rectangular crossheading at fully-mechanized caving workface as the background. Based on both the theoretical analysis and experimental implement, the failure zone and evolution law of the surrounding rock around crossheading are achieved. The main influencing factors can be regarded as the magnitude and direction of initial rock stress, the current goaf status of the adjacent workface, mining stresses, rock properties, geological structure and groundwater, respectively. Large surrounding rock deformation, roof falling, rib spalling and asymmetric deformation are treated as the failure sign. The depressurization, reinforced support and yielding pressure supporting principles are proposed in the end, which the supporting scheme is applied along with the FLAC evaluation to guarantee the accuracy. The engineering application shows that the proposed supporting parameters are reasonable and effective. 1. Introduction With the development of deep mining (Liu, Q., et al., 2004), gas leakage, rock blasting, high mine pressure (Wan, Y., et al., 2006), big crossheading strain, hard crossheading repair and the increasing maintenance rate (Ren, J., et al., 2014), the stability and supporting technique of deep crossheading surrounding is currently an urgent issue (Wang, F., et al., 2016), which already causes the worldwide attention (Yang, X., et al., 2013). Due to the stress concentration of rectangular crossheading excavation, rocks around the crossheading gain large deformation (ZHANG, X., et al., 2004), and may even triggered roof fall, floor heaving and rib spalling (Li, D., et al., 2009). These problems seriously restrict the excavation speed (Wang, M., et al., 2016), which directly impact the safe and efficient exploitation of mine construction (Zhang, L., et al., 2014). At the same time, the huge maintenance rate arise from improper supporting not only bring tremendous loss, but also get the whole mine in trouble, or even worse shut down (Fu, J., et al., 2004). It is one of the key problems for the deep mining development and safety excavation to solve the problem of rectangular crossheading supporting technique in deep coal seam. Therefore, it is necessary to analyze the failure properties of the surrounding rock around rectangular crossheading and put forward a reliable repairing support technique in deep thick coal seam, which is of great significance to realize the safety production of the coal mine. 2. Project background The mine is located about in northwest of Binxian 20km away, which belongs to Xianyang city of Shaanxi province. The depth of the coal seam is 600m, the maximum thickness of coal seam is 26.2 meters, which owns an average thickness of 14.49 meters. As a deep thick mine, initial stress of the crossheading is high. Coal tends to possess the same characteristics of soft rock under high stress, which results in large deformation. Taking advantage of the fully mechanized sublevel caving method, the mining crossheading has obvious dynamic pressure along with the workface stoping, which aggravates the deformation and instability of the crossheading. The specific performance is large floor heave in crossheading mining. The deformation of transportation crossheading is also serious, which lead to the failure of supporting system for the initial roadway. The roof sinks, rib spalls, influencing zone expands, which seriously affects the normal safety production. The formation parameters are shown in Table 1.

Abstract In the construction of artificial slopes in landslide areas, soft-ground areas, and urban areas, it is important to place multiple surface and underground sensors in response to geology and displacement characteristics of the area and to identify the changes in the area over time. Accordingly, we have developed a three-dimensional information and communication technology (3D-ICT) system that shows various measurement data in real time on a website and determines the overall ground stability using a newly developed evaluation technique. This report outlines the system, the gives examples of application to open-cut excavation and tunnel construction, and presents details of the system using construction information modeling/management concepts. 1. Introduction The deformation and failure of slopes are important considerations in excavation under difficult ground conditions such as in areas with soft ground, in areas prone to landslides, and in urban areas with neighboring constructions. Therefore, to evaluate the stability of rock slopes and ensure the safety of a construction, many sensors are typically placed to monitor the condition of the ground as the construction progresses. However, it is difficult to integrate data processing with a variety of measuring instruments that are supplied by different manufacturers and placed on the ground and underground. Furthermore, quick and comprehensive processing of measurement data is absolutely imperative for understanding the situation when slope deformation or failure occurs as a result of rain or an earthquake, but a problem with conventional systems is that they require a long time to comprehensively evaluate stability when data from different measuring instruments must be processed separately. To solve this problem, we have developed a slope-measuring three-dimensional information and communication technology (3D-ICT) system, called the Hazama Ando Automatic Monitoring System (HAMONIS), that can handle various measuring instruments and that integrates the measurement data and provide them over the Internet in real time (see Fig. 1). This system was adopted at construction sites where there were problems with the stability of excavation slopes or with displacement in areas with shallow soil cover due to excavation for the tunnel construction. The system contributed to the quality and safety of the site. Various tests are now carried out in such cases by using construction information modeling/management (CIM) as shown in Fig. 2. Specifically, the geological conditions assumed in the feasibility and design stages are put into a 3D model, and the existence of unstable ground is examined in detail, so that such ground is addressed in the excavation plan or countermeasure designs, depending on the situation. In addition, in our approach, observation results on the geological structure on the tunnel face and excavation slope are input into the model during the construction stage so that the work results can be efficiently summarized in order to examine any differences between the preliminary study results and the actual situation and to discuss any necessary changes in the construction plan or design.

Abstract This paper focuses on some key design principles for rockbolting in underground rock excavation. The items discussed include underground loading conditions, the natural pressure arch around the underground opening, design methodologies, the responses of rockbolts under different loading conditions, the failure modes of rockbolts in fields, the determination of bolt length and spacing, determination of the factor of safety, and the compatibility between support elements. Rock support elements are loaded dynamically owing to the energy release during rock excavation and statically by the deadweight of potentially falling blocks afterward. In highly stressed rock masses, the dynamic load on the support elements is not a constant but correlated with the rock deformation. In this case, the released energy has to be taken into account in rock support design. There always exists a natural pressure arch in the surrounding rock of an underground opening after it is excavated. The methodology of rockbolting is associated with the position of the natural pressure arch with respect to the tunnel wall. Rockbolts should be long enough to reach the natural pressure arch when it is not far from the tunnel wall. However, an artificial pressure arch needs to be established in the fracture zone when the natural pressure arch is far from the tunnel wall. Bolt spacing is more important than bolt length in the case of establishing an artificial pressure arch. The factor of safety for a support system is determined by different parameters that are dependent on the loading condition. All support elements in a support system should be compatible in deformability. 1. Introduction Rockbolting is the most commonly method for rock support in underground works. Rockbolting design has been a trial and error business for a long time. In other words, it is mainly based on experiences. It is believed that the empirical rockbolting design will continue to dominate the rock support practice for a long time in the future because input data related to geology and the mechanical properties of the rock mass are always not completed in any rock engineering project. However, theories and new knowledge are helpful in guiding the rockbolting design, such as, in the aspects of the selection of rockbolt type, bolting pattern, bolt length and bolt spacing. Principles and methodologies for rockbolting design will be talked about in this paper, which include the concept of pressure arch, support principles under different rock conditions, the determination of bolt length and spacing, the factor of safety, the compatibility between support devices, etc.

Abstract After clarifying the terminology of man-made caverns and the advantages of mined caverns compared to other storage types, the concept of mined cavern is presented and detailed within a historical frame. The major evolutions of the concept are developed along with the main technical steps, from mining to tunnelling technology. Emphasize is given to the hydrodynamic containment allowing liquid hydrocarbons and LPGs to be stored in unlined caverns for a large range of geological conditions. More recently stability issues of megawedges have been considered in parallel with the general increase of cavern size in the last decades. The conclusion focusses on the contribution of structural geology in this domain. 1. Introduction Underground storages of hydrocarbons are conventionally classified into three major types corresponding to the three geological contexts in which they are developed: storages in porous media, storages in leached salt caverns and storages in mined cavities (Gomez-Montalvo, 1997; Londe, 2017) Storages in porous environments include aquifers (porous geological structures, sealed by a rock cap) and depleted fields (old oil or gas fields, production phase being terminated). The latter type, for which the tightness of the geological structure has not to be demonstrated, is the oldest underground storage of hydrocarbons, with the 1 st facility put in service in 1916 in Zoar, New York (USA). The second type, storage in salt caverns, represents an adaptation of the cavities leached by fresh water in order to dissolve salt arranged in layers or domes and create a stable volume. This type of storage is the cheapest one by working volume (barrel or cubic-metre). The third type, mined caverns, have been first designed and developed in the USA and 73 caverns were created between the years 1950 and 1984 in this country. At that time, the technology was a pure mining one, adapted to and modified for the purpose of underground storage. The room-and-pillar design (Fig. 1, left), classical in mining industry, represents this obvious mining heritage. The excavation used compressed air hammers and then explosive (drill-and-blast method) and even the name "mined caverns" refers to mining industry. Caverns were small even very small and the total volume of the 73 American caverns reached a bit less than 3 300 000 cubic-meters (as compared to present caverns which frequently reach 1 000 000 cubic-metre per cavern). Along with the technological developments, the room height increased as well as the room width, giving higher extraction ratios. The extraction ratio which was originally 0.3 to 0.5 in the 50s reached much higher values, 0.60 to even 0.75 at the end of the 70s. Despite a very interesting development, these values represented a technological limit which led to the abandon of room-and-pillar design for underground storages at the beginning of the 80s. A few room-and-pillar mined caverns have been excavated using road-headers, a technological development which increased productivity but with limited sections.

Abstract The glass solidified body reprocessed by the high level nuclear waste disposal in the process of nuclear fuel recycle was the final design condition to construct the underground disposal tunnel under the over 300m depth in Japan. The design methodology and construction technique of underground facilities had been discussed. However, there has been 17,000 tons of spent nuclear fuel rods kept by the nuclear power plants at poresent. It is necessary to develop the research of the construction method of direct disposal of the spent nuclear fuel rod. In the direct waste disposal facilities, it is assumed to become tunnel section bigger because of workability of construction and operation and closing when consider with underground disposal facilities. Also, the dynamic influence of tunnel invert may be bigger when the waste transport and movement system operated because the weight of spent nuclear fuel rod is bigger than the glass solidified body. Therefore, this study is performed about a dynamic stability of direct waste disposal facilities, during tunnel excavation, the influence of the ground specialty during excavation and the influence of the load from transport and movement system to tunnel invert. This study applied 2D and 3D numerical analysis. 1. Introduction The process of deep geological disposal of high-level nuclear radioactive waste reproducted by nuclear power plant is main disposal method in Japan. According to the report titled "The second progress report on research and development for the geological disposal of High Level Waste in Japan", a basic idea of geological disposal is called "multiple barrier system". In this system, waste liquid generated in the process of nuclear fuel recycle is fixed as the vitrification. The vitrified solid is covered with isolation material in the pit of disposal tunnel located deep underground. The geological disposal facility consists of vertical shafts, disposal tunnel and disposal pits. The design of this facility needs to consider the economic efficiency and the workability at each stage of operation on construction, operation and closure. In the design of a disposal tunnel where waste is transported and stored, there are three factors of design mechanical stability, workability of construction, operation, backfilling and economy. Then, the optimum sectional shape is proposed for the tunnel of the geological disposal facility.

Abstract With the advancement of science and technology, humans endeavoured to build massive caverns underground taking the advantage of physico-mechanical properties of the rockmass. The rockmass has inherent discontinuities in it whose properties vary greatly from the host rock aiding in the development of potential failure zones during and after execution of such projects. The change in rockmass behaviour observed in such zones calls for safety controls to alarm the working personnel inside the caverns. There arises the need for placing geotechnical and geodetic instrumentation inside rockmass to capture changes in its behaviour and promptly take up the remedial measures to prevent failures. To acquire correct data for right interpretation, there must be a right procedure to be adopted for planning the type of sensors and its specifications, location inside caverns, mode and frequency of data acquisition, data communication and data analysis. Similar planning was carried out for the caverns of an underground powerhouse complex of Punatsangchhu-II Hydroelectric Project, Bhutan by the authors. The intrinsic complexities and the problems tackled during planning and execution of such mega project are explained in detail in this paper. 1. Introduction Excavations of underground caverns for storing crude oil, construction of powerhouse, nuclear repositories and mining minerals in recent days have increased tremendously throughout the world, thereby maximizing the utilization of underground space. But since, rock is a discontinuous, inhomogeneous and anisotropic material, the reliability of structural integrity remains uncertain. The act of excavation against nature destabilizes the surrounding rockmass which leads to development of potentially unstable zones which deforms with time and if not properly treated or supported, leads to progressive failure of the structure itself. Based on the scale of excavations, the risk associated with the project to lives and property is assessed. In order to prevent any mishap, underground projects call for geotechnical and geodetic instrumentation, that helps in early detection of such unstable zones and any abnormal behavior of the rockmass. Generally, instrumentation in underground rockmass is implemented to accomplish the needs of diagnosis, prediction, legislation and research i.e. verification of design parameters, suitability of any new construction technique, diagnosing cause of an adverse event or verification of continued satisfaction behavior of the rockmass to different operations (Dunnicliff, 1998).

Abstract The three most predominant methods for hard rock excavation and fragmentation are the use of explosives, mechanical impact/cutting and hydraulic fracturing. However, those methods have inherent drawbacks, such as non-applicability or poor performance in extremely hard and abrasive rocks. Novel rock fracturing and fragmentation methods are in need to either work individually or in combined forms to break rocks. Research shows that some rock forming minerals and water can be heated up rapidly by microwave, to induce microcracks and fractures in rocks. Microwave therefore can be regarded as a promising technology of hard rock fracturing and fragmentation, with the potential of energy and cost efficiency. This keynote first provides a brief review of the research on microwave effects on rock fracturing, followed by descriptions of experimental studies of crack formation in different rocks treated by a low power industrial microwave. Possible fracturing mechanisms by microwave treatment are discussed, and the applications of microwave treatment assisting rock excavation coupled with mechanical means are outlined at the end of this keynote. 1. Introduction Fracturing and fragmentation of hard rocks is one of the most important tasks in rock engineering in the fields of mining, petroleum, and tunneling industries. While drilling and blasting remains to be the most powerful and effective method to break hard rocks, the uses of explosives are often limited by various constraints. Hard rocks are excavated and fractured by other means, including mechanical cutting and drilling, hydraulic fracturing and heating. Mechanical cutting becomes a dominant method for large scale rock excavation, particularly in tunneling, with the extensive uses of tunnel boring machines (TBMs) and roadheaders. However, hard rocks can pose great challenges to mechanical cutting and fracturing due to the extreme hardness and abrasiveness, which often lead to reduced advance rates and increased cutting tool wear. Other methods of rock fracturing and cutting have been investigated, including waterjet, laser, millimeter wave and microwave, as a sole mean for rock cutting or in combination with mechanical excavation. New technologies explored for possible use of rock fracturing and cutting are the electrical methods and the electromagnetic (EM) methods. The electrical methods include plasma blasting, electron beam and electric current. The EM methods use electromagnetic waves, including laser cutting, infrared irradiation, torch heating, and microwave heating. Mostly those technologies are investigated as a sole mean for rock fracturing or vaporization. The study presented in this keynote uses microwave technology to induce microcracks in hard rocks hence to ease the mechanical excavation. The objective is on adapting and developing low powered microwave tools to assist mechanical excavation of hard rocks, as an economical alternative.