Mining at deep-seated deposits is associated with high stresses, which upon redistribution during mining, may cause induced seismic events. This mining-induced seismicity can lead to rockbursts that pose a threat to mining activities. Rockburst failures are classified based on the location of their occurrences, sources, and severity. Rockbursts have been recorded in deep and hard rock mines in various countries around the world. In Sweden, rockburst failures have been prominent since the year 2000, attracting researchers to study Swedish deep mines experiencing these failures. An investigation and mapping of failures during seismic events at the Kiirunavaara mine in Sweden from 2014 to 2017 evaluated seismicity parameters but did not focus on the failures and performance of rock supports. This paper analyzes failures mapped and investigated during seismic events in four blocks at the Kirunavaara mine, focusing on the failure mechanisms of the rock and rock supports. It considers geological factors, rock support, and provides remedial suggestions. It was found that most damages were either directly or indirectly influenced by stress changes due to seismic events. This analysis can provide insights for designing rock supports to mitigate rockburst occurrences.

Rockbursts are a major problem in deep hard rock mines and one of the ways to mitigate them is to apply a well-designed rock support system. The rock support design needs to withstand not only potential energy, but also kinetic energy, which has been related to the ejection velocity of the rock mass, assumed to be equal to the peak particle velocity. It has been observed that peak particle velocities can be higher on the surface of an excavation than in solid rock and velocity amplifications up to 10 times were already recorded. Although extensive research has been done on estimation of the velocity amplification of seismic waves, there is no standard methodology to evaluate the velocity amplification factor. Seismic events recorded in Kiirunavaara Mine were studied and methodologies for estimation of the amplification of the particle velocity in time and frequency domains were developed. The velocity amplification factors calculated in both time and frequency domains by using the methodologies were compared considering dynamic response of the excavation in both roof and sidewalls, and dominant frequencies in time and frequency domains. The amplifications studied in this paper were further verified by using previous numerical simulation results. It was concluded that the dynamic response of the excavation depends on the loading situations and the fracturing status of the rock; high frequencies are filtered by fractures, and dominant frequency is higher than the corner frequency of the seismic source; amplification usually occurs in frequency range from 100 Hz to 600 Hz; it is easier and faster to calculate velocity amplification factor in time domain with less fluctuation and less overestimations in comparison with the amplification factor in frequency domain.

The numerical analysis results from an improved design of large-scale dynamic test of rock support (Test 6) is presented in this paper. The improved field test was designed based on the results obtained from field tests and the numerical analysis of the earlier tests (Tests 1 – 5) conducted at LKAB Kiirunavaara mine. The performed numerical analysis investigates how the improvements including minimizing the expansion of blasting gases into the burden, avoiding the complete damage of the burden, and creating sub-planar waves were achieved under the improved design of the test. Furthermore, the response of supported and unsupported excavations as well as the complex interaction of stress waves and rock support was numerically studied. The numerical analysis comprised of two stages (i) the explosion stage modelled with the finite element code LS-DYNA and (ii) the wave propagation stage which was modelled using UDEC with the results from LS-DYNA as input. The accuracy of the developed models was investigated by comparison of the UDEC models results to the data obtained from the field test. The numerical analysis results confirmed that the improved designed burden has assisted in reducing the areas of tensile yielding in the burden and as a result, the gas expansion and complete damage of the burden was avoided. The simulation results showed that the used support system (Swellex Mn 24 and reinforced shotcrete) has effectively limited the displacement of the test wall and prevented ejection during the dynamic loading. The combined numerical technique has shown its advantage when simulating blasting as well as interaction between waves and opening and it can thus be used as a tool for evaluating rock support performance.

The numerical analysis of four dynamic large-scale field tests conducted at LKABKiirunavaara mine are presented in this paper. The aim was to numerically studythe behavior and response of the burden and the tested walls in field Tests 1, 2, 4and 5. For this purpose, two numerical methods were combined, i.e. the finiteelement code LS-DYNA and the distinct element code UDEC. The LS-DYNA wasused to calculate the blast load, and the UDEC was used to propagate the calculatedload in the model where the geological conditions of the test site and the installedrock support in the field tests were modelled. The model was calibrated bycomparing the velocity and displacement calculated on the surface of the opening,and the zones yielded in tension were used to study the failure mechanismdeveloped in the burden. The numerical models were able to mimic the behavior ofthe jointed rock mass and the rock support fairly well. It is concluded that thenumber of major joint sets was the main reason to the difference between the failuredevelopment in Tests 1 and 2 and Tests 4 and 5. The numerical analysis of Tests 1and 2 confirmed that the gas pressure in the vicinity of the test wall in those testswas minimum. In Tests 4 and 5, it was observed that, the generated fractures in theburden combined with the natural joint condition of the burden, increased thepossibility for blocks to rotate and move within the burden. The complete burdendamage in Tests 4 and 5 was concluded to be the be due to the ejection of rockblocks in the vicinity of the test wall upon the arrival of stress wave, and ejection ofthe remaining portion of the rock blocks in the burden by the gas expansion. 

Seven large scale dynamic tests of rock support were conducted at LKAB’s KiirunavaaraMine in order to develop a methodology for the design of dynamic load resistant rocksupports. The tests were performed to increase the understanding of the response of rocksupport installed on the walls and roof in a tunnel and subjected to strong dynamicloading. The tests were originally numerically simulated by using two numerical tools,LS-DYNA and UDEC, in a former PhD Project at Luleå University of Technology, LTU(Shirzadegan, 2020), to investigate the performance of the test set up, the response ofsupported and unsupported excavations during the field tests as well as the complexinteraction of stress waves and rock support. The comparison of the numerical results andthe results obtained from field tests showed that the combination of LS-DYNA and UDECcould satisfactorily simulate the field tests. This work is the continuation of the LTUproject with LS-DYNA and UDEC.Since the analyses were previously conducted with 2D models, while in reality thestructural geology, ground motion measurements, fracture investigations and supportmotion, as well as the deformation measurements, occur in a 3D space, in the presentproject three-dimensional analysis are carried out using 3DEC that can assist in studyingthe interaction of the dynamic load with rock joints, rock blocks, tunnel wall and installedrock support. In this project the analyses are carried out in two stages: (i) the explosionstage is modelled in 3D with the finite element code LS-DYNA and (ii) the wavepropagation stage is modelled in 3D with the program 3DEC, where the results from LS-DYNA are used as input.The numerical simulations in this project can be used to analyse the behaviour of rockmasses and rock support systems under dynamic loading conditions. These simulationsprovide valuable information about the response of the rock mass to dynamic loading andcan be used to optimize the design of rock support systems. The numerical results evaluatedifferent rock support performance (Swellex Mn24 bolt and D -bolt) under dynamicloading conditions.3DEC modelling in the present study shows that: a) numerically calculated velocitypatterns match well those observed in the field for Test No. 6, b) larger displacementsoccurred for certain rock blocks in the unsupported crosscut and, c) the patterns of themost displaced rock blocks in the model very much resemble the positions where falloutswere observed in the field close to the footwall drift in the unsupported crosscut. (PDF) Dynamic modelling of rock bolts at Kiirunavaara mine. 

During the winter season, ice causes major problems in many Swedish railway tunnels. Ice, rock and shotcrete in the roof and on the walls may come loose and fall down, installations and cables can break due to ice loads and the tracks can become covered with ice. To maintain safety and prevent traffic disturbances, many tunnels require frequent maintenance. The removal of ice, loose rock and shotcrete is expensive and potentially risky work for the maintenance workers. To reduce maintenance costs, it is important to improve our knowledge of frost penetration inside tunnels and investigate the effect of ice pressure and frost shattering on loadbearing constructions. The aim of this investigation was to gather information about the problems caused by water leakage and its effect on the degradation of a rock tunnel when subjected to freezing temperatures. There are many factors that determine whether frost or ice formations will appear in tunnels. To collect information on ice formation problems, field observations were undertaken in five of Sweden’s railway tunnels between autumn 2004 and summer 2005. For one of the tunnels, follow-up observations also took place in March during the years 2005, 2006 and 2007.

Rock masses are far from being continuum and consist essentially of intact rock and discontinuities such as joints. Presences of discontinuities affects the propagation of the stress wave in rock mass. In this paper the impact of joints properties and features on the dynamic response of underground cross-cuts to seismic loading induced during dynamic large-scale field test in Kiirunavaara mine, was numerically investigated. The numerical methods used comprise the finite element code LS-DYNA and the 2D Universal Distinct Element Code (UDEC). The LS-DYNA model simulated the blasting and acquired the crushed zone and the vibration velocities around the crushed boundary. The vibration velocities from LS-DYNA were then used as an input velocity in the UDEC model. The studies of parameters such as joint normal and shear stiffness, joint spacing and joint orientation were conducted. The vibration responses at the wall of the underground cross-cut from UDEC were analyzedand compared to observed field test results. The results show that the normal stiffness has large effects on the peak particle velocity (PPV) while the shear stiffness contributes less influence. However, changes on joint space and orientation affect the PPV at the wall of the cross-cut. The joint stiffness explains the quality of the joint to transmit the stress wave while the joint spacing, joint orientation describe the blocky in burden which explain number of times the stress wave will be reflected before reaching cross-cut wall. The analysis can be useful during designing of the blast, burden as well as cross-cut support.

Canals and tunnels in hydropower plants must be able to receive high shock-like flows without damaging either the dam or the rock foundation. Although the canals often consist of rock, erosion can occur when water is released. The natural riverbeds and lakes in Sweden usually run along large faults and other zones of weakness in the rock. This is because the water could more easily erode its way along these weakness zones. Spillways of hydropower dams are generally unlined thereby exposing the bedrock to erosion during floods.This study focuses on block erosion mechanisms and characteristics in unlined spillway canals that comprises hard rock mass systems. Two hydropower dam spillway canals were investigated as case studies; identified as Dam1 and Dam 2. The spillway canals of these two dams have uniquely different bed rock characteristics. At Dam 1 the rock mass is very blocky with visually estimated GSI classification in the range of 50 to 70, while Dam 2 is composed of massive rock mass with visually assessed GSI classification of 70 to 90.The erosion characteristics observed in these two spillway canals are uniquely different. The rock mass is obviously the principal factor contributing to these observations. However, there are also other factors, namely the hydraulic factors, as well as the geometrical factors of the canals. In this report these factors have been described in detail.  Three main mechanisms of block erosion were observed, (i) removal or plucking of rock blocks, (ii) fracturing of intact rock blocks and (iii) abrasion. At Dam 1 spillway canal all three mechanisms were observed to be significantly evident. At Dam 2, abrasion is the dominant mechanism of erosion. Hydraulic parameters, water pressure and velocity, affect the criticality of the erosion.Numerical simulations of the spillway canals were conducted using 3DEC. These simulations show that block displacements greater than 10 m are experienced within 1 to 2 minutes of flow. This observation is consistent with observations made during an actual discharge from a dam. Numerical simulations indicated that blocks with sizes less than 1 m3 would easily be plucked and transported downstream. If they are intact and with unfavourable geometry, they can be easily fractured by the spill water loads. Field investigations support these observations.Remedial measures would first require classification of a spillway canal into erosion domains based on erosion vulnerability. For example, the upstream sections of the channels are typically vulnerable to high intensity erosion. Hydraulic jumps, plunge pools, stilling basins, etc, have been typically used to break up the energy before the water flows downstream. However, erosion still occurs further down since the energy is still very large. Reinforcing the bedrock with artificial supports such as rock bolting, widening and levelling of canals, diverting the flow to less vulnerable areas of the canal, etc, have been some means to reduce block erosion. This study concludes that, remedial measures must start with identifying the mechanisms of block erosion, three of which have been described above. Domaining of the channels into erosion critically domains may also assist in monitoring and application of remedial measures. Empirical methods, such as Pells (2016) can be applied in each domain to identify their erosion potential. This study also concludes that the hydraulic pressure and displacements that occur around a rock block needs to be further investigated, either by field measurements in a spillway or by using physical models. In this way, it will be possible to better understand the conditions around blocks in a spillway and erosion mechanisms during a discharge.

Despite extensive grouting efforts to prevent water from leaking into tunnels, water seepages remain. When exposed to freezing temperatures, ice formations occur. During the winter, the Swedish Transport Administration's railway tunnels are affected by major problems caused by ice, such as icicles from roof and walls, ice loads on installations, ice-covered tracks and roads, etc. To ensure safety and prevent traffic disruptions, many tunnels require extensive maintenance. Improved knowledge about frost penetration in tunnels is required to reduce maintenance of the tunnels. Frost insulated drain mats are often used at leakage spots to prevent ice formation along the tunnels. To find out which parts of a tunnel are exposed to freezing temperatures, the University of Gävle and the Royal Institute of Technology in Stockholm conducted a laboratory model test on behalf of the Swedish National Rail Administration (now the Swedish Transport Administration). The laboratory model test aimed to find a method to determine the expected temperature conditions along a tunnel to decide which parts of the tunnel require frost insulation to protect the drainage system from freezing and prevent ice formation. To evaluate the laboratory model test, the Swedish Transport Administration in collaboration with Luleå University of Technology have performed field surveys in two Swedish railway tunnels. The field measurements involved monitoring temperatures in air, rock surfaces and rock mass, as well as measuring wind direction, wind and air velocity and air pressure. The measurements in the tunnels show that the frost penetrates further into the tunnels than was expected from the laboratory model test, which was based on a completely uninsulated tunnel. Frost insulated drains do not only prevent the cold air from reaching the rock mass, but also prevent the rock from emitting geothermal heat that warms up the cold tunnel air. Consequently, the frost penetrates further into the tunnel than it would do if the heat from the rock mass was allowed to warm up the outside air on its way into the tunnel. The number of frost insulated drains and how much of the tunnel walls and roof are covered thereby affect the length of the frost penetration.

Rock tunnels excavated using drilling and blasting technique in jointed rock masses often have a very uneven and rough excavation surface. Experience from previous studies shows that the unevenness of a rock surface has a large impact on the support effect of shotcrete lining. However, clear conclusions regarding the effect of 2D and 3D uneven surfaces were not obtained due to limited studies in the literature. The numerical analyses reported in this paper were made to investigate the influence of the surface unevenness of a circular tunnel opening on the support effect of shotcrete using a 3D numerical code (3DEC). The models were first calibrated with the help of observations and measured data obtained from physical model tests. The influential factors were investigated further in this numerical study after calibration had been achieved. The numerical analyses show that, in general, the unevenness of a tunnel surface produces negative support effects due to stress concentrations in recesses (compressive) and at apexes (tensile) after excavation. However, shotcrete sprayed on a doubly waved uneven surface has better support effect compared to shotcrete sprayed on a simply waved tunnel surface. The development of shear strength (specifically frictional strength) on the uneven interface between the shotcrete and the rock contributes to this effect, in the condition where bonding of the shotcrete does not work effectively. The interface is a crucial element when the interaction between the rock and shotcrete is to be simulated. When an entire tunnel surface is covered by shotcrete with high modulus, more failures will occur in the shotcrete especially when rock surface is uneven. Based on the numerical model cases examined, some recommendations on how to incorporate tunnel surface conditions (2D or 3D unevenness) in the design of a shotcrete lining are given.