A method to determine the five elastic constants of a transversely isotropic rock from a single-orientation core using strip load test was proposed. When the strip load test is used, the five elastic constants can be determined either by strain inversion method or artificial neural networks. However, it was found that elastic constants resulted from strip load test show sensitive reaction to rock heterogeneity. One of the ways to solve this problem is to measure larger number of strain values. It is known that the digital image correlation (DIC) method can measure large amount of strain values of rock specimen, replacing the strain gauge attachment method. Comparing with the strain gauge attachment method through numerical simulation, this study investigated how stably elastic constants can be determined when the DIC method measures strain values. The results show that determined elastic constants using the DIC method are more stable than those from the strain gauge attachment method

Representative elementary volume (REV) is defined as the rock mass volume with respect to the size of geotechnical structures, above which the rock mass is considered homogeneous and isotropic. The REV of a jointed rock mass can be determined using finite numerical analysis, but the effect of using different finite element (FE) model settings has not been widely studied. This paper aims to compare various finite element codes and their settings for determining the REV size of an excavated jointed rock mass. We used the scriptable, free program ADONIS and Rocscience's RS2 and RS3 to numerically analyse circular excavation in a rock mass intersected by orthogonal joint sets with specified joint spacing. We examined the effect of implementing the extended finite element (XFEM) method, shear strength reduction, and 3-dimensional analysis. The determined REV sizes in terms of the opening diameter to joint spacing ratio are generally comparable for all analyses and programs, except when XFEM is enabled in RS2 such that the explicit joint interface does not conform with the FE mesh, the REV size becomes relatively larger. XFEM, SSR, and 3-dimensional analyses required significantly higher computational time. The lattermost could exceed the computation capacity when analysing a densely jointed rock mass. Considering computation efficiency, a 2-dimensional, efficient and representative FE model for a complete range analysis to obtain a reliable REV size of a jointed rock mass is preferred over a 3-dimensional analysis.

Knowledge of three-dimensional (3D) in-situ stress distribution plays a crucial role in the safety and productivity assessment of numerous rock engineering projects. The stress distribution analysis in coal seams in particular poses considerable challenges because they can present a complex variation in stress direction and magnitude despite the lack of major structural features such as large fractures/faults or intrusions. The uncertainties with coal stress simulations is partly linked to i) the lack of borehole stress constraints in the coal seam, ii) the lack of a proper optimisation technique to meet available constraints in over-underlying strata of coal seam and iii) the scarcity of open-source numerical simulators capable of handling complex structural models and optimisation techniques. Here, the stress state of a coal mine is simulated through developing an efficient in-house Finite Element (FE) numerical simulator (3DiStress) capable of importing complex geological models (structural and property). The 3DiStress is equipped with an automatic optimisation algorithm to meet the local stress constraints during the 3D Finite Element stress simulation. Two geological models of different sizes are simulated to eliminate any potential boundary effects on simulated stresses. The obtained numerical results confirm the considerable variations, particularly in the orientation of the horizontal in-situ stresses well-aligned with the local stress map of the region and are found to be linked to non-uniformity in the structural geometry and heterogeneity in elastic properties of the coal seam and its surrounding formations.

Understanding the flow dynamics in rock masses presents a significant scientific and practical challenge. These rock masses are characterized by several discontinuity features such as bedding, joints, fractures, and faults, serving as important reservoirs for various geo-resources like water, oil, steam, CO₂, and methane. However, comprehending the complex interaction of fluid and gas flows within the networks of rock mass discontinuities remains a challenge. The study of flow dynamics within these discontinuities holds broad applications in fields such as geothermal energy, tunnelling, oil and gas extraction, nuclear waste disposal, and CO₂ storage. This research focuses on investigating fracture permeability through direct field surveys and remote techniques such as laser scanning, ground, and drone photogrammetry. Field work in compulsory nor understanding the discontinuity network, in order to collect the essential data enabling the construction of representative 3D discontinuity networks. The development of these networks aids in understanding the flow behaviour within specific geological contexts. Notably, our approach incorporates advanced image analysis techniques to extract the trace of discontinuities from photogrammetric images, contributing to the refinement of the geological model. The findings of this research have practical implications, particularly in identifying potential hazards associated with water and methane inflow during tunnelling, geothermal energy and oil and gas operations, nuclear waste disposal, and CO₂ storage. Furthermore, this approach supports environmental protection efforts by providing a better understanding of flow dynamics, thereby contributing to the safeguarding of natural resources. Compliance with regulations such as the EU Water Framework Directive and the DNSH rule is essential in promoting environmental compatibility. This research offers a valuable methodology for understanding the flow dynamics within rock masses, focusing on discontinuity networks. By employing advanced data collection techniques and integrating advanced image analysis processes, this approach serves as a valuable tool for various applications. It represents the intersection of theoretical understanding and practical application, with implications for managing natural hazards, resource exploitation, and environmental safeguarding.

Coastal communities are increasingly exposed to the impending hazards of climate change and global warming. More intense and frequent extreme weather events, sea level rise and tidal inundations are making not only sandy coasts but also rocky coasts highly vulnerable to both erosion processes and instability phenomena. Rockfalls and cliff collapses are increasingly induced by the higher frequency-magnitude of atmospheric and marine processes, such as storm water events, nearshore current actions, hydrodynamic impacts of wind‐induced waves and sea spray, which are responsible for rock weathering, basal erosion (undermining and notching), loss of defensive beaches and removal of protective fallen debris from the lower cliff face. The occurrence of such instability phenomena require a deeper understanding of the failure mechanisms of coastal cliffs in order to develop appropriate management plans and coastal zone governance, so as to be able to increase public safety and reduce land loss and damage to structures, infrastructures and economic activities (tourism, industries, fishing, aquaculture, etc.). In this paper, a parametric study is performed on soft rock cliffs with basal notches, in order to investigate the effects of the undermining on the stability of the rock masses. To this aim, 2D FEM numerical analyses are carried out with the RS2 code from Rocscience. The cliffs are assumed to have different heights and joints with variable depth and persistence.

Discontinuities such as joints, bedding planes and faults govern the mechanical strength and deformation of rock masses. Thorough knowledge and proper simulation of discontinuity mechanical behaviour are of paramount importance in all rock engineering projects. Nowadays, many computational codes allow to explicitly model discontinuities rather than considering their role within the context of an equivalent continuum representation of the rock mass. Over the past decades, several theoretical and empirical constitutive models have been proposed and implemented in numerical codes. The accuracy of reproducing the discontinuity mechanical response and, in turn, the complexity of these models have increased conjointly with advances in computational methods. Even if the usage of advanced constitutive models to realistically reproduce the discontinuity behaviour is more attractive, the strong nonlinearity of these models may provide difficulty for their implementation, with consequent numerical convergence and stability problems and, in addition, the definition and calibrations of the required parameters might be toilsome. A compromise between the complexity (realism) of a constitutive model, the challenge of its numerical implementation and the definition of the parameters characterizing the model response is thus needed. Therefore, simplified models are still more commonly adopted in practical rock engineering due to their user-friendliness and easy-to-determine parameters, but their adoption might result in non-fully optimized design solutions because they might not thoroughly capture the discontinuity behaviour as experimentally observed. This paper discusses the results of a numerical study that examines the influence of adopting different constitutive models for simulating the behaviour of a fractured rock mass. The paper initially provides an overview of the main constitutive models proposed in the literature by focusing on their theoretical consistency (suitability) and practical values (complexities and limitations). It then introduces the features of a proposed constitutive approach aimed at guaranteeing the theoretical rigorousness and overcoming potential implementation issues of the empirically derived formulation of the Barton-Bandis criterion. The proposed model has been implemented in the finite element code PLAXIS and its performance is inspected through numerical analyses. The results are compared with those obtained with an elasto-plastic constitutive relationship with strain-softening based on the Coulomb yield criterion, providing insights to better constrain the implications and suitability of adopting different constitutive models for assessing the stability of engineering works in fractured rock masses.