ALBERT

Notes: The failure of rock mass under loading is resulting from preexisting flaws, such as cracks,pores and other defects. However, the propagation and coalescence mechanism among multi-groupcracks is still a puzzle, especially to the engineering rocks in site. In this study, the failure of rocksamples with two groups of preexisting parallel cracks under the axial load were numericallyinvestigated by the Rock Failure Process Analysis code (RFPA) from a mechanics point of view. Thesimulated results reproduce the rock failure process: at the first loading stage, the particle is stressedand energy is stored as elastic strain energy with a few randomly isolated fractures. As the loadincreases, the isolated fractures are localized to form a macroscopic crack. At the peak load, theisolated fractures unstably propagate in a direction parallel to the loading direction following tortuouspaths and with numerous crack branches. Finally, the major crack passes through the rock sample andseveral coarse progeny cracks are formed. Moreover, in the vicinity of the contacting zone the localcrushing is always induced to cause fines. On the base of the simulated results, it is found that thedominant breakage mechanisms are catastrophic splitting and progressive crushing. It is pointed outthat the particle breakage behavior strongly depends on the heterogeneous material property, theirregular shape and size, and the various loading conditions. Because of heterogeneity, the crackpropagates in tortuous path and crack branching becomes a usual phenomenon. The failure process ofrock sample demonstrated that due to the high stress concentration at the cracks tip or some weakerstrength elements which are not on the cracks surface initiate some micro-fractures, those cracks andfractures may gradually become larger and larger, more and more with the progress of loading so thatjoin into the branch cracks leading to the rock failure in the end. Not only did the output of thenumerical simulation study compare well with the experiment results, but also the further insights ofthe mechanism of cracks propagation and coalescence process in rock mass were obtained

Notes: The periodically distributed fracture spacing phenomenon exists in the failure process ofthe reinforced concrete prism under uniaxial tension. In this paper, A numerical code RFPA3D (3DRealistic Failure Process Analysis) is used to simulate the three-dimensional failure process of plainconcrete prism specimen and reinforced concrete prism specimen under uniaxial tension. Thereinforced concrete is represented by a set of elements with same size and different mechanicalproperties. They are uniform cubic elements and their mechanical properties, including elasticmodulus and peak strength, are distributed through the specimens according to a certain statisticaldistribution. The elastic modulus and other mechanical properties are weakened gradually when thestresses in the elements meet the specific failure criterion. The displacement-controlled loadingscheme is used to simulate the complete failure process of reinforced concrete. The analyses focus onthe failure mechanisms of the concrete and reinforcement. The complete process of the fracture forthe plain concrete prism and the fracture initiation, infilling and saturation of the reinforced concreteprism is reproduced. It agrees well with the theoretical analysis. Through 3D numerical tests for thespecimen, it can be investigated the interaction between the reinforcement and concrete mechanicalproperties in meso-level and the numerical code is proved to be an effective way to help thoroughlyunderstand the rule of the reinforcement and concrete and also help the design of the structuralconcrete components and systems

Notes: A series of numerical model tests were performed to investigate the behaviour of theanisotropic rock surrounding circular excavations under high confining pressures. The aim was toprovide information on the formation of fractures and failure around deep level rock tunnels undercontrolled conditions. Solid cubes containing a circular hole were confined to a vertical pressure withsame as the confinement in the horizontal directions. In this modeling, the inhomogeneous rock isgenerated by using Weibull parameters which are related to the microstructural properties determinedby crack size distribution and grain size. The fracture angle is assumed to be 45o. The observed failurezone around the excavation was simulated using both the maximum tensile strain criterion andMohr-Coulomb criterion respectively (as the damage threshold). And RFPA (Realistic FailureProcess Analysis) code was used as the calculating tool in this modelling, three opening modes aresimulated and compared. Computational model predictions that include crack propagation and failuremodes of rock show a good agreement with those of the observation in site. It is pointed out that thedamage evolution of EDZ strongly depends on the inhomogeneous, the excavation mode, anisotropicproperty, and the various loading conditions. Concerning the existence of a weak plane, the amount ofdisplacement at the side wall of the tunnel was quite large, since the shear deformation occurred inEDZ. The model is implemented in RFPA code and is able to represent the change in fracture patternsbetween the solid and jointed parts. This provides confidence for the application of the numericalmodel to the design of rock tunnels at great depth

Notes: Numerical simulation of the blanking process of a sheet metal, i.e. the formation ofroll-over and a burnished surface, and crack initiation and propagation, is done using an elastic plasticfinite element model. The Normalized Cockcroft and Latham’s expression is used as a ductile fracturecriterion. Numerical study was conducted to investigate how the tool-wear affects the burr height byvarying the punch nose radius. The simulation results obtained under the effect of the same processvariables are in good agreement with the experimental results. Six punches with various nose edges,0.00mm, 0.05mm, 0.10mm, 0.20mm, 0.30mm and 0.40mm were used for blanking simulations. Theresults indicate that the burr height is directly proportional to the punch nose radius

Notes: Six types of numerical specimens containing two notches are set up to numericallyinvestigate the effect of element size on rock shear strength and failure pattern using RFPA2D (rock failure process analysis) code. These specimens are of the same geometrical dimension 180 mm×180 mm and have been discretized into 61×61, 122×122, 183×183, 244×244, 305×305, and 366×366 elements.The width of notches is about 2.95 (180/61) mm and the length is 45mm. The specimens are placed in a direct shear box. A lateral confining pressure with a value of 0.15MPa is invariably loaded in the vertical direction and an increasing horizontal displacement with 0.002mm/step is applied in the horizontal direction. The whole shear failure progress and associated stress field for the specimens are visually represented. Results show that the crack propagation is mostly influenced by the stress field in the vicinity of the notch tip, the required element size would be necessary in order to obtain good results. In general, for a coarse mesh, the stress field close to the notch tip can’t be represented accurately and shear strength obtained by such discretisation is slightly higher than the accurate value. For a fine mesh, the notch tip spreads through a relatively large number of elements and the stress field in vicinity of notch tip is well represented by the finite element approximation, therefore the failure pattern is consistent with real physical fracture mode

Notes: Based on the experimental results of uniaxial time-dependent ratcheting behavior ofSS304 stainless steel at room temperature and 973K, three kinds of time-dependent constitutivemodels were employed to describe such time-dependent ratcheting by using the Ohno-Abdel-Karimkinematic hardening rule, i.e., a unified viscoplastic model, a creep-plasticity superposition modeland a creep-viscoplasticity superposition model. The capabilities of such models to describe thetime-dependent ratcheting were discussed by comparing with the corresponding experimentalresults. It is shown that the unified viscoplastic model cannot provide reasonable simulation to thetime-dependent ratcheting, especially to those with certain peak/valley stress hold and at 973K; theproposed creep-plasticity superposition model is reasonable when the creep is a dominant factor ofthe deformation, however, it cannot provide a reasonable description when the creep is weak; thecreep-viscoplastic superposition model is reasonable not only at room temperature but also at hightemperature

Notes: Equally spaced opening-mode fractures always evolve in top layer attached to underlying layer. With a newly developed Material Failure Process Analysis code (MFPA2D), we have firstly investigated the stress distribution between two adjacent fractures as a function of the fracture-spacing-to-layer-thickness ratio using a two-layer model with a fractured top layer. The numerical results indicate the horizontal stress perpendicular to the fractures near the top surface changes from tensile to compressive when the fracture-spacing-to-layer-thickness ratio changes from greater than to less than a critical value. Then, the process from fracture initiation to fracture saturation is numerically modeled. The modeling of fracture process shows that the fractures initiate at the top surface and propagate to the interface between the two layers in the first stage. In the following stage, new fractures can infill between the earlier formed fractures and they always initiateat the interface and propagate to the top surface. Numerical simulation clearly demonstrates that the stress state transition precludes further infilling of fractures and the fracture spacing reaches a constant state, i.e. the so-called fracture saturation

Notes: Reinforced concrete structures are generally designed to allow cracking under service loading. Accurate modeling of crack formation and propagation at lower load levels is therefore important. In this paper, a Material Failure Process Analysis code (MFPA2D) is used to model the crack initiation and propagation in reinforced concrete bridge pier subjected to eccentric loading. In our numerical model, the reinforced concrete is assumed to be a three-phase composite composed of concrete, reinforcement and interfaces between them. Numerically obtained results of cracking loads and global load-displacement response agree well with experimentally measured values. It has been found that the fracture of the concrete observed at the macroscopic level is predominated by tensile damage at the mesoscopic level

Notes: A numerical code RFPA3D (Realistic Failure Process Analysis) is used to simulate the crackinitiation and propagation in FRP-strengthened concrete beam under external loading. In our model,the FRP-strengthened concrete is assumed to be a three-phase composite composed of concrete, FRP,and interface between them. The displacement-controlled loading scheme is used to simulate thecomplete failure process of FRP-strengthened concrete the numerical simulation of failure process ofthe specimens. It is found that the main failure mode is the interfacial debonding and the interfacialdebonding may propagate either within the adhesive layer or through concrete layer in the vicinity ofbond interface. The simulation results agree well with the experiment observations. The width of theFRP sheet is considered an important factor not only to significantly influence the debondingpropagation type and crack distribution but also to control the ultimate load-capacity and ultimatestrain. This study is focused on the failure process of the FRP-strengthened concrete beam and theeffects of the width of FRP sheet on the failure mode and on the structural load-carrying capacity ofconcrete structures