Search

Fan vaulting is a magnificent innovation of English Gothic architecture that has been admired since the 14th century. However, how this breathtaking structural form works under mechanical loads (like self-weight and support displacement) is still poorly understood. One of the many interesting open issues is the mechanical role of backfill. This paper studies how the existence and the height of the backfill affect the horizontal and vertical reactions expressed on the fan vault by the tas-de-charge and the lateral longitudinal wall. In order to analyse this, discrete element models are prepared for small-span and large-span fan vaults constructed of jointed masonry, with Voronoi-shaped elements forming the backfill, and then the structures are submitted to self-weight and gradually increasing quasi-static outwards support displacements. The ratio of horizontal versus vertical reactions is applied to quantify how stable these vaults are in the case of different levels of backfill. The effect of shell thickness is also addressed. [ABSTRACT FROM AUTHOR]

This study focuses on an old but still unresolved problem of automatically calibrating the constitutive parameters of discrete element models. Instead of the troublesome and time-consuming manual trial-and-error method, which is typical today, the authors suggest using artificial intelligence techniques. A masonry arch is analysed, whose experimental static load–displacement behaviour is known from the literature. An attempt is made to match this behaviour with discrete element models, through finding appropriate quantitative values for the parameters. Two methods (Genetic Algorithm (GA) and Particle Swarm Optimisation (PSO)) are tested and, since PSO turns out to be more reliable, a further improved version, 'Trust-Based Particle Swarm Optimisation' (TBPSO), is proposed. The results show that (1) TBPSO quickly leads to suitable alternative parameter sets that make the discrete element model match the behaviour of the real experiments and (2) the optimal values of the parameters strongly depend on the loading velocity and the discretisation method used. [ABSTRACT FROM AUTHOR]

An interlocking block is a concave polyhedron with non-planar joints connecting the blocks together. The possibility of the fracture within a masonry interlocking block is a major challenge that has remained rather unexplored yet. Different fracture scenarios can be taken into account through considering the crack planes at which the block can be set apart. The plastic failure inside the block can also be represented through the continuum plastic deformation of the block composed of continuum finite elements. For an interlocking block with a cuboid projection above (lock), this paper intends to analyse the torsion–shear behaviour of the lock experimentally and numerically based on the discrete element method. Two strategies are developed to model a concave block: the lock and main body of an interlocking block are set to be rigid and connected with a cohesive contact in between; the concave interlocking polyhedron is set to be deformable with elasto-plastic behaviour. Given the same material properties, the torsion–shear capacities of the lock obtained by the two numerical models and the experimental test are compared to each other. A parametric analysis is then provided to calibrate the deformable model. [ABSTRACT FROM AUTHOR]

The paper addresses the question of whether the number of particles in a noncemented granular assembly will affect the mechanical characteristics of the assembly: its strength and stiffness. The question is answered by applying the discrete element method to assemblies of different sizes. To isolate the effect of assembly size, apart from the scatter that usually accompanies such simulations, multiple assemblies were tested. The two-dimensional assemblies had nearly identical initial porosities and fabrics, and they were all loaded in biaxial compression. Two different boundaries were tested: periodic and wall boundaries. We find that the peak compressive strength decreases with assembly size for both types of boundaries and over a range of assembly sizes that contain 256 particles to over 66,000 particles. Stiffness is only slightly reduced and only with wall boundaries. Deformation is less uniform in the larger assemblies, with deformation concentrated in a smaller fraction of the assembly area. An analysis of deformation patterning shows that at least a few thousand particles are required for realistic microband patterning. CE Database subject headings: Granular media; Size effect; Discrete elements; Particles; Localization; Simulation.

The paper presents a definition of rolling between a pair of two-dimensional or three-dimensional particles with a compliant contact. The definition of rolling movement is based upon the shapes of the objects' surfaces as described with differential geometry. A pseudoinverse of the surface curvatures is used for producing a rolling vector that is tangent to the two objects at their contact. Matrix expressions are presented for the efficient computation of the vector. The roiling vector is objective and is independent of the reference points that are used to track the particle motions. The definition of rolling is applied in a discrete element method simulation of the triaxial compression of three large, dense cubic assemblies: one packing of spherical particles and two packings of nonspherical particles. At small strains, particle rolling was slightly less with the nonspherical particles, but the packing with the greatest coordination number had much less rolling. CE Database subject headings: Granular materials; Contrasts; Kinematics; Deformation.

The paper presents a definition of particle rolling for the interactions of two and three-dimensional particles of arbitrary shape, in case of infinitesimal particle translations and rotations. The definition is based on a purely kinematical analysis, and it is shown to satisfy the objectivity condition.