This paper describes a self-contained parallel fluid-structure interaction solver based on a finite volume discretisation, where a strongly coupled partitioned solution procedure is employed.The incompressible fluid flow is described by the Navier-Stokes equations in the arbitrary Lagrangian-Eulerian form, and the solid deformation is described by the Saint Venant-Kirchhoff hyperelastic model in the total Lagrangian form. Both the fluid and the solid are discretised in space using the second-order accurate cell-centred finite volume method, and temporal discretisation is performed using the second-order accurate implicit scheme. The method, implemented in open-source software OpenFOAM, is parallelised using the domain decomposition approach and the exchange of information at the fluid-solid interface is handled using global face zones. The performance of the solver is evaluated in standard two-and threedimensional cases and excellent agreement with the available numerical results is obtained.

Tapered-double cantilever-beam joints were manufactured from aluminium-alloy substrates bonded together using a single-part, rubber-toughened, epoxy adhesive. The mode I fracture behaviour of the joints was investigated as a function of loading rate by conducting a series of tests at crosshead speeds ranging from 3.33 x10 -6 m/s to 13.5 m/s. Unstable, (i.e. stick-slip crack) growth behaviour was observed at test rates between 0.1 m/s and 6 m/s, whilst stable crack growth occurred at both lower and higher rates of loading. The adhesive fracture energy, G Ic , was estimated analytically, and the experiments were simulated numerically employing an implicit finite-volume method together with a cohesive-zone model. Good agreement was achieved between the numerical predictions, analytical results and the experimental observations over the entire range of loading rates investigated. The numerical simulations were able very readily to predict the stable crack growth which was observed, at both the slowest and highest rates of loading. However, the unstable crack propagation that was observed could only be predicted accurately when a particular rate-dependent cohesive zone model was used. This crack-velocity dependency of G Ic was also supported by the predictions of an adiabatic thermal-heating model (ATM).

Correct calculation of stresses at the interface of bonded or otherwise joined materials plays a significant role in many applications. It is therefore, important that traction at the material interface is calculated as accurately as possible. This paper describes procedures that can be employed to achieve this goal by using centre-based finite volume method. Total traction at the interface is calculated by decomposing it into normal and tangential components, both being calculated at each side of the interface, and applying the continuity assumption. The way in which the traction approximation is achieved depends on calculation of tangential gradient of displacement at the interface. To this end, three different methods are proposed and validated against problems with known solutions. It was shown than all methods can be successfully used to simulate problems with multi-material domains, with the procedure based on finite area method being most accurate.

This paper describes a finite volume method for orthotropic bodies with general principal material directions undergoing large strains and large rotations.The governing and constitutive relations are presented and the employed updated Lagrangian mathematical model is outlined. In order to maintain equivalence with large strain total Lagrangian methods, the constitutive stiffness tensor is updated transforming the principal material directions to the deformed configuration. Discretisation is performed using the cell-centred finite volume method for unstructured convex polyhedral meshes. The current methodology is successfully verified by numerically examining two separate test cases: a circular hole in an orthotropic plate subjected to a traction and a rotating orthotropic plate containing a hole subjected to a pressure. The 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 variable material orientation and parallel processing.

No abstract

This paper describes the development and application of a frictionless contact stress solver based on the cell-centred finite volume method. The contact methodology, implemented in the open-source software OpenFOAM, is derived from the penalty method commonly used in finite element contact algorithms. The solver is verified on two benchmark tests using the available Hertzian analytical solutions.

No abstract

This paper explores the potential methods for evaluating a healing system for asphalt pavements. The healing system under investigation involves compartmented calcium-alginate fibres encapsulating an asphalt binder healing agent (rejuvenator). This system presents a novel method of incorporating rejuvenators into asphalt pavement mixtures. The compartmented fibres are used to distribute the rejuvenator throughout the pavement mixture, thereby overcoming some of the problems associated with alternate asphalt pavement healing methods, i.e., spherical capsules and hollow fibres. The asphalt healing efficiency methods to be evaluated in this paper include: (i) standard test methods for asphalt pavements, such as the Indirect Tensile Strength test and the 4 Point Bending Fatigue test; and (ii) alternative fracture tests such as the Semi Circular Bend test. The study employs fracture theory in order to evaluate the efficiency of the damage repair. The research findings demonstrate that including compartmented calcium-alginate fibres encapsulating a rejuvenator into an asphalt pavement mix does not significantly improve the healing properties of the asphalt pavement. Nevertheless, the findings indicate that, with further enhancement, compartmented calcium alginate fibres may present a promising new approach for the development of self-healing asphalt pavement systems. Additionally, the test results indicate that the 4 point bend fatigue test is the most suitable test for evaluating the performance of self healing asphalt pavements.