Browsing by Author "Jones, Bevan W S"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Open Access
    An Algebraic Volume of Fluid Method for Strongly Coupled Spacecraft Fuel Slosh Modelling
    (2020) Jones, Bevan W S; Malan, Arnaud G
    The increase in the number of commercial space missions has resulted in the increased need for efficient and effective spacecraft designs. A key contributor to the accuracy of space vehicle simulation is the prediction of fuel slosh loads during in-orbit manoeuvres, particularly due to the large fuel-to-solid mass ratios involved. To this end, this thesis details a high resolution mathematical model capable of predicting the dynamic interaction between fuel slosh and the rigid structure of a spacecraft. The Volume of Fluid (VoF) method provides a framework in which Computational Fluid Dynamics (CFD) can be used to model the fluid dynamics of two phase fuel slosh in a mass conservative manner. To be applicable to industrial geometries, an unstructured finite volume median dual cell methodology is employed for spatial discretisation. This gives rise to the first novel contribution of this work, namely the development of a new volume conservative VoF initialisation method for arbitrary interfaces on unstructured meshes. The scheme, called the Arbitrary Grid Initialiser (AGI), is rigorously validated and proven conservative to machine precision [1]. An algebraic, as opposed to geometric, VoF advection method is used due to being similarly well suited to unstructured grids. Improvements to the algebraic VoF method is therefore the next contribution of this thesis; where the CICSAM [2] and HiRAC [3] VoF methods are improved, and the first conservative HiRAC method presented. The improved CICSAM and HiRAC methods are shown to be competitive with their geometric counterparts on unstructured grids while being mass conservative. Both CICSAM and HiRAC are then coupled (HiRAC for the first time) to a well balanced Continuum Surface Force (CSF) surface tension discretisation. The surface tension implementation, for which standard height functions are used, is shown to be well-balanced with an accuracy that compares favourably to existing methods. In the final part of the thesis, the complete spacecraft model is constructed. A numerical rigid body code is developed for this purpose, which can additionally track its orientation. The rigid body and fluid schemes are finally coupled together in a strong, stable, and partitioned manner using the Aitken's ∆2 method [4]. The model is demonstrated to be numerically stable for large liquid-to-solid ratios via a benchmark test case.
  • Thumbnail Image
    Item
    Restricted
    A ball-vertex approach to r-refinement for accuracy enhancements in CFD Calculations
    (2015) Jones, Bevan W S; Malan, Arnaud G; van Rooyen, Jacobus A
    Presented in this paper is an accuracy enhancing r-refinement scheme based on the ballvertex method [1, 2]. Due the success of the method in mesh movement, this work seeks to extend its use to error minimization for computational fluid dynamics calculations. To this end, element edges are loaded via error driven monitor functions. The latter are estimated from both field gradients and curvature. Boundary nodal movement is facilitated via the use of automated B`ezier surface reconstruction. The application study involves both analytical as well as industrial test-cases. The latter involves 2D and 3D transonic flow calculations. When compared with the mesh independence solution, an error reduction in the computed coefficient of lift and moment of 60% was achieved even on relatively coarse meshes and close to an order of magnitude on finer meshes. Finally, a mesh deformation, moving boundary, problem was completed to demonstrate duality as both a mesh optimisation and deformation tool.
  • Thumbnail Image
    Item
    Open Access
    Mesh adaptation through r-refinement using a truss network analogy
    (2015) Jones, Bevan W S; Malan, Arnaud G
    This project investigates the use of a truss network, a structural mechanics model, as a metaphor for adapting a computational fluid dynamics (CFD) mesh. The objective of such adaptation is to increase computational effi- ciency by reducing the numerical error. To drive the adaptation, or to give the scheme an understanding of accuracy, computational errors are translated into forces at mesh vertices via a so-called monitor function. The ball-vertex truss network method is employed as it offers robustness and is applicable to problems in both two and three dimensions. In support of establishing a state-of-the-art adaptive meshing tool, boundary vertices are allowed to slide along geometric boundaries in an automated manner. This is achieved via feature identification followed by the construction of 3rd order bezier surface patches over boundary faces. To investigate the ability of the scheme, three numerical test cases were investigated. The first comprised an analytical case, with the aim of qualitatively assessing the ability to cluster vertices according to gradient. The developed scheme proved successful in doing this. Next, compressible transonic flow cases were considered in 2D and 3D. In both cases, the computed coefficient of lift and moment were investigated on the unrefined and refined meshes and then compared for error reduction. Improvements in accuracy of at least 60% were guaranteed, even on coarse meshes. This is viewed as a marked achievement in the sphere of robust and industrially viable r-refinement schemes.