Development of Aerodynamic Optimisation

我对流体动力学建模的兴趣是在我的学位和随后的工业就业期间发展起来的，现在已经成熟到我可以在该领域规划研究生学习的程度。CFD本身是一个成熟的领域，因此我的意图是利用CFD进入形状优化的下一个前沿领域。关于使用CFD设计最佳几何形状的激动人心的问题在许多方面仍未得到解答。例如，基于流体的几何优化的两个最基本的限制是形状的拓扑结构通常保持不变，并且定义形状的参数集在分辨率方面不会改变。因此，我的目标将集中在探索拓扑自由参数化，也允许几何细化。当前感兴趣的一个领域是实体参数化的体积。将使用这些参数化，以便在优化过程中添加额外的设计变量，从而使该方法具有充分的范围来探索各种形状。我们的目标是创造一种工具，能够自动生成性能更好的新空气动力学设计。实现这一目标的工具部分存在于布里斯托大学内部，以内部CFD工具的形式存在，以及预先存在的参数化开发历史。现有的CFD工具将与新的参数化相结合，并具有几何细化的能力。最后一步将是使这些方法为工业界的从业者所使用，以便它们可以很容易地应用于现实世界的问题。我对流体动力学建模的兴趣是在我的学位和随后的工业就业期间发展起来的，现在已经成熟到我可以在该领域规划研究生学习的程度。CFD本身是一个成熟的领域，因此我的意图是利用CFD进入形状优化的下一个前沿领域。关于使用CFD设计最佳几何形状的激动人心的问题在许多方面仍未得到解答。例如，基于流体的几何优化的两个最基本的限制是形状的拓扑结构通常保持不变，并且定义形状的参数集在分辨率方面不会改变。因此，我的目标将集中在探索拓扑自由参数化，也允许几何细化。当前感兴趣的一个领域是实体参数化的体积。将使用这些参数化，以便在优化过程中添加额外的设计变量，从而使该方法具有充分的范围来探索各种形状。我们的目标是创造一种工具，能够自动生成性能更好的新空气动力学设计。实现这一目标的工具部分存在于布里斯托大学内部，以内部CFD工具的形式存在，以及预先存在的参数化开发历史。现有的CFD工具将与新的参数化相结合，并具有几何细化的能力。最后一步将是使这些方法为工业界的从业者所使用，以便它们可以很容易地应用于现实世界的问题。

My interest in fluid dynamic modelling developed during my degree and subsequent industrial employment, and has now matured to the extent that I feel comfortable planning postgraduate study within the field. CFD itself is a mature field, and my intention is therefore to move in to the next cutting-edge area of shape optimisation using CFD. Exciting questions relating to designing optimal geometries using CFD remain unanswered in many respects. For example, two of the most fundamental restrictions to fluid-based geometric optimisation are that the topology of the shape usually remains unchanged, and also that the parameter set defining the shape does not change in terms of resolution. My objectives will therefore be focused on exploring topology free parameterisations that also permit geometric refinement. One current area of interest is volume of solid parameterisations. These parameterisations will be used so that additional design variables may be added during the optimisation process, and so that the method has full scope to explore a wide variety of shapes. The goal will be to create tools which are able to automatically generate new aerodynamic designs with improved performance. Tools for achieving this exist in part within the University of Bristol, in the form of in-house CFD tools, as well as a pre-existing history of developing parameterisations. The existing CFD tools will be coupled to new parameterisations, and the capability to geometrically refine will be included. The final step will be to make these methods accessible to practitioners in industry so that they can be readily applied to real-world problems. My interest in fluid dynamic modelling developed during my degree and subsequent industrial employment, and has now matured to the extent that I feel comfortable planning postgraduate study within the field. CFD itself is a mature field, and my intention is therefore to move in to the next cutting-edge area of shape optimisation using CFD. Exciting questions relating to designing optimal geometries using CFD remain unanswered in many respects. For example, two of the most fundamental restrictions to fluid-based geometric optimisation are that the topology of the shape usually remains unchanged, and also that the parameter set defining the shape does not change in terms of resolution. My objectives will therefore be focused on exploring topology free parameterisations that also permit geometric refinement. One current area of interest is volume of solid parameterisations. These parameterisations will be used so that additional design variables may be added during the optimisation process, and so that the method has full scope to explore a wide variety of shapes. The goal will be to create tools which are able to automatically generate new aerodynamic designs with improved performance. Tools for achieving this exist in part within the University of Bristol, in the form of in-house CFD tools, as well as a pre-existing history of developing parameterisations. The existing CFD tools will be coupled to new parameterisations, and the capability to geometrically refine will be included. The final step will be to make these methods accessible to practitioners in industry so that they can be readily applied to real-world problems.