An integrative design-through-analysis paradigm for higher-order computational aerodynamics and aeroelasticity

点击翻译按钮获取中文摘要

To guarantee safety and fine-tune performance of next-generation light-weight aircraft, integrating aerodynamic and aeroelastic simulations early in the design process is essential. This necessitates the transfer of computer-aided geometric designs into physics-based computer simulations, currently based on geometry processing and mesh generation methods that are often time-consuming, error-prone and require frequent manual intervention. This constitutes a considerable barrier to creating rapid design-through-analysis workflows, with significant negative impact on cost and turnaround times of aircraft design processes. The overarching hypothesis of the proposed project is that a new integrative design-through-analysis paradigm based on parametric geometry modeling, isogeometric structural analysis, and embedded domain computational fluid dynamics can significantly improve the integration of design and simulation. While parametric modeling techniques are available in many commercial design tools and isogeometric analysis is currently growing into a mature simulation methodology, embedded domain CFD remains plagued by serious technology gaps that prevent its reliable and efficient application for high-fidelity aerodynamic simulations. Prominent problems include the detrimental impact of elements with small cuts on conditioning and time step size, low-order accuracy of quadrature rules in cut elements, sub-optimal near-boundary accuracy, and algorithms that are unsuitable for emerging heterogeneous computing architectures. From a technology viewpoint, the objective of the proposed project is therefore to devise and test a range of new methods that can effectively close these gaps and remove long-held concerns of the computational sciences community against embedded domain CFD. These improvements constitute the basis for a new class of next-generation embedded domain CFD technologies that enable robust, efficient and higher-order accurate aerodynamic simulations at very large scales.From the design-through-analysis perspective, the overarching goal of the proposed project is to employ the envisioned integrative paradigm, spurred by the improved embedded domain CFD technologies, in such a way that a single engineer is enabled to conceive, create, analyze, and interpret a large ensemble of simulations in a time-critical period. This is demonstrated by aerodynamic and aeroelastic simulations of a series of challenging aeronautics and turbomachinery scenarios, showcasing the potential of the integrative framework for further automating design-through-analysis workflows at an industry scale.