Surrogate modeling plays a crucial role in connecting high-fidelity data generation techniques with everyday engineering tasks. These surrogate models act as quick substitutes for more expensive, high-fidelity methods, albeit with some loss in accuracy. The success of these models largely hinges on the quality of the data used to train them. This data should offer enough information to educate the surrogate model about the crucial aspects within the domain of interest. However, as problems become more intricate and involve higher dimensions, it becomes increasingly challenging to determine in advance the ideal data distribution necessary to accurately capture the non-linearities in the domain. Consequently, employing adaptive training methods to pinpoint where additional data is required becomes critical for producing a high-quality surrogate model. This effort focuses on demonstrating the application of adaptive training to enhance the underlying data distribution used for training a machine learning-based surrogate model. The primary goal here is to reduce the time required to obtain a solution, rather than aiming for a perfectly optimized data distribution.

The rapid prediction of loads for engineering applications is of high interest in the aerodynamic community. An approach for obtaining these rapid predictions is through the use of surrogate modeling. Surrogate models enable a faster turn-around time for various engineering needs that high-fidelity computational models cannot accommodate. A machine learning (ML) framework to support surrogate modeling of integrated aerodynamic loads predictions for aircraft is investigated in this effort. The ML framework includes core Deep Neural Network (DNN) components built to support surrogate models for both steady and unsteady aerodynamics. A vital aspect of a successful surrogate model is the prediction accuracy when non-linear flow phenomena such as flow separation and transonic effects impact the flowfield. For highly separated flows – such as dynamic stall – a two-step, novel physics-state predictor approach is laid out. The dual framework employs an intermediate physics-state predictor that enables the surrogate model to more accurately model dynamic separated and non-linear flow patterns. Parameter studies are conducted to investigate the impact of different input feature sets on the surrogate model's predictive capabilities. A discussion on the potential benefits of incorporating Convolutional Neural Network (CNN) based DNN architectures for the surrogate model is also included. Initial 2D/3D verification results for steady and unsteady aerodynamic problems are presented. The eventual goal is to develop trusted surrogate models for real-time system assessments supporting Digital Engineering initiatives.

Results for the First AIAA Stability and Control Prediction Workshop using the HPCMP CREATE(TM)-AV Kestrel KCFD solver are presented. The workshop geometry is the Common Research Model, with the addition of a vertical tail designed by ONERA. Results are presented for Navier-Stokes predictions, using RANS turbulence models, for a mesh-independence study at Mach 0.83, as well as a Mach sweep for Mach 0.70, 0.83, 0.85, 0.87 and 0.90. Mesh independence is shown for CRM configurations with and without the windtunnel sting. Good mesh independence is obtained for all integrated forces and moments except for pitching moment. Computational results agree well with experimental forces and moments for low Mach. Pitching moment is least accurate, especially at the higher Mach conditions. Comparison of computed surface pressure at several wing and tail locations indicates that KCFD predictions for shock location are likely reasons for the disagreement between computations and experiment.

The application of hypersonic flow simulation tools to realistic flight scenarios will require the coupling of multiple physical effects to the baseline fluid dynamics. Such multiphysics effects can include the aerooelastic response of the airframe or engine components, dynamic transport of atmospheric particles, the deformation of solid-fluid interfaces that can ablate, pyrolyze, or erode, as well as a host of other processes, all of which are governed by unique sets of physical equations and models. Coupling multiple (and potentially disparate) physics solvers to a robust compressible flow solver poses additional challenges related to the stability, performance and scalability of the combined solver. The choices made during the software design process can therefore lead to a variation in simulation efficiency across different computer architectures. In this paper, we will consider two representative multiphysics hypersonic flow scenarios: the interaction of solid particulates with the flow field created by a hypersonic lifting body and the aerooelastic deformation of a model airframe under high-Mach-number flow conditions. For these simulations we explore the behavior of several hypersonic simulation tools, including Kestrel, FUN3D, US3D, and the JENRE(R) Multiphysics Framework, on several high performance computing systems containing various CPU and GPU architectures.

It is critical to understand how hypersonic simulation tools perform on a range of computational platforms. This information will aid in the acquisition of appropriate hardware and the potential refactoring of hypersonic codes to run on different systems. In this paper, we consider two representative high-speed reacting flow cases: a model Mach 8 hypersonic waverider glide vehicle and a model hydrocarbon-fueled hypersonic ramjet propulsion system. In both scenarios, the flow fields are in chemical non-equilibrium and are modeled by the multi-species reacting Navier-Stokes equations. For these simulations we use several hypersonic simulation tools, including US3D, Kestrel, FUN3D, and the JENRE(R) flow solver. We explore several high performance computing systems containing Intel(R) Xeon(R) Platinum processors, AMD EPYC(TM) 7702 processors, and NVIDIA(R) Tesla V100 devices. We compare performance and strong scaling between the different systems.

View Video Presentation: https://doi.org/10.2514/6.2021-0234.vid Kestrel is a multidisciplinary simulation tool in the HPCMP CREATE(TM) program that couples aerodynamics, thermochemistry, stability and control, structures, propulsion, and store separation for a large range of freestream operating conditions. It provides a robust and accurate capability targeting fixed-wing aircraft and is being used extensively in government and industry organizations within the DoD acquisition community. This paper presents a summary of key production features introduced during the 2020 calendar year (roughly v10.3 to v11.1) as well as some emerging code capabilities. One of the more notable among these new features is the introduction of a new workflow targeting multiphysics trajectory problems by generalizing the frequency and manner in which different solvers are coupled in time. A discussion of ongoing code credibility activities is also included.

Kestrel is an integrating product in the HPCMP CREATE(TM) program that allows crossover between simulation of aerodynamics, thermochemistry, dynamic stability and control, structures, propulsion, and store separation. It provides a robust and accurate multidisciplinary simulation capability targeting fixed-wing aircraft and is being used extensively in the DoD aircraft acquisition process. This paper presents a summary of recent code capabilities that have entered into production via multiple feature releases over the last year as well as emerging features that will be available in future releases.

The opinions of design, test and analysis engineers from industry and government laboratories on the status and future of stability and control prediction are presented in this paper. Air transport, fighter and flying wing configurations, considered to be representative of the platforms of concern to a vast majority of the commercial and military industries, are discussed. Static and dynamic stability as well as control effectiveness are used to describe the stability and control characteristics. Achievements necessary to advance the current status are decomposed into relatively straightforward and technologically challenging obstacles to overcome. Finally, a roadmap to address these obstacles is presented.

This computational aerodynamics textbook is written at the undergraduate level, based on years of teaching focused on developing the engineering skills required to become an intelligent user of aerodynamic codes. This is done by taking advantage of CA codes that are now available and doing projects to learn the basic numerical and aerodynamic concepts required. This book includes a number of unique features to make studying computational aerodynamics more enjoyable. These include:The computer programs used in the book's projects are all open source and accessible to students and practicing engineers alike on the book's website, www.cambridge.org/aerodynamics. The site includes access to images, movies, programs, and moreThe computational aerodynamics concepts are given relevance by CA Concept Boxes integrated into the chapters to provide realistic asides to the conceptsReaders can see fluids in motion with the Flow Visualization Boxes carefully integrated into the text.

This computational aerodynamics textbook is written at the undergraduate level, based on years of teaching focused on developing the engineering skills required to become an intelligent user of aerodynamic codes. This is done by taking advantage of CA codes that are now available and doing projects to learn the basic numerical and aerodynamic concepts required. This book includes a number of unique features to make studying computational aerodynamics more enjoyable. These include:The computer programs used in the book's projects are all open source and accessible to students and practicing engineers alike on the book's website, www.cambridge.org/aerodynamics. The site includes access to images, movies, programs, and moreThe computational aerodynamics concepts are given relevance by CA Concept Boxes integrated into the chapters to provide realistic asides to the conceptsReaders can see fluids in motion with the Flow Visualization Boxes carefully integrated into the text.

The F-16XL flight vehicle has been used extensively to study the aerodynamics of medium- to high-angle-of-attack flight. Flight-test data, including surface pressures, are available for a wide range of flight conditions, allowing direct comparison between computational fluid dynamics and full-scale flight data. The most recent comprehensive study, made possible by NASA, was conducted by NATO Task Group AVT-113 and had participants from many different countries. One of the conclusions of the study was that unsteady flow simulations with high-resolution turbulence treatment was necessary to compare well with the high-angle-of-attack flight-test data. The current work applies the high performance computing CREATE (TM)-Air Vehicles Kestrel fixed-wing simulation tool to the F-16XL configuration at high-angle-of-attack flight conditions. Kestrel couples a near-body spatially second-order solver with an offbody, spatially third- or fifth-order solver to achieve a very high-resolution computation of the aerodynamic features of the flowfield. Adaptive mesh refinement is also performed in the offbody domain as the solution progresses to maintain the vortical structures away from the body. The resulting simulation data are compared to flight-test data, and they are found to be in very good agreement for this flight condition.