This paper presents a parametric reduced-order model (ROM) based on manifold learning (ML) for use in steady transonic aerodynamic applications. The main objective of this work is to derive an efficient ROM that exploits the low-dimensional nonlinear solution manifold to ensure an improved treatment of the nonlinearities involved in varying the inflow conditions to obtain an accurate prediction of shocks. The reduced-order representation of the data is derived using the Isomap ML method, which is applied to a set of sampled computational fluid dynamics (CFD) data. In order to develop a ROM that has the ability to predict approximate CFD solutions at untried parameter combinations, Isomap is coupled with an interpolation method to capture the variations in parameters like the angle of attack or the Mach number. Furthermore, an approximate local inverse mapping from the reduced-order representation to the full CFD solution space is introduced. The proposed ROM, called Isomap+I, is applied to the two-dimensional NACA 64A010 airfoil and to the 3D LANN wing. The results are compared to those obtained by proper orthogonal decomposition plus interpolation (POD+I) and to the full-order CFD model.
This article gives an overview of reduced order modeling work performed in the DLR project Digital-X. Parametric aerodynamic reduced order models (ROMs) are used to predict surface pressure distributions based on high-fidelity computational fluid dynamics (CFD), but at lower evaluation time and storage than the original CFD model. ROMs for steady aerodynamic applications are built using proper orthogonal decomposition (POD) and Isomap, a manifold learning method. Approximate solutions in the so obtained low-dimensional representations of the data are found with interpolation techniques, or by minimizing the corresponding steady flow-solver residual. The latter approach produces physics-based ROMs driven by the governing equations. The steady ROMs are used to predict the static aeroelastic loads in a multidisciplinary design and optimization (MDO) context, where the structural model is to be sized for the (aerodynamic) loads. They are also used in a process where an a priori identification of the critical load cases is of interest and the sheer number of load cases to be considered does not lend itself to high-fidelity CFD. An approach to correct a linear loads analysis model using steady CFD solutions at various Mach numbers and angles of attack and a ROM of the corrected Aerodynamic Influence Coefficients (AICs) is also shown. This results in a complete loads analysis model preserving aerodynamic nonlinearities while allowing fast evaluation across all model parameters. The different ROM methods are applied to a 3D test case of a transonic wing-body transport aircraft configuration. Keywords reduced order model • proper orthogonal decomposition • isomap • manifold learning • multidisciplinary design and optimization • aerodynamic influence coefficients • loads analysis • CFD
Reduced-order models (ROMs) become increasingly popular in industrial design and optimization processes, since they allow to approximate expensive high fidelity computational fluid dynamics (CFD) simulations in near real-time. The quality of ROM predictions highly depends on the placement samples in the spanned parameter space. Adaptive sampling strategies allow to identify regions of interest, which feature e.g. nonlinear responses with respect to the parameters, and therefore enable the sensible placement of new samples. By introducing more samples in these regions, the ROM prediction accuracy should increase. In this contribution we investigate different adaptive sampling strategies based on cross-validation, Gaussian mean-squared error, two methods exploiting the CFD residual and a two manifold embedding methods. The performance of those strategies is evaluated and measured by their ability to successfully identify the regions of interest and the resulting sample placement in terms of different quantitative statistical values. We further discuss the reduction of the ROM prediction error over the adaptive sampling iterations and show that depending on the adaptive sampling strategy, the number of required samples can be reduced by 35–44% without deteriorating model quality compared to a Halton sequence sampling plan.
SummaryReduced‐order models (ROMs) are more and more considered for use in aerodynamic applications. Benefits of these methods can be expected for optimization problems or predicting aerodynamic loads for the entire flight envelope. For these applications it is often possible to perform computations for various parameter combinations before any ROM evaluations are needed. The order reduction of the CFD solutions in this article is done using proper orthogonal decomposition. Coupled with an interpolation method predictions for unknown parameter combinations can be made. The CFD solutions are computed using a discontinuous Galerkin finite element method combined with a nonlinear multigrid scheme. The nonlinear multigrid solver algorithms depend on a good initial guess of the flow field to be computed. Typically, the initial guess on a fine mesh is obtained by solving the problem on a agglomerated mesh. Alternatively, an initial guess could be obtained from a ROM prediction, if available. The main objective is to identify the benefits of using a ROM in a higher‐order multigrid environment. Integral as well as distributed surface quantities of the ROM predictions originating from several multigrid levels will be compared with fully converged flow solutions on the top level of the multigrid algorithm. Furthermore, initializing the flow solver with predictions from several multigrid levels will be analyzed in comparison to full multigrid computations as well as to a scenario where already converged solutions for similar parameter combinations are used as initial flow solutions on the top multigrid level.
Reduced-order models (ROMs) based on proper orthogonal decomposition (POD) are widely used in industry. Due to the rigid requirements on the input data, these methods struggle with discontinuous parameters, e.g., optional rear spoiler on a car. In order to also include these types of parameters, a new method is presented that splits the full-order model (FOM) domain with its discontinuous parameters into multiple ROM subdomains. The resulting subdomains then again comply with the ROM requirements, and the established and proven ROM methods can be applied. The steps involved in computing a ROM based on the proposed method, by setting up the subdomains, mapping the FOM data into the domains, as well as computing the ROMs on the domains, are shown in detail in this paper. The method is employed on two use cases. The academic one-dimensional use case focuses on how the steps involved are employed and analyzes the introduced errors. The second use case’s FOM is based on the DrivAer body with an optional rear spoiler computed using computational fluid dynamics (CFD) and demonstrates the usage in an industrial environment.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.