Skip to main content

About this book

This book reports on the German research initiative AeroStruct, a three-year collaborative project between universities and the aircraft industry. It describes the development of an integrated multidisciplinary simulation environment for aircraft analysis and optimization using high-fidelity methods. This system is able to run at a high level of automatism, thus representing a step forward with respect to previous ones. Its special features are: a CAD description that is independent from the disciplines involved, an automated CFD mesh generation and an automated structure model generation including a sizing process. The book also reports on test cases by both industrial partners and DLR demonstrating the advantages of the new environment and its suitability for the industry. These results were also discussed during the AeroStruct closing Symposium, which took place on 13-14 October 2015 at the DLR in Braunschweig, Germany. The book provides expert readers with a timely report on multidisciplinary aircraft design and optimization. Thanks to a good balance between theory and practice, it is expected to address an audience of both academics and professional, and to offer them new ideas for future research and development.

Table of Contents


Use-Case FORSwing


Chapter 1. Integrated Process Chain for Aerostructural Wing Optimization and Application to an NLF Forward Swept Composite Wing

This contribution introduces an integrated process chain for aerostructural wing optimization based on high fidelity simulation methods. The architecture of this process chain enables two of the most promising future technologies in commercial aircraft design in the context of multidisciplinary design optimization (MDO). These technologies are natural laminar flow (NLF) and aeroelastic tailoring using carbon fiber reinforced plastics (CFRP). With this new approach the application of MDO to an NLF forward swept composite wing will be possible. The main feature of the process chain is the hierarchical decomposition of the optimization problem into two levels. On the highest level the wing planform including twist and airfoil thickness distributions as well as the orthotropy direction of the composite structure will be optimized. The lower optimization level includes the wing box sizing for essential load cases considering the static aeroelastic deformations. Additionally, the airfoil shapes are transferred from a given NLF wing design. The natural laminar flow is considered by prescribing laminar-turbulent transition locations. Results of wing design studies and a wing optimization using the process chain are presented for a forward swept wing aircraft configuration. The wing optimization with 12 design parameters shows a fuel burn reduction in the order of 9% for the design mission.
Tobias Wunderlich, Lars Reimer

Chapter 2. Automated Structural Design of Composite Forward Swept Wings

This article describes the structural design process within a multidisciplinary environment. A forward swept wing configuration is considered where static divergence has to be avoided by using anisotropic properties of stiffened panels made of CFRP (carbon fibre reinforced plastic). The structural design includes parametric model generation and automated sizing of composite wings. An analytical formulation of stiffened panels is used to investigate different stiffener concepts, where fast analytical failure criteria are applied. The goal is to minimize weight and provide accurate deformations for a coupled process. A parametric study shows the flexibility of the approach as well as the validity of the design concept and the approach for bend twist coupling. Furthermore, the influence of neglecting the load redistribution due to the wings deformation on the wing mass is shown.
Sascha Dähne, Lars Heinrich

Chapter 3. Design Procedure for Optimum Fiber Composite Airframe Structures Within an Automated Multidisciplinary Design and Optimization Process

This paper presents a concept and setup of a design procedure for optimum fiber composite airframe structures within an automated multidisciplinary design and optimization process. The optimization procedure is based on the Three-Columns-Concept, using a state-of the-art Finite-Element-Method (FEM) software for structural analysis purposes embedded into a multidisciplinary optimization software. The optimization model is formulated for traditional civil aircraft designs, providing maximum design flexibility in order to tailor-made the characteristics of fiber composite materials, including aeroelastic tailoring. Therefore, the design domain includes both, geometry and laminae related design variables. In doing so, a strict separation of the design model from the analysis model is implemented. The evaluation model includes all important structural design objectives, like mass, deformations (e.g. wing bending and twisting), local and global buckling behavior (classical eigenvalue analysis), strength as well as damage tolerance. Aggregation methods are applied to reduce the tremendous number of constraints and to improve the robustness of the optimization process. Finally, optimization results of a forward swept wing configuration, optimized with respect to structural design objectives, are presented.
Michael Seibel

Chapter 4. Development of Flight Control Functions for Integration in Gust Load Simulations

Within the joint project AeroStruct, the Department of Flight Mechanics, Flight Control and Aeroelasticity at TU Berlin developed generic flight control modules for considering the influence of flight control functions in preliminary aircraft design. In order to improve the methods used for the design process, an automated tool chain was set up, that builts a flight-mechanical model of the airplane, calculates controller parameters according to consistent criteria and automatically implements flight control functions in a high fidelity simulation environment.
Alexander Hamann

Use-Case FlexCraft


Chapter 5. Introduction to Airbus Use-Case “FlexCraft”

Within the joint project AeroStruct Airbus studied a concept for best exploitation of wing or aircraft flexibility for aircraft performance purposes. An important aspect was to improve the collaboration between the disciplines like Aerodynamics, Loads or Structure. There was a lot of learning on requirements on fidelity and accuracy, but also about effects of uncertainties on the design. Multiple layer models of different fidelity helped to understand and optimize the design process. For the design case aerodynamic optimization was conducted on the basis of a surrogate model using a quite large number of parametric variations of shape and flow parameters. Loads and flutter studies were performed with advanced CFD methods like LFD (linear frequency domain). The design of a more flexible wing was evaluated from an overall aircraft point of view, in particular taking into account the drag and weight changes. All together the AeroStruct project provided not only a much better insight to potential future design but also delivered a better understanding of product behavior and discipline interactions.
Frank Theurich, Klaus Becker

Chapter 6. On Recent Advances in Industrial High-Fidelity Aeroelasticity

The paper shows for steady and unsteady tasks the usage of high-fidelity CFD-CSM interaction. While for steady scenarios CFD is the standard method for nowadays aerodynamic prediction, the level of detail in the structural model is often limited to linear FEM models with, by definition, rigid airfoils. Here the usage of multi-body models with nonlinear body motion and single-body nonlinear structural models is presented. For dynamic applications like flutter prediction, an approach which allows using purely unsteady CFD data, is presented. In this connection the combination of linearised CFD and unsteady data recycling results in a very efficient and accurate unsteady aerodynamic ROM.
Bernd Stickan, Frank Schröder, Sebastian Helm, Hans Bleecke

Chapter 7. New CFD Practices for Modelling High-Speed Flows

Computational Fluid Dynamics (CFD) is used extensively in modelling high-speed flows for commercial transport aircraft. Current production CFD has deficiencies in modelling specific flow features. In this paper unitary flow cases are considered for specific flow features such as boundary layers, vortical flow, flow separation and shock boundary layer interaction. In the paper new CFD practices in meshing and turbulence modelling are investigated in order to better represent these flow features. In order to measure improvements metrics are introduced which measure an improvement with respect to a reference (from wind tunnel experiment or LES computation). In the paper it is shown that application of new CFD practices give an improved prediction of boundary layer and vortical flow.
Jan Willem van der Burg, Matthias Lühmann, H. Jakob, J. Benton



Chapter 8. Application of Adjoint Based Optimization on a MALE Platform

Within the AeroStruct project an adjoint based optimization framework has been established based on Reynolds averaged Navier Stokes aerodynamics and FEM structural mechanics analysis using the same geometric model. A parametric geometry kernel (Descartes) has been developed on the basis of the CPACS dataformat as the core of the system. The system has been setup in a flexible and modular fashion such that components can be exchanged and the system can be applied to applications needing only a subset of the functionality. This article describes some of the integration details as well as applications with results.
Kolja Elssel, Kaare Sørensen, Ögmundur Petersson

Chapter 9. Treating Non-conforming Sensitivity Fields by Mortar Mapping and Vertex Morphing for Multi-disciplinary Shape Optimization

This study investigates the sensitivity filtering properties of the Mortar Mapping method and correlates it to the Vertex Morphing method in order to demonstrate the advantages of such a procedure in the context of shape optimization. It points out the importance of a common design control approach in a Multi-Disciplinary Optimization (MDO) environment. In particular, individual components of MDO have nonmatching interfaces when Fluid-Structure Interaction (FSI) problems are of interest. Since the numerical models of dissimilar discretizations deliver nonconforming sensitivity fields with respect to the design variables defined at their interfaces, the shape optimization of the common surfaces necessitates a third field which unifies the optimization variables and acts as a control field. This approach not only covers this necessity by facilitating the Mortar Mapping method but also reveals that such a procedure acts as a sensitivity filter similar to the Vertex Morphing method without altering the optimality of the solution.
Altuğ Emiroğlu, Roland Wüchner, Kai-Uwe Bletzinger

Chapter 10. Interfacing MSC Nastran to the CFD-Solver DLR-Tau for Unsteady FSI Analyses with Nonlinear Aircraft Structures

For the partitioned analysis of fluid structure interactions the coupling of fluid and structural analysis codes is a basic requirement. For this purpose the MSC Nastran solver provides a programming interface OpenFSI, which is used and extended here for a flexible usage. Target is the integration of this nonlinear structural solver into the parallel programming environment FlowSimulator in which the DLR-Tau code is used for fluid analysis. The developed component based architecture and implementation is explained in detail. Two applications show the characteristics of nonlinear behaviour of aircraft structures. A beam-like behaviour shows in the nonlinear aeroelastic loading case a stiffening due to larger deflections and rotations and a correct kinematic shortening of the wing-span. A tin-walled wing structure shows skin buckling, which reduces the overall stiffness of the wing and alters the flowfield locally. Both effects can not be reproduced by the classical linear theory.
Matthias C. Haupt, Klemens Lindhorst, Peter Horst

Use-Case DIMENSyon-P


Chapter 11. Numerical Computations of Isolated and Installed Engine Jet Flows

The integration of efficient high bypass ratio turbofans under the wing of transport aircraft configurations necessitates a very close coupling between the engine and the airframe. One concern arising from this is the possible increase of the noise emissions due to the interaction of the engine jet with the aircrafts’ high-lift system. A coupled aerodynamic-aeroacoustic assessment approach is necessary to help understand the underlying flow physics and also support the optimization of such integration scenarios. In the frame of a comprehensive parameter study, a best practice approach for the use of the DLR TAU-Code to investigate the complex aerodynamics of jet-flap interactions was devised which also ensures the usability of the RANS (Reynolds Averaged Navier–Stokes) results in a subsequent aeroacoustic analysis.
Arne Stürmer, Carlos Màrquez-Gutierrez, Peer Böhning

Chapter 12. Application of Reynolds-Stress-Models on Free Shear Layers

This contribution presents the application of the JHh-v2 and JHh-v3 Reynolds-Stress-Model on different test cases with free shear layers. Based on a two-dimensional mixing layer, the need for improvement of the JHh-v2 turbulence model for free shear layers is worked out. An additional sink term within the length-scale equation is implemented and calibrated, resulting in the JHh-v3 model. Simulations of practically relevant test cases with free shear layers are performed using the JHh-v2 and JHh-v3 turbulence models. For comparison, experimental data and results of simulations with the Menter-SST eddy-viscosity model are shown.
Tim Landa, René-Daniel Cécora, Rolf Radespiel

Chapter 13. Further Development of CAA Simulation for Isolated and Installed Nozzle Configurations

A hybrid RANS/CAA approach with stochastic source modeling is used in this work for noise prediction. Previously, this methodology has been successfully applied to isolated configurations of jet and airframe with different source models. For the prediction of jet noise, the source model as proposed by Tam and Auriault has been used for isolated nozzle configurations. A modeling approach for isolated airframe noise configurations is relying on the vorticity based Lamb vector source model. The CAA simulation of installed configurations with the combination of jet and airframe requires however consideration of both noise generation mechanisms simultaneously. Thus, a vorticity based ‘Eddy Relaxation’ source model has been recently proposed as further development of this methodology, which is expected to capture the installation effect.
Andrej Neifeld, Roland Ewert

Cross-Cutting Subjects


Chapter 14. Structural Optimization of 3D Wings Under Aerodynamic Loads: Topology and Shell

New methods in manufacturing and novel challenges and usages require the exploration of the potential of new wing designs. This is the goal of this paper. We propose novel computational methods for the robust optimization of wings under aerodynamic loads. We restrict the discussion to the optimization of the linear-elastic properties of wings concerning several load cases and with treatment on deformations and regularization. The degrees of freedom for the design itself are the interior structure of the wing leading to topology optimization aspects and the structure of the wing hull in terms of composite material. Thus, this paper aims at mathematical methods for topology optimization of the wing interior made of isotropic material, the optimization of orthotropic composite material in the wing hull and the proper treatment of practical deformation aspects and multiple loads in this context.
Volker Schulz, Roland Stoffel, Heinz Zorn

Chapter 15. Accelerating Unsteady CFD Simulations Using a Minimum Residual Based Nonlinear Reduced Order Modeling Approach

Reduced-order modeling is evaluated as a means to speed up unsteady computational fluid dynamics (CFD) simulations while maintaining the desired level of accuracy. In the reduced order modeling approach, proper orthogonal decomposition (POD) is applied to some computed response time history from a compressible, unsteady CFD solver to compute a set of orthogonal basis vectors. An approximate flow solution for the next time step is predicted by minimizing the unsteady flow solver residual in the space spanned by the POD basis. This is done by solving a non-linear least-squares problem. This approximate flow solution is then used to initialize the flow solver at this time step, aiming to reduce the number of inner iterations of the dual time stepping loop to convergence compared to the conventional choice of initializing with the previous time step solution or an extrapolation in time. This procedure is repeated for all following time steps. Results for the pitching LANN wing at transonic flow conditions show a more than twofold reduction in the number of inner iterations of the flow solver to convergence. Despite the overhead caused by evaluating the reduced-order model (ROM) at every time step, the method results in a 38% savings in computational time without compromising accuracy, thus improving the overall efficiency for unsteady aerodynamics applications. Finally, several means to further improve the performance are also discussed, including updating the POD basis after every new time step.
Matteo Ripepi, Stefan Görtz

Chapter 16. Surrogate-Based Aerodynamic Shape Optimization of a Wing-Body Transport Aircraft Configuration

Aerodynamic shape optimization driven by high-fidelity computational fluid dynamics (CFD) simulations is still challenging, especially for complex aircraft configurations. The main difficulty is not only associated with the extremely large computational cost, but also related to the complicated design space with many local optima and a large number of design variables. Therefore, development of efficient global optimization algorithms is still of great interest. This study focuses on demonstrating surrogate-based optimization (SBO) for a wing-body configuration representative of a modern civil transport aircraft parameterized with as many as 80 design variables, while most previous SBO studies were limited to rather simple configurations with fewer parameters. The freeform deformation (FFD) method is used to control the shape of the wing. A Reynolds-averaged Navier-Stokes (RANS) flow solver is used to compute the aerodynamic coefficients at a set of initial sample points. Kriging is used to build a surrogate model for the drag coefficient, which is to be minimized, based on the initial samples. The surrogate model is iteratively refined based on different sample infill strategies. For 80 design variables, the SBO-type optimizer is shown to converge to an optimal shape with lower drag based on about 300 samples. Several studies are conducted on the influence of the resolution of the computational grid, the number and randomness of the initial samples, and the number of design variables on the final result.
Zhong-Hua Han, Mohammad Abu-Zurayk, Stefan Görtz, Caslav Ilic

Chapter 17. A Method for the Calculation of Large Deformations in Aeroelastic Applications

A modal-based method that calculates the geometric nonlinear effects in the regime of large deformations of wing-like structures is applied to the modeling of a highly flexible 3D wingbox of high aspect ratio. The proposed method features higher-order stiffness terms and calculates the nodal deformation field not only by normal modes but also by additional modal components. In this way, a nonlinear force-displacement relationship and a geometrically nonlinear displacement field are accounted for. Static and dynamic results for the two configurations are presented together with results from a nonlinear finite element solver. The validations highlight the capability of the method to capture the nonlinear effects and demonstrates its power to model a 3D wingbox structure made of composite shell elements with anisotropic material characteristics.
Markus Ritter


Additional information

Premium Partner

image credits