A non-intrusive geometrically nonlinear augmentation to generic linear aeroelastic models
Introduction
Next-generation passenger aircraft will need to meet much stricter emissions and fuel efficiency targets. This will undoubtedly accelerate the push towards increased used of advanced lighter materials to reduce operating weight and configurations with very high aspect ratio wings to reduce induced drag. Aeroelastic effects will thus dominate many aspects of wing design, although higher flexibility may also have positive effects in damping some external disturbances, such as gusts. As a result, airframe design processes may need to be revised (Palacios et al., 2014), and, in particular, geometrical-nonlinear effects may need to be accounted for in stages of the analysis where they are currently neglected. This has already been the case in more radical configurations, such as large solar-powered aircraft (Noll et al., 2007), where linear analysis yielded non-conservative results as they missed critical nonlinear aeroelastic couplings (Patil and Hodges, 2004). While fully-coupled nonlinear aeroelastic simulation is now available (Qiao et al., 2018), certification of a new air vehicle currently requires 100,000s of load case simulations (Kier, 2017), as it considers maneuvers and gust loads at different velocities and altitudes, and for a range of mass cases and configurations. Statistical methods may help reduce the total simulation burden, yet there is still a need for a computationally-efficient models that provide an acceptable level of fidelity to describe the vehicle dynamics. Furthermore, design of control strategies also requires reduced models, and standard approaches with linear models can fail under large wing deformations (Wang et al., 2018).
Two main approaches have been considered so far to construct computationally-efficient aeroelastic models with geometrical nonlinearity, namely, descriptions based on beam theory, and model-order reduction methods. Geometrically nonlinear beams have been extensively used to study these problems (see, for instance, the review by Afonso et al., 2017). While they have been very useful to understand key physics and explore the design space, they impose substantial limitations when detailed finite-element (FE) models are already available. Extraction of the equivalent beam properties is needed through either a homogenization process (Dizy et al., 2013) or a stick modeling approach (Riso et al., 2020). For large structures, it has been shown that either approach can lead to significant errors when compared to nonlinear simulations using built-up finite-element models (Palacios and Cea, 2019, Medeiros et al., 2020). Alternatively, system identification approaches have been proposed (Mignolet et al., 2013) that capture the nonlinear response of structure to construct reduced-order models (ROMs) for fast computations in nonlinear dynamic aeroelasticity (Medeiros et al., 2020). The main issue here lies in the prior calculation of a training database made of a large number of static nonlinear computations, which then acts as the model internal physics in the actual simulations. This approach also needs the full model to be suitable for nonlinear simulations, which could be a real constrain, e.g. on linear models calibrated against Ground Vibration Tests (GVT).
The goal of this work is to develop an efficient computational framework to study the geometrically nonlinear effects on already existing (linear) industrial-scale aeroelastic models, building on an alternative approach (Wang et al., 2015, Palacios and Cea, 2019) that does not require prior nonlinear computations on a full 3D finite-element model. Instead, the geometric layout of the structure and the coupling between modes in the intrinsic beam equations (Hodges, 2003) – that, by construction, take sectional velocities and internal forces as main variables – are employed. A feature of this approach is that mass and stiffness properties of the initial configuration require no updating as the model deforms. Furthermore, only quadratic terms of the main variables are needed to capture the complete space of geometric nonlinearities. As for the modeling of the aerodynamic forces, although 2-D airfoil unsteady aerodynamics are still often used for high-aspect-ratio wing aeroelasticity, 3-D effects have been shown to play an important role (Modaress-Aval et al., 2019) and will be considered here. The aerodynamic forces are obtained here from the Doublet-Lattice Method (DLM), which solves a linearization of the compressible, inviscid, unsteady flow equations in the 3-dimensional domain, and it is commonly used in linear aircraft aeroelastic analysis (Kim, 2019). It will be shown how key geometrically nonlinear effects (wing inextensionality, follower force effects), can be taken into account despite the linear – i.e. not updating with geometry – aerodynamic matrices obtained from the DLM. The Unsteady Vortex Lattice Method (UVLM) (Murua et al., 2012) is another potential method for geometrically-nonlinear aeroelastic simulations, as it fully accounts for large deformations both in the structure and the wake. However, it is substantially more computationally demanding and it is limited to incompressible conditions, therefore not appropriate for the typical flow regimes of commercial aircraft. The current description can also accommodate higher fidelity aerodynamic models, such as 3D panel methods (Kier, 2017) that include thickness effects, CFD-generated aerodynamic influence coefficient matrices (Güner et al., 2019), and it will be shown that it is suitable for an industrial loads and aeroelastic environment.
The structure of this paper is as follows. Section 2 presents the methodological approach proposed in this work; Section 3 shows static and dynamic aeroelastic results with three different models, all of which were originally built as linear MSC Nastran aeroelastic models; and Section 4 presents the main conclusions of the work.
Section snippets
Theoretical formulation
The starting point in the formulation is an arbitrarily complex FE model for which structural dynamic condensation techniques are employed to reduce the 3-dimensional components into nodes along all major load paths (a skeleton-like substructure). The resulting linear normal modes (LNMs) of the reduced structure are evaluated in intrinsic variables (velocities and strains) (Hodges, 2003) and used as a basis of a Galerkin-projection of the geometrically-nonlinear intrinsic beam equations in
Numerical studies
Three numerical studies are included to verify our implementation and demonstrate the suitability of the proposed approach to solve geometrically-nonlinear aeroelastic problems. Firstly, nonlinear vibrations are numerically investigated on a representative wing-box undergoing large displacements to explore the impact of different condensation techniques. The second example shows the nonlinear aeroelastic equilibrium and the dynamic response of a simple high aspect ratio wings, where results are
Conclusions
A modal-based description has been presented that seamlessly introduces geometrically-nonlinear effects on generic models built for linear dynamic analysis. The proposed approach preserves the linear solution, which is then augmented with the nonlinear effects in beam theory, namely geometric stiffening, follower forces and changes of inertia properties. To achieve this, firstly, a condensed structure along the main load-path of the full 3D model is obtained using techniques of modal
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This research is sponsored by an Industrial Cooperative Award in Science and Technology, United Kingdom between the UK Engineering and Physical Sciences Research Council and Airbus Group Innovation, United Kingdom .
References (36)
- et al.
A review on non-linear aeroelasticity of high aspect-ratio wings
Prog. Aerosp. Sci.
(2017) - et al.
Homogenisation of slender periodic composite structures
Int. J. Solids Struct.
(2013) - et al.
Geometrically nonlinear aeroelastic behavior of pretwisted composite wings modeled as thin walled beams
J. Fluids Struct.
(2018) - et al.
Unsteady aerodynamic modeling methodology based on dynamic mode interpolation for transonic flutter calculations
J. Fluids Struct.
(2019) Flutter prediction methodology based on dynamic eigen decomposition and frequency-domain stability
J. Fluids Struct.
(2019)- et al.
A review of indirect/non-intrusive reduced order modeling of nonlinear geometric structures
J. Sound Vib.
(2013) - et al.
A comparative study of nonlinear aeroelastic models for high aspect ratio wings
J. Fluids Struct.
(2019) - et al.
T-tail flutter: Potential-flow modelling, experimental validation and flight tests
Prog. Aerosp. Sci.
(2014) - et al.
Applications of the unsteady vortex-lattice method in aircraft aeroelasticity and flight dynamics
Prog. Aerosp. Sci.
(2012) - et al.
On the importance of aerodynamic and structural geometrical nonlinearities in aeroelastic behavior of high-aspect-ratio wings
J. Fluids Struct.
(2004)
Limit-cycle oscillations in high-aspect-ratio wings
J. Fluids Struct.
Nonlinear aeroelastic characteristics analysis of composite wing with high aspect ratio based on co-rotational method
J. Fluids Struct.
A method for normal-mode-based model reduction in nonlinear dynamics of slender structures
Comput. Struct.
Nonlinear modal aeroservoelastic analysis framework for flexible aircraft
AIAA J.
DTIC ADP010476: Aeroservoelastic modeling, analysis, and design techniques for transport aircraft
Numerical analysis of store-induced limit-cycle oscillation
J. Aircr.
Unsteady aerodynamic forces mixed method for aeroservoelasticity studies on an F/A-18 aircraft
J. Aircr.
Nonlinear folding wing tips for gust loads alleviation
J. Aircr.
Cited by (8)
A novel reduced-order method using mixed nonlinear kinematics for geometrically nonlinear analysis of thin-walled structures
2024, Computer Methods in Applied Mechanics and EngineeringAeroelastic reduced-order modeling for efficient static aeroelastic analysis considering geometric nonlinearity
2024, Journal of Fluids and StructuresDual audible-range band gaps in three-dimensional locally resonant phononic crystals
2023, Physical Review BImpact of Low-Order Modeling on Aeroelastic Predictions for Very Flexible Wings
2023, Journal of AircraftGeometrically Nonlinear Effects on the Aeroelastic Response of a Transport Aircraft Configuration
2023, Journal of Aircraft
- 1
Both authors contributed equally to the manuscript.