130 results found
Venkatesan-Crome C, Palacios R, Kattmann T, et al., 2019, Discrete Adjoint for Unsteady Incompressible Flows Using a Density-based Formulation, Publisher: American Institute of Aeronautics and Astronautics
Bao Y, Zhu HB, Huan P, et al., Numerical prediction of vortex-induced vibration of flexible riser with thick strip method, Journal of Fluids and Structures, ISSN: 0889-9746
We present numerical prediction results of vortex-induced vibration (VIV) of a long flexible tensioned riser subject to uniform currents. The VIV model of long length-to-diameter ratio is considered and ‘thick’ strip technique based on high-order spectral/hp element method is employed for computational simulation. The model parameter of the riser for the simulation is chosen according to the dimensional counterparts used in the experimental tests in Lehn (2003). The numerical results are displayed in terms of motion responses, hydrodynamic forces and wake patterns as well and compared and discussed with the available data in the literature.
Maraniello S, Palacios R, State-space realizations and internal balancing in potential-flow aerodynamics with arbitrary kinematics, AIAA Journal, ISSN: 0001-1452
This paper presents a general state-space realization of the unsteady vortex-latticemethod together with a bespoke model-order reduction strategy. The aim is to providea computationally-efficient aerodynamic description suitable for integration in aeroelasti-city with arbitrary kinematics. The state-space realization is obtained from linearization,around arbitrary geometries and static loading conditions, of lifting surfaces and their freewakes. All components of the forces are evaluated in time domain using Joukowski’s the-orem. As the wake is also modelled, however, the state-space description has a large num-ber of states. High-order modelling of the apparent mass, projection on low-dimensionaldegrees-of-freedom, and balanced residualization are combined to reduce the system dimen-sionality. A parallelized low-rank square root algorithm for balancing is also introduced toreduce the computational cost. Numerical investigations on airfoils and cantilever wingslinearised around non-zero reference conditions are used to exemplify the approach.
Qi P, Zhao X, Palacios R, 2019, Preview-based Altitude Control for a Very Flexible Flying Wing with Lidar Wind Measurements, 57th IEEE Conference on Decision and Control (CDC), Publisher: IEEE, Pages: 4289-4294, ISSN: 0743-1546
Venkatesan-Crome C, Carrusca Gomes P, Palacios R, 2019, Optimal Compliant Airfoils Using Fully Non-Linear FSI Models, Publisher: American Institute of Aeronautics and Astronautics
Cea A, Palacios R, 2019, Nonlinear Modal Aeroelastic Analysis from Large Industrial-Scale Models, Publisher: American Institute of Aeronautics and Astronautics
Del Carre A, Palacios R, 2019, Efficient Time-Domain Simulations in Nonlinear Aeroelasticity, Publisher: American Institute of Aeronautics and Astronautics
Qi P, Zhao X, Palacios R, 2019, Autonomous Landing Control of Highly Flexible Aircraft based on Lidar Preview in the Presence of Wind Turbulence, IEEE Transactions on Aerospace and Electronic Systems, Pages: 1-1, ISSN: 0018-9251
Palacios Nieto R, Cea A, 2018, Nonlinear modal condensation of large finite element models: An application of Hodges’ intrinsic theory, AIAA Journal, ISSN: 0001-1452
A method to obtain geometrically nonlinear reduced-order descriptions of structures defined by a large finite element model is presented. The full model is used to identify all the coefficients in a modal projection of Hodges’s intrinsic beam equations, with the geometric reduction introduced through static or dynamic condensation along the main load paths on the original structure. The only information retrieved from the full model is the linear normal modes as well as condensed mass and stiffness and nodal coordinates. The approach aims to solve geometrically nonlinear problems of industrial complexity in an efficient manner, while preserving the linear model under small displacements. Examples of increasing complexity will be shown with the built-up finite element models made with beams and shells and both lumped and distributed masses. Nonlinear static and dynamic analyses, including rigid-body dynamics, are then demonstrated using the resulting nonlinear modal description.
Qi P, Zhao X, Wang Y, et al., 2018, Automatic Landing Control of a Very Flexible Flying Wing, Pages: 2599-2604, ISSN: 0743-1619
© 2018 AACC. This paper investigates the landing control of a highly flexible flying wing model. An automatic landing control system is developed based on the reduced-order linear model, which employs a two-loop control scheme. The outer loop employs the LADRC (linear active disturbance rejection control) and PI control algorithms to track the reference landing trajectory and vertical speed respectively, and generate the attitude command based on which the inner loop uses H∞ control to compute the control inputs to the corresponding control surfaces. A landing trajectory generator is designed to generate real-time commands for the landing control system with a pre-compensator introduced to improve the dynamic tracking performance. Simulation results based on the full-order nonlinear model show that the automatic landing control system is able to land the flying wing effectively and safely, showing good performances in tracking and robustness against the wind disturbances.
Wang Y, Wynn A, Palacios R, 2018, Nonlinear aeroelastic control of very flexible aircraft using modelupdating, Journal of Aircraft, Vol: 55, Pages: 1551-1563, ISSN: 0021-8669
A nonlinear modeling strategy is introduced to drive online optimization-based controllers for trajectory tracking and stabilization of very flexible aircraft. This is achieved thanks to a compact residualized modal projection of the aeroelastic equations of motion, including large (geometrically nonlinear) wing deflections, that is used to describe the vehicle dynamics. This study shows that a model-predictive control system can then be built with an internal model based on successive linearization of the equations of motion around the instantaneous vehicle geometry. Significant improvements in stability against large disturbances and maneuverability are demonstrated on numerical simulations of a very flexible flying wing when geometrically nonlinear effects are included in the internal model of the controller.
Maraniello S, Palacios R, 2018, State space realisation and model reduction of potential-flow aerodynamics for HAWT applications, 7th Conference on Science of Making Torque from Wind (TORQUE), Publisher: Institute of Physics (IoP), ISSN: 1742-6588
This paper presents a general state-space realisation of the unsteady vortex-lattice method and combines it with a novel model-order reduction strategy. The aim is to provide a computationally efficient aerodynamic description suitable for integration in horizontal axis wind turbine aeroelasticity. The consistent linearisation is in the 3D components of the vortexlattice geometry. The resulting linear-time invariant system can, therefore, resolve all the component of forces, hence being suitable for linearisation around arbitrary wake shapes and blade geometries/deformations. The wake modelling captures unsteady aerodynamic effects but it results in large state-space models. Projection on low-dimensional degrees-of-freedom and balanced residualisation are, therefore, employed to reduce the model dimensionality. An iterative balancing algorithm based on Smith’s method is also developed so as to contain the computational cost of the process. The paper also presents an initial numerical investigation on aerofoils linearised around non-zero reference conditions, showing that this approach can reduce the size of the problem by several orders of magnitude and at a lower computational cost than standard direct methods for system balancing.
Palacios R, Venkatesan-Crome C, Sanchez R, Aerodynamic Optimization using FSI Coupled Adjoints in SU2, 6th European Conference in Computational Mechanics
Qi P, Zhao X, Wang Y, et al., 2018, Aeroelastic and Trajectory Control of High Altitude Long Endurance Aircraft, IEEE Transactions on Aerospace and Electronic Systems, ISSN: 0018-9251
IEEE We investigate the aeroelastic and trajectory control of an High Altitude Long Endurance (HALE) aircraft model in the presence of gust and turbulence disturbances. The model is derived from geometrically-nonlinear beam theory using intrinsic degrees of freedom and linear unsteady aerodynamics, which results in a coupled structural dynamics, aerodynamics and flight dynamics description. The control design employs a two-loop PI/LADRC (linear active disturbance rejection control) and H<formula><tex>$\infin$</tex></formula> control scheme in both the longitudinal and lateral channels, based on a reduced-order linearised model. In each channel, the outer loop (position control) employs a PI/LADRC technique to track the desired flight routes and generate attitude command to the inner loop, while the inner loop (attitude control) uses H<formula><tex>$\infin$</tex></formula> control to track the attitude command generated from the outer loop and computes the control inputs to the corresponding control surfaces. A Particle Swarm Optimization algorithm is employed for parameter optimization of the weighting matrices in the H<formula><tex>$\infin$</tex></formula> control design. The simulation tests conducted on the full-order nonlinear model show that the aeroelastic and trajectory control system achieves good performance with respect to robustness, trajectory tracking and disturbance rejection
Broughton-Venner JJ, Wynn A, Palacios R, 2018, Aeroservoelastic Optimisation of Aerofoils with Compliant Flaps via Reparameterization and Variable Selection, 58th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference, Publisher: AMER INST AERONAUTICS ASTRONAUTICS, Pages: 1146-1157, ISSN: 0001-1452
Bao Y, Palacios R, Graham M, et al., 2018, A strip modelling of flow past a freely vibrating cable, ERCOFTAC Series, Vol: 24, Pages: 221-227, ISSN: 1382-4309
© 2018, Springer International Publishing AG. Vortex-induced vibration of long flexible structures with cylindrical cross-section are widely encountered in various engineering fields.
Palacios R, Buoso S, Dickinson B, 2017, Bat-inspired integrally actuated membrane wings with leading-edge sensing, Bioinspiration and Biomimetics, Vol: 13, ISSN: 1748-3182
This paper presents a numerical investigation on the closed-loop performance of a two-dimensional actuated membrane wing with fixed supports. The proposed concept mimics aerodynamic sensing and actuation mechanisms found in bat wings to achieve robust outdoor flight: firstly, variable membrane tension, which is obtained in bats through skeleton articulation, is introduced through a dielectric-elastomer construction; secondly, leading-edge airflow sensing is achieved with bioinspired hair-like sensors. Numerical results from a coupled aero-electromechanical model show that this configuration can allow for the tracking of prescribed lift coefficient signals in the presence of disturbances from atmospheric gusts. In particular, disturbance measurements through the hair sensor (a feedforward control strategy) are seen to provide substantial advantage with respect to a reactive (feedback) control strategy determining a reduction of the oscillations of the lift coefficient.
Sanchez R, Albring T, Palacios R, et al., 2017, Coupled Adjoint-Based Sensitivities in Large-Displacement Fluid-Structure Interaction using Algorithmic Differentiation, International Journal for Numerical Methods in Engineering, Vol: 113, Pages: 1081-1107, ISSN: 0029-5981
A methodology for the calculation of gradients with respect to design parameters in general Fluid-Structure Interaction problems is presented. It is based on fixed-point iterations on the adjoint variables of the coupled system using Algorithmic Differentiation. This removes the need for the construction of the analytic Jacobian for the coupled physical problem, which is the usual limitation for the computation of adjoints in most realistic applications. The formulation is shown to be amenable to partitioned solution methods for the adjoint equations. It also poses no restrictions to the nonlinear physics in either the fluid or structural field, other than the existence of a converged solution to the primal problem from which to compute the adjoints. We demonstrate the applicability of this procedure and the accuracy of the computed gradients on coupled problems involving viscous flows with geometrical and material non-linearities in the structural domain.
Qi P, Wang Y, Zhao X, et al., 2017, Trajectory Control of a Very Flexible Flying Wing, American Control Conference (ACC), Publisher: IEEE, Pages: 4480-4485, ISSN: 0743-1619
Maraniello S, Palacios R, Optimal rolling manoeuvres with very flexible wings, AIAA Journal, Vol: 55, Pages: 2964-2979, ISSN: 1533-385X
The single-shooting method is used to identify optimal manoeuvres in the lateral dyna-mics of partially-supported flexible wings. Efficient actuation strategies are sought whenlarge wing deflections substantially modify the geometry of the vehicle during the man-oeuvre. For that purpose, geometrical nonlinear models are first built using compositebeams and an unsteady vortex lattice, and the optimal control problem is then solved via agradient-based algorithm. A flight-dynamics model based on elastified stability derivativesis shown to capture the relevant dynamics either under slow actuation or for stiff wings andit is used as a reference. Embedding the full aeroelastic description into the optimisationframework expands the space of achievable manoeuvres, such as quick wing response withlow structural vibrations or large lateral forces with minimal lift losses.
Buoso S, Palacios R, 2017, On-demand Aerodynamics in Integrally Actuated Membranes with Feedback Control, AIAA Journal, Vol: 55, Pages: 377-388, ISSN: 1533-385X
This paper is a numerical investigation on model reduction and controlsystem design of integrally actuated membrane wings. A high-fidelityelectro-aeromechanical model is used for the simulation of the dynamicfluid-structure interaction between a low-Reynolds-number flow and a dielectricelastomeric wing. Two reduced-order models with different levels ofcomplexity are then derived. They are based on the projection of the fullorderdiscretisation of fluid and structure on modal shapes obtained fromeigenvalue analysis and Proper Orthogonal Decomposition. The low-ordersystems are then used for the design of Proportional-Integral-Derivative andLinear Quadratic Gaussian feedback schemes to control wing lift. When implementedin the full-order model, closed-loop dynamics are in very goodagreement with the reduced-order model for both tracking and gust rejection,demonstrating the suitability of the approach. The control lawsselected in this work were found to be effective only for low-frequency disturbancesdue to the large phase delay introduced by the fluid convectivetime-scales, but results demonstrate the potential for the aerodynamic controlof membrane wings in outdoor flight using dielectric elastomers.
Sanchez R, Palacios R, Economon TD, et al., 2017, Optimal actuation of dielectric membrane wings using high-fidelity fluid-structure modelling
© 2017, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. This paper describes a computational framework for the analysis and design of electromechanically actuated membrane wings operating at low flight speeds. A fluid-structure interaction formulation with large deformations and complex material behavior has been developed, which is suited for integrally-actuated wings built with dielectric elastomers. A coupled adjoint-based methodology based on algorithmic differentiation has also been developed to determine optimal actuation profiles. Coupled sensitivities are shown to be accurately computed on the wing response, both under constant pressure and immersed in a reattached laminar flow. A simple optimization problem on the equilibrium position of the membrane is finally solved to exemplify the proposed design methodology.
Maraniello S, Palacios R, 2017, Geometrically-nonlinear effects in lateral manoeuvres with coupled flight dynamics and aeroelasticity
© 2017, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. An investigation on the aeroelastic effects in lateral manoeuvres with very flexible wings is presented. The aim is to identify efficient actuation strategies from fully coupled non- linear aeroelastic/flight dynamics models, which account for potential large wing deections to improve vehicle manoeuvrability. The flexible vehicle dynamics is described using geometrically-exact composite beams on a body-attached frame and an unsteady vortex lattice with arbitrary kinematics of the lifting surfaces, while rolling manoeuvres are identified through optimal control. A flight-dynamics model based on elastified stability derivatives is used as a reference, and it is observed to capture the relevant dynamics either under slow actuation or for stiff wings. Embedding the full aeroelastic description into an optimal control framework is shown to expand the space of achievable manoeuvres, such as quick wing response with low structural vibrations or large lateral forces with minimal lift losses. It is also seen to provide a general methodology to identify unconventional manoeuvres that utilize large wing geometry changes to meet multiple simultaneous control objectives.
Palacios R, Simpson RJS, Maraniello S, 2017, State-space realizations of potential-flow unsteady aerodynamics with arbitrary kinematics
© 2017, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. We introduce a nondimensional state-space formulation of the unsteady vortex-lattice method for time-domain aerodynamics. It deals with 2- or 3-dimensional geometries, re- solves frequencies up to a spatio-temporal Nyquist limit defined by the wake discretization, and has a convenient form for linearization, model reduction and coupling with structural dynamics models. No assumptions are made relating to the kinematics of the fluid-structure interface (inputs) and use of Joukowski's theorem to compute forces naturally resolves all components of the unsteady aerodynamic forcing (outputs). Linearized expressions are written about arbitrary non-zero reference geometries, velocities and loading distributions and as such yield models that are as general as possible given the assumptions in the un- derlying uid mechanics. The implementation is verified against classical solutions in the unsteady aerodynamics, and in aeroelastic stability analysis of cantilever wing configurations.
Ng BF, New TH, Palacios R, 2017, Bio-inspired leading-edge tubercles to improve fatigue life in horizontal axis wind turbine blades
© 2017, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. Bio-inspired leading-edge tubercles are known to improve aerodynamic performances during stall but there could be additional advantages in other operating regimes and their effect on aeroelastic loadings is also less understood. In this study, the effect of leading-edge tubercles on fatigue loadings on wind turbine blades is investigated using an aeroelastic model that couples a composite beam to the unsteady vortex-lattice method. To accommodate the leading-edge tubercles, spanwise structural properties and aerodynamic geometries are varied, which resulted in a reduction in both torsional frequencies and aerodynamic lift despite keeping the same planform area. The reductions in structural frequencies and aerodynamics have counteracting effects on fatigue responses, with the former increasing and the latter reducing loads. On a turbine blade with tubercles (amplitude of 0.2c and wavelength of 0.5c) occupying 20% to 95% span of the leading-edge, flapwise root-bending moment was found to be 6% lower than the unmodified configuration, which can be further enhanced with a trough termination at the blade tip. Torsional moment was 17% lower due to the reduction in leading-edge suction along tubercle crests, which are further away from the elastic axis. In terms of tubercle positioning, having tubercles close to the blade tip enables performance enhancement during episodes of stall from large tip deflections and has significant contributions to root-bending moment due to a larger moment arm and higher relative flow speed. On the other hand, positioning tubercles close to the blade root may also be favoured as this region is prone to stall from low speeds, yet having little effect on fatigue responses.
González-Salcedo A, Aparicio-Sanchez M, Munduate X, et al., 2017, A computationally-efficient panel code for unsteady airfoil modelling including dynamic stall
© 2017, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. A new approach based on inviscid panel methods has been developed for airfoils undergoing unsteady kinematics. The model is tailored to practical applications related to wind turbines, taking into account accuracy and computational cost considerations. It is able to represent rigid or deformable airfoils, attached and separated flow focusing also on unsteady cases including dynamic stall. An extensive validation has been carried out including steady and unsteady conditions, while simulations of airfoils with trailing edge flaps have also been performed.
Sanchez R, Palacios R, 2017, Computing derivatives in nonlinear aeroelasticity using algorithm differentiation
Copyright 2018, IADC/SPE Drilling Conference and Exhibition. This paper introduces a new method to calculate design sensitivities in high-fidelity fluid-structural interactions problems. As the intended application is on aeroelastic systems, characterized by many design variables and a small number of objective functions, gradients are computed from the fully-coupled adjoint equation. This is cast here as an iterated sequence for which convergence is guaranteed. The system is finally obtained using algorithm differentiation on the fully-coupled primal problem and its solution is then sought using a block Gauss-Seidel method, which preserves the partitioned structure of the primal coupled solver. This solution architecture has been implemented in the open-source SU2 software suite for the solution of industrial-scale aeroelastic optimization problems with viscous fluids and structures with nonlinear geometric and material response.
Palacios R, Simpson RJS, Maraniello S, 2017, State-space realizations and model reduction of potential-flow unsteady aerodynamics with arbitrary kinematics
Copyright 2018, IADC/SPE Drilling Conference and Exhibition. We introduce a nondimensional state-space formulation of the unsteady vortex-lattice method for time-domain aerodynamics. It deals with 2- or 3-dimensional geometries, resolves frequencies up to a spatio-temporal Nyquist limit defined by the wake discretization, and has a convenient form for linearization, model reduction and coupling with structural dynamics models. No assumptions are made relating to the kinematics of the fluid-structure interface (inputs) and use of Joukowski’s theorem to compute forces naturally resolves all components of the unsteady aerodynamic forcing (outputs). Linearized expressions are written about arbitrary non-zero reference geometries, velocities and loading distributions and as such yield models that are as general as possible given the assumptions in the underlying fluid mechanics. The implementation is verified against classical solutions in the unsteady aerodynamics, and in aeroelastic stability analysis of cantilever wing configurations.
Sanchez R, Kline HL, Thomas D, et al., 2016, Assessment of the fluid-structure interaction capabilities for aeronautical applications of the open-source solver SU2, ECCOMAS Congress 2016, Publisher: Institute of Structural Analysis and Antiseismic Research, Pages: 1498-1529
We report on an international effort to develop an open-source computational environmentfor high-fidelity fluid-structure interaction analysis. In particular, we will focus onverification of the implementation for application in computational aeroelasticity. The capabilitiesof the SU2 code for aeroelastic analysis have been further enhanced both by developingnatively embedded tools for the study of largely deformable solids, and by wrapping it usingPython tools for an improved communication with external solvers. Both capabilities will bedemonstrated on relevant test cases, including rigid-airfoil solutions with indicial functions,the Isogai Wing Section, test cases from the AIAA 2nd Aeroelastic Prediction Workshop, andthe vortex-induced vibrations of a flexible cantilever in the wake of a square cylinder. Resultsshow very good performance both in terms of accuracy and computational efficiency. The modularityand versatility of the baseline suite allows for a flexible framework for multidisciplinarycomputational analysis. The software libraries have been freely shared with the community toencourage further engagement in the improvement, validation and further development of thisopen-source project.
Broughton-Venner JJ, Wynn A, Palacios R, 2016, Aeroservoelastic optimisation of an aerofoil with active compliant flap via reparametrisation and variable selection, 17th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, Publisher: AIAA
To aid in the investigation of new simultaneous optimisation strategies for exible vehicles and their control systems, a two-dimensional aerofoil optimisation which demands minimal computational effort is studied. The aeroservoelastic system consists of a two-dimensional, potential flow over a deforming aerofoil; an actively controlled, but saturated compliant trailing edge; a dynamic observer that uses a series of pressure sensors on the aerofoil; and a heave/pitch linear spring model. Although computationally simple, the design allows for optimisation over multiple disciplines: the structure can be designed by varying the stiffness of the springs; the control architecture through weightings in a LQR controller; the observer by means of the placement of pressure sensors; and the aerodynamics via the shaping of the compliant trailing edge. Optimising the weight and a metric of performance over all these fields simultaneously is compared to a sequential methodology of optimising the open-loop characteristics first and subsequently adding a closed-loop con-troller. Parametrisation of the design vector and variable selection often require user input and are fixed during optimisation. Our research aims to automate this process. Further-more, we investigate whether varying the parametrisation and number of design variables during the optimisation can lead to improvements in the final design. To accomplish this, a new basis for the design vector is created via Proper Orthogonal Decomposition (POD) using the trajectories of initial optimisation paths as a “training set". This parametrisation is shown to make the optimisation more robust with respect to the initial design, and facilitate an automated variable selection methodology. This variable selection allows for the dimension of the problem to be reduced temporarily and it is shown that this makes the optimisation more robust.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.