Aerostructural optimization of a morphing wing for airborne wind energy applications

Abstract

Airborne wind energy (AWE) vehicles maximize energy production by constantly operating at extreme wing loading, permitted by high flight speeds. Additionally, the wide range of wind speeds and the presence of flow inhomogeneities and gusts create a complex and demanding flight environment for AWE systems. Adaptation to different flow conditions is normally achieved by conventional wing control surfaces and, in case of ground generator-based systems, by varying the reel-out speed. These control degrees of freedom enable to remain within the operational envelope, but cause significant penalties in terms of energy output. A significantly greater adaptability is offered by shape-morphing wings, which have the potential to achieve optimal performance at different flight conditions by tailoring their airfoil shape and lift distribution at different levels along the wingspan. Hence, the application of compliant structures for AWE wings is very promising. Furthermore, active gust load alleviation can be achieved through morphing, which leads to a lower weight and an expanded flight envelope, thus increasing the power production of the AWE system. This work presents a procedure to concurrently optimize the aerodynamic shape, compliant structure, and composite layup of a morphing wing for AWE applications. The morphing concept is based on distributed compliance ribs, actuated by electromechanical linear actuators, guiding the deformation of the flexible—yet load-carrying—composite skin. The goal of the aerostructural optimization is formulated as a high-level requirement, namely to maximize the average annual power production per wing area of an AWE system by tailoring the shape of the wing, and to extend the flight envelope of the wing by actively alleviating gust loads. The results of the concurrent multidisciplinary optimization show a 50.7% increase of extracted power with respect to a sequentially optimized design, highlighting the benefits of morphing and the potential of the proposed approach.

Structural analysis and optimization of a tethered swept wing for airborne wind energy generation

“…Three‐dimensional solid or shell‐based finite element analysis of such composite structures requires detailed knowledge of geometry, properties of the laminate material, and fibre orientation, and moreover is computationally expensive. Particular to AWE, there has been some recent work on the analysis and optimisation of rigid composite kites, with the focus on morphing wings . However, this kind of detailed analysis, given the computational costs involved, is uncommon in the initial iterative design stage.…”

“…Particular to AWE, there has been some recent work on the analysis and optimisation of rigid composite kites, with the focus on morphing wings. 9 However, this kind of detailed analysis, given the computational costs involved, is uncommon in the initial iterative design stage. An alternative methodology would be to model the kite structure as an equivalent composite beam, while accounting for the unconventional wing-box topologies by determining characteristic 2D cross-sectional stiffness while taking into account the varied shape and composite lay-up along the span.…”

Aero-structural optimization and analysis of a camber-morphing flying wing: Structural and wind tunnel testing

“…Owing to the utilization of electromechanical actuators, the concept possesses much more scalability than the previously used piezoelectric actuators. Therefore, it can be envisioned to utilize this concept for larger aircraft, subject to higher flight speeds and higher aerodynamic loading (Fasel et al, 2017;Keidel et al, 2017). Stronger electromechanical or hydraulic actuators would be required to achieve sufficiently large deflections and control moments.…”

Data-driven nonlinear aeroelastic models of morphing wings for control

“…A linear state-space model is usually preferred for response analysis and control design [45][46][47]. In the field of AWE, the benchmark problem for the method in this paper, researchers have applied several time-domain models, ranging from lifting line methods [48] to quasi-steady approximations of a 3D panel method, based on source and doublet distributions [11]. However, none of these approaches considered the full unsteadiness of the flow.…”