Determination of optimal wing twist pattern for a composite digital wing

Recent advancements in the field of metamaterials with tunable properties have led to a renewed interest in developing the adaptive aerospace structures. In earlier research, a pair of cellular composite digital wings were developed, and wind tunnel tested. One of the novelties of the developed digital material wings is that they are internally actuated and that they can be effectively twisted at varying frequencies. The wind tunnel data collected while dynamically twisting the wings reveal an intriguing frequency dependent aerodynamic behavior. It suggests that an improved aerodynamic performance can be attained by modulating the wing twist pattern. This paper is motivated by this observation, and our goal is to exploit the potential in-flight application by determining the optimal wing twist patterns according to the flight profiles. Two simulation models are proposed based on the wind tunnel data. The analytical aeroelastic model is developed and validated by integrating the finite element modeling of the digital wing structure and the vortex lattice modeling of the quasi-steady aerodynamics. A machine learning based neural network model is also developed and validated using the wind tunnel data. The problem of determining the optimal wing twist patterns for these two models can then be formulated as a constrained optimization problem, in which the design objective is to find an optimal twist pattern that minimizes the drag subjected to various physical constraints. This paper demonstrates the applicability of the proposed approach.