Experimental System Identification of a Fully Coupled State Space Model for Mechanism-Free Ornithopters

Abstract A piezoelectric-based mechanism-free ornithopter can eliminate the need for electromagnetic motors and conventional power transmission mechanisms, potentially saving weight and energy consumption, improving durability, and reducing overall mechanical complexity. In the previous research by the authors, the structural and fluid domains of the mechanism-free ornithopters were modeled by the Rayleigh-Ritz method and empirical state space Theodorsen function, respectively. The fully coupled system model is in state space form, which facilitates the application of parametric analyses in the preliminary design stage, multidisciplinary design optimization, and analyses of stability and controllability. This paper aims to: (1) identify the unknown model coefficients and functions, and (2) validate the coupled state space model by using wind tunnel experiments. A prototype of the piezocomposite mechanism-free ornithopter wing is fabricated and subjected to testing in a low-speed wind tunnel. Each piezocomposite wing of the tested prototype is actuated by two Macro-Fiber Composite actuators bonded to a carbon fiber substrate. Initial simulation results indicate that this setup is particularly effective in replicating the heaving (bending) and pitching (twisting) motions characteristic of a natural flapping wing. The placement and alignment of these actuators were previously determined through analysis of the flapping wing model. The wind tunnel experiments involve displacement and force measurements to validate the structural and fluid domain solutions, respectively. Aerodynamic load data is acquired using a six-degree-of-freedom load cell, and displacements are measured using 3D motion-capture camera system. The multipoint motion-capture technique facilitates for reconstructing a 3D representation of the motion of the flapping wing prototype. Tests are conducted across variations in key parameters such as flow speed, angle of attack, and excitation voltage magnitude, frequency, and phase of the excitation voltage between the two actuators. The experimental results are compared to the model predicted solution to evaluate the model’s accuracy and empirically obtain the system-level parameters under a wide range of operating conditions.