Preliminary Finite Element Modeling of a Piezoelectric Actuated Marine Propulsion Fin

Andrew R. Streett

Physical copy available from RIT's Wallace Library at VM751 .S77 2006

Abstract

New technologies surrounding composite materials and autonomous underwater vehicle (AUV) design have led to numerous studies involving the marine propulsion for these AUVs. AUVs traditionally are classified as highly efficient, payload capable, and can be utilized as reconnaissance or surveillance vehicles. Undullatory and oscillatory propulsion devices have been conceived to replace the present propulsion technologies, of propellers, with highly maneuverable, efficient, and quiet propulsion systems. Undullatory and oscillatory propulsion has been around for centuries employed by aquatic life, but only recently have the mini-technologies been available to present such propulsion devices economically and with enough materials research as to mimic biologic life on the same scale.

Piezoelectric properties coupled with a thin plate allow for actuation properties, similar to bimetallic metals. Applying two piezoelectrics to the fixed end of a cantilevered beam or plate, on opposite sides, and actuating them with an opposite phase shift in electrical voltage potential results in transverse motion of the beam from the orthogonal plane to the vertical axis of the piezoelectric device. Coupling this property to a particular fiber orientation, composite thin plate, significantly increases the actuation properties. In addition, placing more than two piezoelectrics along the length of the thin composite plate gives the potential to increase actuation properties and change the motion from oscillatory to undullatory. These motions can again be increased by utilizing the natural vibration modes of the thin composite plate with piezoelectrics near resonance actuation.

The current research is involved with modeling a piezoelectric actuated marine propulsion fin using the Galerkin finite element technique. An experimental proof of concept was developed to compare results. Using fluid-structure interaction (FSI) methods, it is proposed that the fluid and structure programs are resolved within one program. This is in contrast to traditional attempts at FSI problems that utilize a computational fluid dynamics (CFD) solver transferring load data between a structural dynamics/finite element (FE) program.