Energy Harvesting From Turbulence-Induced Vibration in Air Flow: Artificial Piezoelectric Grass Concept

Abstract

Turbulence-induced vibration is generally considered undesirable, and is a phenomenon that if not properly anticipated can lead to catastrophic structural failure. From an energy harvesting perspective however, these types of vibrations have been found to be quite valuable. Turbulence can greatly diminish the performance of traditional fluid flow energy harvesting devices such as turbine, or propeller type designs. Even more recently developed nontraditional harvesters that take advantage of vortex shedding, flutter, or related phenomena in fluid dynamics become extremely inefficient, and in some cases, completely fail to generate useful power in turbulent flow. Motivation for this work comes from the fact that very little research has been done on methods for harvesting energy from turbulence. The primary objective of this research is to design and develop a deploy-and-forget energy harvesting device for use in low velocity (∼0.5 m/s) highly turbulent water flow environments i.e. small rivers and streams. The work presented here focuses on a novel, lightweight, highly robust, energy harvester design referred to as “piezoelectric grass”. This biologically inspired design consists of an array of cantilevers, each constructed with piezoelectric material. When exposed to proper turbulent flow conditions, these cantilevers vibrate vigorously due to rapidly varying pressure fields along their faces. Electrical power is generated directly from these vibrations via the piezoelectric effect. Small-scale wind tunnel tests were carried out to validate the concept. Included in this paper are the results of a direct comparative study performed on two types of harvesters in air. The generating elements or “blades of grass” of one design are made of PVDF cantilevers (type-1), and those of the other design are made with PZT QuickPacks™ mounted to the base of spring steel cantilevers (type-2). Results from these tests show very clearly that optimum harvesting conditions exist. The maximum power output per cantilever was ∼1μW for the type-1 harvester, and increased up to ∼ 1mW for the type-2 harvester which is among the highest found in literature for similar harvesting methods.