Hostname: page-component-7b7567c8c-fvjsw Total loading time: 0 Render date: 2026-11-16T20:17:34.260Z Has data issue: false hasContentIssue false

Flutter instability in an internal flow energy harvester

Published online by Cambridge University Press:  12 March 2021

Luis Phillipe Tosi*
Affiliation:
Division of Engineering and Applied Science, California Institute of Technology, Pasadena, CA 91125, USA
Benedikt Dorschner
Affiliation:
Division of Engineering and Applied Science, California Institute of Technology, Pasadena, CA 91125, USA
Tim Colonius
Affiliation:
Division of Engineering and Applied Science, California Institute of Technology, Pasadena, CA 91125, USA
*
†Email address for correspondence: ltos@caltech.edu

Abstract

Vibration-based flow energy harvesting enables robust, in situ energy extraction for low-power applications, such as distributed sensor networks. Fluid–structure instabilities dictate a harvester's viability since the structural response to the flow determines its power output. Previous work on a flextensional based flow energy harvester demonstrated that an elastic member within a converging-diverging channel is susceptible to the aeroelastic flutter. This work explores the mechanism driving flutter through experiments and simulations. A model is then developed based on channel flow rate modulation and considering the effects of both normal and spanwise flow confinement on the instability. Linear stability analysis of the model replicates flutter onset, critical frequency and mode shapes observed in experiments. The model suggests that flow modulation through the channel throat is the principal mechanism for the fluid-induced vibration. The generalized model presented can serve as the foundation of design parameter exploration for energy harvesters, perhaps leading to more powerful devices in the future, but also to other similar flow geometries where the flutter instability arises in an elastic member within a narrow flow passage.

Information

Type
JFM Papers
Copyright
© The Author(s), 2021. Published by Cambridge University Press
Figure 0

Figure 1.Illustration of flow path and relevant geometry. Values and units are listed in table 5. (a) Axial cross-section of flow geometry illustrating flow path. (b) Cut section a-a.

Figure 1

Table 1.Table of fluid–structure dimensional parameters.

Figure 2

Figure 2.Illustration cantilever beam in a converging-diverging channel geometry (b) with a simple harmonic boundary condition (a).

Figure 3

Table 2.Table of fluid–structure non-dimensional parameters for flextensional boundary and cantilevered beam.

Figure 4

Table 3.Table of calculated mean flextensional (Flex.) properties.

Figure 5

Figure 3.Representative SPOD data for flextensional setting 1 showing self-sustained oscillating regime of mode 1 and under-damped regime of mode 2. (a) Trace of cross-spectral density matrices at discrete frequencies. (b) Phase diagram for mode 1. (c) Phase diagram for mode 2. (d) Spectral proper orthogonal decomposition mode shape for mode 1. (e) Spectral proper orthogonal decomposition mode shape for mode 2.

Figure 6

Figure 4.Video data set showing mode 1 amplitude vs. flow rate for all flextensional settings. (a) Flextensional 1. (b) Flextensional 2. (c) Flextensional 3.

Figure 7

Figure 5.Video data set showing mode 1 frequency response vs. flow rate for all flextensional settings. (a) Flextensional 1. (b) Flextensional 2. (c) Flextensional 3.

Figure 8

Table 4.Table of critical and restoring (fold) values for Flex. settings.

Figure 9

Figure 6.Plots of PZT 1 voltage amplitude vs. flow rate for all flextensional settings. (a) Flextensional 1. (b) Flextensional 2. (c) Flextensional 3.

Figure 10

Figure 7.Plots of PZT 1 voltage PSD primary frequency (frequency associated with largest PSD amplitude peak) vs. flow rate for all flextensional settings. (a) Flextensional 1. (b) Flextensional 2. (c) Flextensional 3.

Figure 11

Figure 8.Instantaneous snapshot of the computational set-up for $Q=208\ \textrm {L}\ \textrm {min}^{-1}$, showing a slice of velocity magnitude. Exemplary observer points are indicated by the red spheres.

Figure 12

Figure 9.Flow evolution for one period of the energy harvester, showing a slice of velocity magnitude.

Figure 13

Figure 10.Flow evolution for half a period of the energy harvester, showing isosurfaces of vorticity coloured by velocity magnitude and zoomed into the diffuser region.

Figure 14

Figure 11.Beam oscillation amplitude as a function of the flow rate for the experiments as well as the numerical simulations.

Figure 15

Figure 12.Evolution of the beam displacement and its PSD.

Figure 16

Figure 13.Evolution of the streamwise velocity profile for three probes.

Figure 17

Figure 14.Plot of normalized PSD of pressure for the observer points.

Figure 18

Figure 15.Spanwise velocity profile. (a) Location of the phase-averaging plane. (b) Phase-averaged spanwise velocity profile.

Figure 19

Figure 16.Three-dimensional control volume illustration for the spanwise quasi-1-D leakage flow model. (a) Projection view of channel control volume. (b) Front view of channel control volume.

Figure 20

Figure 17.Illustration of spanwise quasi-1-D geometry for comparison to experimental results.

Figure 21

Figure 18.Flutter boundary defined by critical flow rate vs. diffuser expansion angle for all three flextensional settings. (a) Flex. 1. (b) Flex. 2. (c) Flex. 3.

Figure 22

Figure 19.Frequencies at critical flow rates vs. diffuser expansion angle for all three flextensional settings. (a) Flex. 1. (b) Flex. 2. (c) Flex. 3.

Figure 23

Figure 20.Mode shape comparison between EB model and experimental SPOD results. (a) Flex. 1. (b) Flex. 2. (c) Flex. 3.

Figure 24

Table 5.Table of flow path parameter dimensions illustrated in figure 1.

Figure 25

Table 6.Table of structural material properties (Boston Piezo Optics Inc. 2018; MatWeb 2018).

Figure 26

Table 7.Table of relevant piezoelectric stack properties.

Figure 27

Table 8.Table of mean force (N) at different displacement values for three flexure settings, with a linear best fit slope describing the stiffness $k_0$$(\textrm {N}\ \textrm {m}^{-1})$.

Figure 28

Table 9.Table of experimental flexure values for flexure dynamic test.

Figure 29

Table 10.Table of critical values for flexure settings listed.

Figure 30

Table 11.Table of solutions to the characteristic equation for clamped-free EB beam.