In the present work we explore some aspects of energy harvesting from unsteady, turbulent fluid flow using piezoelectric generators. Turbulent flows exhibit a large degree of coherence in their spatial and temporal scales, which provides a unique opportunity for energy harvesting. The voltage generated by short, flexible piezoelectric cantilever beams placed inside turbulent boundary layers and wakes of circular cylinders at high Reynolds numbers is investigated. Matching the fluid flow’s predominant frequency with the natural frequency of the piezoelectric generator appears to maximize the piezoelectric output voltage. This voltage is also dependent on the generator’s location inside the flow field. A three-way coupled interaction simulation that takes into account the aerodynamics, structural vibration, and electrical response of the piezoelectric generator has been developed. The simulation results agree reasonably well with the experimental data paving the way of using such a tool to estimate the performance of different energy harvesting devices within unsteady flow fields.
The available power in a flowing fluid is proportional to the cube of its velocity, and this feature indicates the potential for generating substantial electrical energy by exploiting the direct piezoelectric effect. The present work is an experimental investigation of a self-excited piezoelectric energy harvester subjected to a uniform and steady flow. The harvester consists of a cylinder attached to the free end of a cantilevered beam, which is partially covered by piezoelectric patches. Due to fluid–structure interaction phenomena, the cylinder is subjected to oscillatory forces, and the beam is deflected accordingly, causing the piezoelectric elements to strain and thus develop electric charge. The harvester was tested in a wind tunnel and it produced approximately 0.1 mW of non-rectified electrical power at a flow speed of 1.192 m s−1. The aeroelectromechanical efficiency at resonance was calculated to be 0.72%, while the power per device volume was 23.6 mW m−3 and the power per piezoelectric volume was 233 W m−3. Strain measurements were obtained during the tests and were used to predict the voltage output by employing a distributed parameter model. The effect of non-rigid bonding on strain transfer was also investigated. While the rigid bonding assumption caused a significant (>60%) overestimation of the measured power, a non-rigid bonding model gave a better agreement (<10% error).
The idealized interactions of shock waves with homogeneous and isotropic turbulence, homogeneous sheared turbulence, turbulent jets, shear layers, turbulent wake flows, and two-dimensional boundary layers have been reviewed. The interaction between a shock wave and turbulence is mutual. A shock wave exhibits substantial unsteadiness and deformation as a result of the interaction, whereas the characteristic velocity, timescales and length scales of turbulence change considerably. The outcomes of the interaction depend on the strength, orientation, location, and shape of the shock wave, as well as the flow geometry and boundary conditions. The state of turbulence and the compressibility of the incoming flow are two additional parameters that also affect the interaction. 310ANDREOPOULOS Ⅲ AGUI Ⅲ BRIASSULIS curvature, flow separation, dilatational effects, or longitudinal pressure gradients that may be present in flow before or after the interaction with the shock.The outcomes of the shock-turbulence interaction depend on (a) the characteristics of the interacting shock-wave-like strength, relative orientation to the incoming flow, and location and shape, (b) the state of turbulence of the incoming flow as it is characterized by the fluctuation levels of velocity, density, pressure, and entropy and length scales, (c) the level of compressibility of the incoming flow, and (d) the flow geometry and boundary conditions. Basic understanding of the physics of such complex interactions has been obtained through investigations of conveniently selected and reasonably simplified flow configurations. The flows to be considered here include shear free flows, shear layers, and wall-bounded flows. Most of the work in this review is confined to the following cases: (a) homogeneous and isotropic turbulence interactions with shock waves (see Figure 1a), (b) constant shear homogeneous turbulence interactions with shock waves (see Figure 1b), (c) circular jet flows interacting with shock waves (see Figure 1c), (d) plane shear layers interacting with oblique shock waves (see Figure 1d), (e) wake flows interacting with oblique or normal shock waves (see Figure 1e), and ( f ) boundary layer interactions with oblique shock waves (see Figure 1f). This classification of the interactions also reflects the level of increasing complexity from the first to the last flow configuration.We restrict our review to nominally two-dimensional interactions, albeit keeping in mind that all of these interactions are three-dimensional in nature because turbulence in all its complexity is characterized by instantaneous flow variables that exhibit a variation in time and space. Shock wave-boundary layer interactions have been reviewed in the past by Green (1970), Adamson & Messiter (1980), andDodson (1991). Compressibility effects of turbulence, including some shock wave interactions, have been recently reviewed by Lele (1994). Compressibility effects in turbulent boundary layers and shear layers in the absence of shock interactions have been reviewed by Sp...
Experimental results are presented that reveal the structure of a two-dimensional turbulent boundary layer which has been investigated by measuring the timedependent vorticity flux at the wall, vorticity vector, strain-rate tensor and dissipationrate tensor in the near-wall region with spatial resolution of the order of 7 Kolmogorov viscous length scales. Considerations of the structure function of velocity and pressure, which constitute vorticity flux and vorticity, indicated that, in the limit of vanishing distance, the maximum attainable content of these quantities which corresponds to unrestricted resolution, is determined by Taylor's microscale. They also indicated that most of the contributions to vorticity or vorticity flux come from the uncorrelated part of the two signals involved. The measurements allowed the computation of all components of the vorticity stretching vector, which indicates the rate of change of vorticity on a Lagrangian reference frame if viscous effects are negligible, and several matrix invariants of the velocity gradient or strain-rate tensor and terms appearing in the transport equations of vorticity, strain rate and their squared fluctuations. The orientation of vorticity revealed several preferential directions. During bursts or sweeps vorticity is inclined at 35m to the longitudinal direction. It was also found that there is high probability of the vorticity vector aligning with the direction of the intermediate extensive strain corresponding to the middle eigenvector of the strain-rate matrix. The results of the joint probability distributions of the vorticity vector orientation angles showed that these angles may be related to those of hairpin vortex structures. All invariants considered exhibit a very strong intermittent behaviour which is characterized by large-amplitude bursts which may be of the order of 10 r.m.s. values. Small-scale motions dominated by high rates of turbulent kinetic energy dissipation and high enstrophy density are of particular interest. It appears that the fluctuating strain field dominates the fluctuations of pressure more than enstrophy. Local high values of the invariants are also often associated with peaks in the shear stress.
Neuromorphic vision sensing (NVS) devices represent visual information as sequences of asynchronous discrete events (a.k.a., "spikes") in response to changes in scene reflectance. Unlike conventional active pixel sensing (APS), NVS allows for significantly higher event sampling rates at substantially increased energy efficiency and robustness to illumination changes. However, feature representation for NVS is far behind its APS-based counterparts, resulting in lower performance in high-level computer vision tasks. To fully utilize its sparse and asynchronous nature, we propose a compact graph representation for NVS, which allows for end-to-end learning with graph convolution neural networks. We couple this with a novel end-to-end feature learning framework that accommodates both appearancebased and motion-based tasks. The core of our framework comprises a spatial feature learning module, which utilizes residual-graph convolutional neural networks (RG-CNN), for end-to-end learning of appearance-based features directly from graphs. We extend this with our proposed Graph2Grid block and temporal feature learning module for efficiently modelling temporal dependencies over multiple graphs and a long temporal extent. We show how our framework can be configured for object classification, action recognition and action similarity labeling. Importantly, our approach preserves the spatial and temporal coherence of spike events, while requiring less computation and memory. The experimental validation shows that our proposed framework outperforms all recent methods on standard datasets. Finally, to address the absence of large realworld NVS datasets for complex recognition tasks, we introduce, evaluate and make available the American Sign Language letters (ASL-DVS), as well as human action dataset (UCF101-DVS, HMDB51-DVS and ASLAN-DVS). Figure 1: Examples of archery action captured by APS and NVS sensors. APS sensors capture images at fixed frame rates, while NVS sensors output a stream of events.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.