Real world assessment of an auto-parametric electromagnetic vibration energy harvester


The convention within the field of vibration energy harvesting has revolved around designing resonators with natural frequencies that match single fixed frequency sinusoidal input. However, real world vibrations can be random, multi-frequency, broadband and time-varying in nature. Building upon previous work on auto-parametric resonance, this fundamentally different resonant approach can harness vibration from multiple axes and has the potential to achieve higher power density as well as wider frequency bandwidth. This article presents the power response of a packaged auto-parametric VEH prototype (practical operational volume of ∼126 cm−3) towards various real world vibration sources including vibration of a bridge, a compressor motor as well as an automobile. At auto-parametric resonance (driven at 23.5 Hz and 1 grms), the prototype can output a peak of 78.9 mW and 4.5 Hz of −3dB bandwidth. Furthermore, up to ∼1 mW of average power output was observed from the harvester on the Forth Road Bridge. The harvested electrical energy from various real world sources were used to power up a power conditioning circuit, a wireless sensor mote, a micro-electromechanical system accelerometer and other low-power sensors. This demonstrates the concept of self-sustaining vibration powered wireless sensor systems in real world scenarios, to potentially realise maintenance-free autonomous structural health and condition monitoring.

Divisions: College of Engineering & Physical Sciences > School of Engineering and Technology > Mechanical, Biomedical & Design
Additional Information: © Sage 2017. The final publication is available via Sage at
Publication ISSN: 1530-8138
Last Modified: 29 Nov 2023 12:15
Date Deposited: 25 Mar 2019 14:04
Full Text Link: https://www.rep ... dle/1810/273608
Related URLs: https://journal ... 045389X17740964 (Publisher URL)
PURE Output Type: Article
Published Date: 2018-04-01
Published Online Date: 2017-11-15
Accepted Date: 2017-11-01
Authors: Jia, Yu
Yan, Jize
Du, Sijun
Feng, Tao
Fidler, Paul
Middleton, Campbell
Soga, Kenichi
Seshia, Ashwin A



Version: Accepted Version

| Preview

Export / Share Citation


Additional statistics for this record