Modelling and Optimization of a Magnetic Spring Based Electromagnetic Vibration Energy Harvester

    Research output: Contribution to journalArticlepeer-review


    This paper presents the development of an AA battery size electromagnetic vibration energy harvester with an aim to maximize the output power density. A tube shape and stacked opposing permanent magnets with magnetic spring were used to suit the shape constraint as well as to achieve high flux linkages. An initial prototype of electromagnetic vibration harvester with AA battery size was built and tested on a controllable shaker to obtain its output voltage and power level at different frequencies for fixed accelerations. A single magnet was fixed at the bottom of the harvester to provide levitation force in this development in order to lower the resonant frequency. A special time-domain based analytical model was also developed using both Finite Element Analysis and Simulink simulation. The time-domain analytical model is easier to implement than other frequency domain based analytical models which generally applied in literatures for modelling of the electromagnetic vibration energy harvesters. The analytical model was verified by the measured results obtained from the initial prototype. The validated analytical model was successfully applied to optimize the harvester. Two more generator prototypes were further built and tested after the optimization study. The optimized harvester using three stacked opposing permanent magnets could achieve a normalized power density of 12,655 μWcm−3 g−2 at 9.9 Hz frequency with 0.22 g acceleration, which is significantly higher than other reported electromagnetic vibration energy harvesters.
    Original languageEnglish
    Article numberVolume 17
    Pages (from-to)463-474
    JournalJournal of Electrical Engineering & Technology
    Publication statusPublished (VoR) - 20 Sept 2021


    Dive into the research topics of 'Modelling and Optimization of a Magnetic Spring Based Electromagnetic Vibration Energy Harvester'. Together they form a unique fingerprint.

    Cite this