UBC Theses and Dissertations
Vibration-based electromagnetic energy harvesters for MEMS applications Khan, Syed Farid Ullah
This thesis investigates vibration-based electromagnetic energy harvesters (EMEHs) for application in low power autonomous sensors. It makes contributions pertaining to the development of a low cost fabrication technology, analytical modeling, simulations and characterization of EMEHs under harmonic and random vibrations. A novel, low cost, one mask fabrication technology devised in this thesis is used to develop a copper foil-type linear EMEH, and a polydimethylsiloxane (PDMS) membrane type nonlinear EMEH. The voltage and power generated by these harvesters are comparable to existing EMEHs which use more involved fabrication processes. In the membrane type EMEH the inclusion of a more flexible PDMS membrane design reduces the harvester resonant frequency and makes it suitable for extracting energy from low level vibration environments. For acceleration levels greater than 0.1 g, this harvester exhibits a nonlinear behaviour. At higher levels of narrow band random excitations, the device therefore exhibits broadening of the load voltage spectrum in comparison to the response under relatively low levels of narrow band random excitations. Analytical models for linear EMEHs with non-uniform magnetic field for harmonic vibrations are developed. A simple analytical model based on Faraday’s law and uniform gradient of the normal component of the magnetic flux density is developed for EMEHs where the entire coil experiences approximately the same gradient of the normal component of the magnetic flux density. However, for EMEHs where the entire coil is not exposed to the same magnetic flux gradient a more robust model, based on the off-center analytical solution of the magnetic flux density is devised. The simulation results of the developed models are in good agreement with the experimental observations. Analytical models for linear and nonlinear EMEHs under random vibrations are derived. The models are parameterized such that they are applicable to all architectures of EMEHs and can be utilized for designing and performance estimation of EMEHs. Nonlinear harvesters with spring nonlinearity and with combined spring and damping nonlinearity are modeled using the statistical linearization method. The developed models are useful in investigating the effects of the mechanical nonlinearity on the performance and bandwidth of the harvesters under random vibrations.
Item Citations and Data
Attribution-NonCommercial-NoDerivatives 4.0 International