Nonlinear MEMS Piezoelectric Harvesters in the presence of geometric and structural variabilities

This paper investigates the use of an electrostatic device to improve the performance of MEMS piezoelectric harvesters in the presence of geometric and structural variabilities due to the manufacturing process. Different types of uncertain parameters including material and geometric uncertainties have been considered. The variability of these parameters are estimated based on available existing experimental data in the literature. Monte Carlo simulation (MCS) is used for uncertainty propagation and it is shown that the resonance frequencies of the majority of the samples are far away from the excitation frequency and consequently this results in less harvested power. This paper identifies these samples and uses electrostatic devices to improve the performance of the harvester. The proposed device is composed of an unsymmetric arrangement of two electrodes to decrease the resonance frequency of samples through a softening nonlinearity. The unsymmetric arrangement of two electrodes is inevitable and due to geometric variability of the harvester. There are also two arch shape electrodes which can be used to create a hardening effect to increase the resonance frequency of samples which have resonance frequencies smaller than the nominal value.