As the required power for wireless, low-power sensor systems continues to decrease, the feasibility of a fully self-sustaining, onboard power supply, has increased interest in the field of vibration energy harvesting — where ambient kinetic energy is scavenged from the surrounding environment. Current literature has produced a number of harvesting techniques and transduction methods; however, they are all fundamentally similar in that, the harmonic excitation frequency must fall within the resonant bandwidth frequency of the harvesting mechanism to maintain acceptable energy output. The purpose of this research is to investigate the potential for natural frequency tuning by means of passive electrical components, that is, using an imposed electrical inductance to adjust the equivalent stiffness, and resulting resonant frequency of a vibration energy harvester. In past literature, it was concluded that an “active” frequency tuning mechanism would be infeasible, as the power required by an equivalent “stiffening transducer” would require more power to maintain the system at resonance, than the system would actually produce as a result of maintaining resonance, i.e., a net energy loss (Roundy and Zhang 2005). It is believed that the model used in this conclusion can be improved by directly modeling changes in system stiffness as an equivalent mechanical spring, instead of an external inertial loading. Due to the conservative nature of the harmonic spring, the compliance of a harvesting mechanism can be theoretically altered without energy losses, whether the actuation is applied using “active” or “passive” means. This revised model departs from the traditional, base excitation model in most vibration energy harvesting systems, and includes additional stiffness, and damping elements, representative of induced mechanical spring, and related losses. We can validate the feasibility of this technique, if it can be shown that the natural frequency of an energy harvester can be altered, and still maintain energy output similar to its “untuned” natural frequency. If feasible, this tuning method would provide a viable alternative to other bandwidth-increasing techniques in literature, e.g., wideband harvesting, bandwidth normalizing, high-damping, etc. In this research, a change in natural frequency of the experimental energy harvesting system of 0.5 Hz was demonstrated, indicating that adjusting the natural frequency of a vibration energy harvesting system is possible; however, there are many new challenges associated with the revised energy harvesting model, related to the new introduced losses to the system, as well as impedance matching between the mechanical and electrical domains. Further research is required to better quantitatively characterize the relationship between natural frequency shift, and imposed electrical inductance.
FINITE ELEMENT SIMULATION OF MEMS PIEZOELECTRIC ENERGY SCAVENGER BASED ON PZT THIN FILM
