Wireless sensor networks are being used increasingly in wide variety of applications to monitor the surrounding environment. To realize the full potential and autonomy of such sensor networks, the sensor nodes should be able to operate without frequent repair or replacement. Because of the lifetime and energy capacity limitation of electrochemical batteries, finding an alternative energy source has been a topic of great interest lately. Since vibrations occur in most structures, harvesting energy from the ambient vibrations is a promising way to power wireless sensor networks.;In this research work, optimal energy harvesting systems using magnetostrictive material (MsM) are designed to power the Wireless Intelligent Sensor Platform (WISP), developed by North Carolina State University. Compared with the piezoelectric material based harvester, MsM is more flexible, has an inherent low natural frequency, and it provides almost unlimited number of vibration cycles. A linear MsM energy harvesting device was modeled and optimized for maximum power output. The conversion efficiency, converting from magnetic to electric energy, is approximately modeled from the magnetic field induced by the beam vibration. From the measurement, the open circuit voltage is 1.5 V when the MsM cantilever beam is operating at the 2nd order natural frequency 324 Hz. The AC output power is 0.97 mW and power density 279 muW/cm3. Since the MsM device has low open circuit output voltage characteristics, a full-wave quadrupler was designed to boost the rectified output voltage. To implement complex conjugate impedance matching between load and the MsM device, a discontinuous conduction mode (DCM) buck-boost converter was designed. From the measurement of the prototype interfacing circuit and optimized linear MsM energy harvester, the maximum output power after the voltage quadrupler is 705 muW and power density is 202.4 muW/cm3, which is comparable to the piezoelectric energy harvesters in the literature. The output power delivered to a lithium rechargeable battery is around 630 muW and is independent of the load resistance.;The above linear device is most effective when the input vibration frequency closely matches the natural frequency of the device. The output power from such energy harvesters falls off very quickly even if there is a slight mismatch between the natural frequency of the device and the input vibration frequency. In most of the environments, depending on the operating conditions, the frequencies of the driving vibration are random or may change over time. In order to extract more energy from such vibrations, the energy harvester should have sufficient bandwidth to cover peak power frequencies of the input vibrations. A nonlinear MsM based energy harvester is proposed in this work. The cubic nonlinearity of the clamped-clamped buckled beam makes wideband design possible. Since MsM is flexible, it is feasible to use MsM on a buckled beam. The governing equations for the buckled MsM energy device are derived using extended Hamilton's principle. A single-mode Galerkin approximation is used to discretize the nonlinear partial-differential equation and to obtain an ordinary-differential equation with respect to time only. The dynamic equations for the nonlinear energy harvester are solved using numerical methods and the simulation results confirmed that the nonlinear energy harvester has a much wider bandwidth compared with the linear energy harvester. The experiment was carried out for the nonlinear MsM energy harvester built using Metglas 2605SA1. The relationship between the axial load and the important parameters such as the resonance characteristics, the bandwidth, and the power output to a resistive load were measured and discussed for the prebuckled and buckled beam.
展开▼
