The global electric power infrastructure is undergoing a dramatic transformation from a centralized, demand-following network to one that is more dynamic and consumer-interactive. This process requires a very large number of advanced sensors be deployed throughout the entire system. This manuscript documents the development of two enabling technologies for the realization of such large-scale monitoring networks. The first technology pertains to the idea of harvesting ambient energy to power wireless sensor nodes on overhead power lines such that these wireless sensors can be operated continuously without interruptions. A new type of piezoelectromagnetic (PEM) energy harvester is developed. The harvester is comprised of a bimorph piezoelectric cantilever with permanent magnets mounted on its tip. When placed near to an AC current source, the magnets couple to the AC magnetic field and excite the piezoelectric cantilever, which in turn generates a large electrical voltage output. The energy converted by the harvester is then rectified, stored and conditioned by a power conditioning circuit. The power is delivered to a wireless sensor platform at a regulated DC voltage. In this design, an enhanced magnetic coupling method is proposed and verified. A new closed-form lumped parameter model for the bimorph piezoelectric cantilever is devel- oped. The harvester is able to generate a raw power output of 500μW when the device is placed 1cm from the center of a 0.75 inch conductor that carries a current of 20A. At that level of power output, the device is able to allow a low-power wireless sensor node to collect and transmit measurement data to a remote device once every 10 seconds.

The second research topic documented in this thesis is about non-intrusive electricity metering using remote energy sensing techniques. A stick-on energy metering system that can be easily installed on the surface of circuit breaker panels is developed. The system uses magnetic and electric sensor to measure the electromagnetic field produced by the energized conductor in each individual circuit. A signal processing algorithm is invented to accurately reconstruct the current and voltage waveforms from the measurement data. The installation of such stick-on devices does not require the use of electricians. Therefore, the overall cost of deployment is estimated to be one-tenth that of the state-of-the-art solutions.