Vibrational energy harvesting for sensors in vehicles

Licentiate thesis, 2022

The miniaturization of semiconductor technology and reduction in power requirements have begun to enable wireless self-sufficient devices, powered by ambient energy. To date the primary application lies in generating and transmitting sensory data. The number of sensors and their applications in automotive vehicles has grown drastically in the last decade, a trend that seems to continue still. Wireless self-powered sensors can facilitate current sensor systems by removing the need for cabling and may enable additional applications. These systems have the potential to provide new avenues of optimization in safety and performance.

This thesis delves into the topic of vibrations as ambient energy source, primarily for sensors in automotive vehicles. The transduction of small amounts of vibrational, or kinetic, energy to electrical power, also known as vibrational energy harvesting, is an extensive field of research with a plethora of inventions. A short review is given for energy harvesters, in an automotive context, utilizing transduction through either the piezoelectric effect or magnetic induction. Two practical examples, for ambient vibration harvesting in vehicles, are described in more detail. The first is a piezoelectric beam for powering a strain sensor on the engines rotating flexplate. It makes combined use of centrifugal force, gravitational pull and random vibrations to enhance performance and reduce required system size. The simulated power output is 370 µW at a rotation frequency of 10.5 Hz, with a bandwidth of 2.44 Hz. The second example is an energy harvesting unit placed on a belt buckle. It implements magnetic induction by the novel concept of a spring balance air gap of a magnetic circuit, to efficiently harvest minute vibrations. Simulations show the potential to achieve 52 µW under normal road conditions driving at 70 km/h.

Theoretical modeling of these systems is also addressed. Fundamental descriptions of the lumped and distributed models are given. Based on the lumped models of the piezoelectric energy harvester (PEH) and the electromagnetic energy harvester (EMEH), a unified model is described and analyzed. New insights are gained regarding the pros and cons of the two types of energy harvester run at either resonance or anti-resonance. A numerical solution is given for the exact boundary of dimensionless quality factor and dimensionless intrinsic resistance, at which the system begins to exhibit anti-resonance. Regarding the maximum achievable power, the typical PEH is favored when running the system in anti-resonance and the typical EMEH is favored at resonance. The described modeling considers all parameters of the lumped model and thus provides a useful tool for developing vibrational energy harvester prototypes.