Modern internet–based technologies require devices that can generate, acquire, and
compute data. Typically, these devices rely on conventional batteries, which limit device
size, lifespan, and introduce maintenance costs due to replacements.
A promising alternative is using piezoelectric energy harvesters, which convert mechanical
stress into voltage via the piezoelectric effect. These harvesters need to be activated
by external vibrations, but most current systems operate in a frequency range of 50 to
300 Hz. However, ambient vibrations often occur at lower frequencies (5 to 40 Hz) and
are not constant,making it challenging to harvest energy efficiently when the frequency
mismatch occurs.
The objective of this research is to develop a design that can harvest energy from a broad
range of low and non–constant frequencies and develop a testing methodology for its
dynamic characterization that represents a real–world scenario, by using square wave
and noise excitation.
The proposed design is based on the frequency up–conversion principle. The design
features a driving oscillator, a bi–stable oscillator that resonates at low frequencies and
collects energy during its snap–through motion. The generating oscillator, on the other
hand, resonates at higher frequencies and contains the piezoelectric material. The energy
harvesting is achieved by the transfer of energy from the driving to the generating
oscillator.
The dynamic characterization of the design resulted in several findings. The proposed
design can up–convert frequency by a factor of 208 and harvest energy at low input vibration
(0.1–20 Hz) and noise excitation (0–300 Hz).
Overall this research contributes to the field of energy harvesting by introducing a new
design that can harvest energy fromambient vibrations and a new testing procedure that
replicates a real–world scenario.