Inertial Energy Harvesting For Wearables System Architectures, Power Generation Limits And Transducers

The growing prevalence and variety of wearable devices has led to a renewed interest in harvesting energy from the motion of the human body. We have been working on inertial energy harvesters in which the human body motion excites a proof mass whose motion is then transduced to electrical power. This paper synthesizes many of our findings over the past few years. We first quantitatively compare different mechanical system architectures concluding that, at least for a walking excitation, a sprung rotational structure has significant performance advantages. Secondly, we consider the theoretical upper bound for inertial rotational energy harvesters and compare existing devices to this upper bound. We highlight factors that can explain the gap between existing device performance and the theoretical maximum. We then consider the practical effects of piezoelectric, electromagnetic, and electrostatic transduction. Finally, we demonstrate a rotational electromagnetic prototype with performance exceeding existing devices, but still well below the theoretical maximum.