Synthetic assessment of self-powered energy-harvesting including robustness evaluation

We have developed a smart energy harvester that generates electrical energy from multi-modal vibrations. The harvester consists of a digital processor and a piezoelectric sensor, which allows the application of a technical method to improve the energy-conversion efficiency. The method is implemented by measuring the vibration displacements, processing the data digitally, and adequately regulating electric switches. These operations are managed by a built-in digital processor. The driving power for the digital processor is satisfied with a part of the energy harvested from structural vibrations. Thus, the harvester operates flexibly with the digital processor to enhance electrical energy generation, and requires neither batteries nor an external power supply. We refer to the proposed device as a self-powered energy harvester. An advantage of digital processing is that observations by a Kalman filter can be used to estimate modal structural vibrations. In addition to reducing sensor noise, the digital filter extracts modal values from the measured displacement data. Here, we describe the basic configuration of the proposed harvester and demonstrate energy harvesting from multi-modal vibrations for a structure with 2 degrees of freedom (DOF). We assess the internal energy consumption of self-powered control devices, such as the digital processor and DC/DC converter. In addition, we show the robustness of the proposed harvester by conducting a harvesting experiment with electrical noise. The results demonstrate that the self-powered energy harvester generates more electrical energy from 2DOF vibrations than does a conventional harvester, and accurately operates under noisy conditions.