Reduced-order modeling with multiple scales of electromechanical systems for energy harvesting

New technologies that aim at powering wireless nodes by scavenging the energy from ambient vibrations can be a practical solution for some structural monitoring applications in the near future. In view of possible large-scale applications of piezoelectric energy harvesters, an accurate modeling of the interfaces in these devices is needed for more advanced and reliable simulations, since they might have large influence on functionality and performance of smart monitoring infrastructures. In this perspective, a novel multiscale and multiphysics hybrid approach is proposed to assess the dynamic response of piezoelectric energy harvesting devices. Within the framework of the presented approach, the FE2 method is employed to compute stress and strain levels at the microscale in the most critical interfaces. The displacement-load curve of the whole device (so-called capacity curve or pushover curve) is then obtained by means of the application of a suitable pattern of static forces. Finally, the parameters of a reduced-order model are calibrated on the basis of the nonlinear static analysis. This reduced-order model, in turn, is employed for the efficient dynamic analysis of the energy harvesting device.