Robust optimization of a dual-stage bistable nonlinear vibration energy harvester considering parametric uncertainties

Abstract

This study investigates robust optimization of a dual-stage bistable energy harvester (BEH) considering parametric uncertainties, so as to efficiently search the robust optimal design parameters for enhancing vibration energy harvesting performance in a working bandwidth. Dynamic model of the dual-stage BEH and numerical method of predicting the broadband power transmissibility are presented, the fidelity of which is verified by the analytical solutions and experimental results. Then, the optimization problem is defined, where the performance index for optimization is calculated based on the power transmissibility in a working bandwidth. To ensure the robustness of optimization results, instead of straightforward maximizing the performance index, the genetic algorithm is used to maximize the mean value of the performance indices corresponding to the parametric uncertainties which satisfy normal distribution. Results show that compared to the optimization disregarding the uncertainties, the proposed robust optimization method can significantly enhance the overall performance and robustness of the dual-stage BEH subjected to the parametric uncertainties. Therefore, the comparative results validate the effectiveness of the proposed method for robust optimization of the dual-stage BEH.

Active vibration isolation performance of the bistable nonlinear electromagnetic actuator with the elastic boundary

“…Inspired by the discoveries of the previous researches that additional dynamic coupling can enhance the vibration responses of the bistable structure [21][22][23], our research group recently proposed a new bistable electromagnetic actuator with the elastic boundary (simply named EB-bistable actuator) [24][25][26]. The EB-bistable actuator adds another linear spring and joint mass to form the elastic boundary (EB) at one end of the bistable nonlinear spring, and thus the bistable nonlinearity is dynamically changed during the actuation [24].…”