Design optimization of multi-resonant piezoelectric energy harvesters

Abstract

In this work, we perform a two-stage design optimization of the folded beam multi-resonant piezoelectric energy harvester. The optimization routine involves a global pre-optimization using evolutionary algorithm and a subsequent local optimization by classical non-linear programming. Optimization resulted in designs, which shows promising characteristics in simulation. For the validation of simulation results, we fabricated the optimal designs and performed experimental characterization. Simultaneously, we developed a comparable array of single beam piezoelectric harvesters, which we compared with the optimized folded beam designs. Both new designs significantly outperform the initial one in terms of frequency spacing. Furthermore, the experimental data demonstrate that the optimized folded beam designs generate considerably higher maximum voltage compared to the proposed array-structured designs.