Multiphysics modeling and data-driven MOR of flow-induced piezoelectric energy harvesters
At LMSSC, Paris, December 7th 2021, 2 p.m.
Doctorate, University of Luxembourg, Esch-sur-Alzette
We propose a high-fidelity model for flow-driven piezoelectric energy harvesters, where an incompressible viscous fluid, a large deformation piezoelectric composite beam and an integrated electric circuit are accounted for. A numerical solution to multiple fields is obtained simultaneously using the finite element method, which is reliable but at the cost of high computational expense.
Artificial neural network-aided model order reduction is employed to alleviate this issue. Preliminary results of this approach will be presented in the seminar.