Tuning Nonlinear Model Parameters in Piezoelectric Energy Harvesters to Match Experimental Data

Abstract

Abstract A framework that allows the use of well-known dynamic estimators in piezoelectric harvesters (PEHs) (i.e., deterministic performance estimators) and that accounts for the random error associated with the mathematical model and the uncertainties of model parameters is presented here. This framework may be employed for Posterior Robust Stochastic analysis, such as when a harvester can be tested or is already installed and the experimental data are available. In particular, the framework detailed here is introduced to update the electromechanical properties of PEHs using Bayesian techniques. The updated electromechanical properties are identified by adopting a Transitional Markov Chain Monte Carlo. A well-known device with a nonlinear constitutive relationship is employed for experiments in this study, and the results demonstrated the capability of the proposed framework to update nonlinear electromechanical properties. The importance of including model parameter uncertainties to generate robust predictive tools is also supported by the results. Therefore, this framework constitutes a powerful tool for the robust design and prediction of PEH performance.