A distributed parameter model for the piezoelectric stack harvester subjected to general periodic and random excitations

Vibration-based energy harvesting using piezoelectric stacks has received increasing attention in recent years for its higher mechanical-to-electrical energy conversion capability in the d33 mode. However, most research on this type of harvesters is based on either the simplified one-degree-of-freedom model or transfer matrix model. This paper presents a distributed-parameter model for the multilayer piezoelectric stack transducer based on the axial vibration theory of a continuous bar. The presented analytical model is free from the assumption that the generated current is identically distributed over all the piezoelectric layers, which has been widely used by the existing models in literature. A first-order numerical model is also introduced to validate the performance of the analytical model on the prediction of voltage, current and power outputs of the harvester under different types of external excitations. Both the analytical and numerical models are firstly validated by experiments and then are used to predict the electrical responses of the stack harvester under general periodic and random excitations with different intensities. Experiment and simulation results demonstrate that the proposed distributed-parameter model has a good accuracy and reliable performance in the prediction of the electrical responses.