Despite the rapid development and widespread adoption of physics-informed neural networks (PINNs) in various engineering fields, their applications in microelectromechanical coupling systems (MEMS) remain relatively unexplored. In this study, we demonstrate a novel implementation of PINNs for modeling and characterizing a piezoelectric microactuator. By leveraging the beam-like structure, the governing equations for a multi-layered piezoelectric actuator are first derived and subsequently incorporated into the PINNs model to accurately predict the deformation of the piezoelectric actuator in response to a given voltage input. Furthermore, by integrating experimental deflection data obtained from Laser Doppler Vibrometer measurements into the neural network, we further demonstrate the potential of PINNs in identifying the piezoelectric material coefficient through inverse analysis. Our contribution in applying PINNs to models and characterizing piezoelectric actuators in MEMS serves as a promising starting point for the broader utilization of machine learning techniques in this field.
As smart structures are becoming increasingly ubiquitous in our daily life, the need for efficient modeling electromechanical coupling devices is also rapidly advancing. Smart structures are often made of piezoelectric materials such as lead zirconate titanate (PZT), which exhibits strong nonlinear behavior known as hysteresis effect under a large applied electric field. There have been numerous modeling techniques that are able to capture such an effect; some techniques are suitable for obtaining physical insights into the micro-structure of the material, while other techniques are better-suited to practical structural analyses. In this paper, we aim to achieve the latter. We propose a simplified phenomenological macroscopic model of a nonlinear ferroelectric actuator. The assumption is based on the direct relation between the irreversible strain and irreversible electric field, and the consequently irreversible polarization. The proposed model is then implemented in a finite element framework, in which the main features such as local return mapping and the tangent moduli are derived. The outcomes of the model are compared and validated with experimental data. Therefore, the development presented in this paper can be a useful tool for the modeling of nonlinear ferroelectric actuators.