Contact Tip Trajectory in Steady-State Regime Prediction Using Deep Learning for Piezoelectric Actuators

Including friction modeling in simulation can be challenging and requires time domain simulation, which is time consuming. We propose to separate the friction modelling from the FEA (Finite Element Analysis). The approach is defined for UCM (Ultrasonic Ceramic Motor) a type of piezoelectric actuator strongly relying on friction coupling between a vibrating tip and a sliding part. The tip's trajectory is studied both with and without contact. A DNN (Deep Neural Network) is trained to predict the contact trajectory from the contactless one, which can be obtained in frequency domain. Used as a complement of contact-free simulation of the UCM, the DNN can be used to predict the behavior of the complete actuator.