Identification of friction in inerter with constant and variable inertance

This paper presents the experimental identification of the friction force in the inerter with constant and variable inertance. The change of intertance is possible due to the implementation of continuously variable transition in the gear train. As a result of our study, we created robust models that can be applied to various devices equipped with inerters. We designed and built the experimental rig, which enables us to enforce the motion of examined systems along the assumed path of the motion, with maximum velocity up to 1[m/s]\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$1\,[\mathrm {m/s}]$$\end{document}. With the proper analysis, we extract the friction force. Then, we propose mathematical models of friction force based on static Coulomb and dynamic LuGre models. Additionally, we use the artificial neural network to model the friction force characteristic. Finally, we validate created models with additional sets of experimental data. The LuGre model and the artificial neural network replicate the real response efficiently, and both approaches can be used for precise modeling of the systems’ motion.