Online Monitoring Dynamic Characteristics in Thin-Walled Structure Milling: A Physics-Constrained Bayesian Updating Approach

Abstract

Accurate estimation of modal parameters of thin-walled structures during the milling process is crucial since it determines the accuracy of deformation and vibration prediction and then the feasibility of selected machining parameters. However, time-varying dynamic characteristic due to material removal brings a great challenge in the pursuit. This article presents an online monitoring strategy for the estimation of modal parameters of thin-walled structures with measured cutting forces and single acceleration signal. The detailed physical constraints are presented, and the analytical form of the modal shape of thin-wall structures is derived. A Bayesian filter framework with slack variables is established for updating the modal parameters by simultaneously considering the physical constraints and the measured signals. Validation experiments are conducted on two kinds of thin-walled structures with different structures. The results of the proposed method are compared with the results of the standard modal test at certain stages. It indicates that the proposed approach is effective for accurately online monitoring of the spatiotemporally varying modal parameters in milling of thin-walled structures.