Ayberk Zorlu

Topic

Operational Modal Analysis for Chatter Prediction in Milling

Department of Mechanical Engineering

Date & location

• Monday, July 15, 2024

• 10:00 A.M.

• Virtual Defence

Reviewers

Supervisory Committee

• Dr. Keivan Ahmadi, Department of Mechanical Engineering, University of Victoria (Supervisor)

• Dr. Yang Shi, Department of Mechanical Engineering, UVic (Member) 

External Examiner

• Dr. Xiaoliang Jin, Department of Mechanical Engineering, University of British Columbia 

Chair of Oral Examination

• Dr. Martin Segger, Department of Art History and Visual Studies, UVic

   

Abstract

Unstable vibrations during machining can harm both the tool and the workpiece, requiring careful selection of process parameters to avoid them. These parameters are usually set based on vibration models of the machining process. However, due to unmodeled dynamics or process variations, chatter can still occur, highlighting the need for online chatter monitoring systems. Existing methods often detect chatter only after it occurs, so there is a need for monitoring systems that can predict chatter before it occurs to ensure high-quality machining.

This thesis presents a new method to identify the dynamics of regenerative chatter from the measured process vibrations in milling. This method combines the synchronous once per-revolution sampling of stable process vibrations with Operational Modal Analysis to estimate the Floquet multipliers of the delayed linear time-periodic dynamics in milling, all from the natural process vibrations without external excitation. The identified multipliers quantify vibration stability, enabling chatter prediction before it occurs. Additionally, they can be used to calibrate physics-based chatter models based on vibration measurements solely within the stable region.

The method’s accuracy in identifying Floquet multipliers is validated through extensive numerical simulations and two experimental case studies. The results show that chatter due to both Hopf and period-doubling bifurcations can be predicted from the process vibrations during stable cuts. Moreover, the experimental case studies demonstrate a vibration measurement system for implementing the presented method in standard milling operations and confirm its effectiveness in practice.