Title

A Comparison of Model-Based Machining Force Control Approaches

Abstract

Machining force regulation provides significant benefits in productivity and part quality. Adaptive techniques have typically been utilized due to the tremendous parameter variations that are found in machining processes. While adaptive controllers provide greater stability as compared to fixed-gain controllers, they have found very little headway in industry due to complexity in design, implementation, and maintenance. Recently, model-based techniques, with and without process compensation (i.e., the ability to directly adjust controller gains given known changes in process parameters), have been explored. This paper provides a comparison of four model-based machining force controllers; namely, linearization, log transform, nonlinear, and robust. These controllers are compared to an adaptive machining force controller in terms of transient performance and stability robustness with respect to parameter variations, and in terms of stability robustness with respect to unmodeled dynamics via simulation and experimental studies. The developed stability analyzes for the model-based controllers provide excellent predictions of the stability boundaries in the parameter space. Thus, stability robustness in terms of both model parameter variation and controller parameter adjustments can be systematically explored. Also, the results demonstrate that the stability robustness of the model-based controllers is insensitive to unmodeled servomechanism dynamics. While each force control approach performed satisfactorily in a laboratory environment, it can be generally concluded that their implementation should be dictated by the economics of the production environment.

Department(s)

Mechanical and Aerospace Engineering

Keywords and Phrases

Adaptive Control; Machining Force Controllers; Machining Processes; Milling; Model-Based Control; Robust Control

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2004 Elsevier, All rights reserved.

Share

 
COinS