26 Vibration-Based Tool Condition Monitoring Systems C. Scheffer

Back

University of Stellenbosch

P.S. Heyns

University of Pretoria

26.1 Introduction ........................................................................ 26-1

26.2 Mechanics of Turning ......................................................... 26-2

General Terms † Chatter Vibrations † Tool Wear

26.3 Vibration Signal Recording ................................................ 26-7

Direct and Indirect Systems † Sensor Requirements for

Tool Wear Monitoring † Force Measurement † Acceleration

Measurement † Acoustic Emission Measurement † Sensor

Comparisons

26.4 Signal Processing for Sensor-Based Tool

Condition Monitoring ........................................................ 26-11

Feature Extraction † Feature Selection

26.5 Wear Model/Decision-Making for Sensor-Based

Tool Condition Monitoring ............................................... 26-15

Trending, Threshold † Neural Networks † Fuzzy Logic †

Other Methods

26.6 Conclusion ........................................................................... 26-20

Summary

Despite the high level of technology built into every aspect of modern metal cutting operations, the phenomenon of

tool wear still hampers the reliability and complete automation of machining processes. Tool wear is the loss of

material on the edge of the cutting tool. This chapter concerns sensor-based tool condition monitoring (TCM), and

specifically those methods that are based on vibration related properties such as force, acceleration, and acoustic

emission (AE). References are made to systems proposed in the literature and also to commercially available

hardware. The chapter focuses on turning operations. The mechanics of turning are briefly discussed. Various

methods of obtaining vibration signals from turning operations are described. The vibration signal has to be

processed in order to estimate the level of wear in the cutting edge of the tool, and several state-of-the-art approaches

are discussed. Effective methods of constructing a model relating sensor data and the tool wear, using processed

vibration signals, are described. The chapter concludes by indicating some important points that should be

considered when using vibration-based systems for TCM, and some interesting topics for future research in this field

of study. Chapters 25 and Chapter 27 present further information on the present subject.

Навигация