Wavelet Neural Network – Based Research on Online Wearing Prediction of TI6AL4V Cutter in High Speed Milling

Abstract:

TI6AL4V is a kind of hard-machining material, which has bad thermal conductivity, good chemical reactivity, little elastic modulus, great friction coefficient, severe work hardening, short cutter life, low machining efficiency and poor machining surface quality. To improve the machining efficiency, reduce machining cost and improve products quality, the cutting tool wear is the key factor affecting machining quality, machining efficiency and production safety. In this paper, a test system which takes TI6AL4V as the research object, and the dynamic milling force during the high speed milling as the detection signal is built for online tools wear prediction. The method of wavelet packet transform and neural network are presented to diagnose and predict the situation of tools wear. The practical example shows that this system has good practicability and could identify the tools wear states exactly through verification tests.

• Periodical: Key Engineering Materials (Volumes 431-432)
• Pages: 205-208
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.431-432.205
• Online since: March 2010
• Authors: Hong Yu Zhu, Wei Jin Chen, Ying Li
• Keywords: Cutter Wearing Online Prediction, High Speed Milling (HSM), Neural Network (NN), Ti6Al4V, Wavelet Packet (WP)

References

[1] T. Ôzel and et al.: International Journal of Machine Tools and Manufacture, Vol. 40 (2000), p.713.

[2] N. Fan: Modular Machine Tool ＆ Automatic Manufacturing Technique, Vol. 5 (2008), p.69.

[3] M. Wang: Studies on Tool Wear in Milling of Titanium Alloys (Ph.D. Nanjing Institute of Aeronautics, 1985).