Linear discriminant function analysis of acoustic emission signals for cutting tool monitoring

摘要: Abstract A principal setback to automation of the machining process is inability completely monitor condition cutting tool in real time. Whereas several techniques developed date are useful specific applications, no universally applicable sensor yet available. Acoustic emission one most promising be recently for on-line monitoring. However, signal analysis still an area that requires further investigation enhance potential acoustic emission. For this purpose, frequency-based pattern recognition concepts using linear discriminant functions have been used analysing signals generated during distinguish between different sources, specifically chip formation, fracture, and noise. Five features were classification frequency range 100 kHz 1 MHz, with each feature consisting a 20 bandwidth, selected class mean scatter criterion. The coefficients obtained by training system sources interest. An AISI 1018 steel was machined titanium carbide-coated tool. Cutting speeds ranged from 200 800 ft/min (1 4 m/sec) feed rats 0·0005 0·0075 in/rev (0·0133 mm/rev 0·191 mm/rev) depth cut 0·17 (4·32 mm). results show successful rate 90% breakage, while those formation noise 97 86% respectively.