Cutting Condition Monitoring in Milling within Enclosed Environments Using a Microphone Array and Random Forest
This study demonstrates that monitoring cutting conditions in highly reverberant milling environments is significantly more effective when using a microphone array combined with octave-band decomposition, Wiener filtering, and Random Forest classification, achieving 95.5% accuracy compared to 64.7% with conventional broadband methods by preserving frequency-selective process information that global metrics lose to reverberation.