Feature extraction of rotating apparatus using acoustic sensing technology

This article presents a feature extraction of rotating apparatus using acoustic sensing technology (AST). The kernel algorithm is based on an acoustic signal enhancement filter (ASEF). The acoustic feature extraction algorithm is implemented by using Mel-scale frequency cepstral coefficient (MFCC) theory. The system utilizes an National Instruments (NI) cRIO-9067 embedded controller and a real-time signal sensing module to analyze rotation performance and predict malfunctions in rotating apparatus. AST can adopt low noise array microphone which the effective bandwidth is 20 to 10000 Hz. Experimental results showed that the acoustic signal method could effectively perform real-time early fault detection and prediction in proposed system. Smart AST was proposed that can distinguish the acoustic feature differences of normal and abnormal ones.