Fault Classification and Degradation Assessment Based on Wavelet Packet Decomposition for Rotary Machinery

Abstract

This paper presents a novel method for fault classification and degradation assessment in rotary machinery through wavelet packet decomposition and data-driven regression methods. Wavelet Packet Decomposition is applied to extract the coefficient and energy based features from vibration signals. During the experiment, we used several machine-learning methods, including Artificial Neural Networks, Support Vector Machine, and K-Nearest Neighbor Classification for degradation assessment and compared the numerical results.