Research on Fault Diagnosis for Petrochemical Running Equipments Based on ICA
Zhi‐Gang Chen, Xiao Jiao Lian, Ming Zhou · Applied Mechanics and Materials · 2011
For solving the difficulty of feature signal extraction from vibration signals, a new method based on Independent Component Analysis (ICA) is proposed to realize separation and filtering for multi-source vibration signals. Firstly, the principal and algorithm of ICA used to separate mixed signals is introduced. Secondly, application in signal separation and filtering with ICA is studied in diagnosis. In addition, imitation and field examples are given. The experiments show it is feasible to separate and extract feature signal from multi-source vibration signals and it is an effective method in signal preprocessing in fault diagnosis.