Study on Fault Eigenvalue Reduction Approaches Based on Rough Sets Theory

An extraction and reduction method of fault eigenvalue was proposed in this paper. Firstly, fault feature factor was extracted from stator current signal, axial vibration signal and radial vibration signal. Then, using differential matrix in rough sets theory, the extracted fault feature factor was reduced to achieve a more sensitive, more effective fault eigenvalue. Through analysis on these reduced fault eigenvalue, signal type which was sensitive to fault was obtained. The experiment result shows the reduction method has great significance in selecting reasonable fault eigenvalue and simplified fault diagnosis process.