Based on the information entropy theory

a fault diagnosis methodology for rolling element bearings was proposed. It is a fusion of wavelet energy spectrum exergy and Mahalanobis distance. Inner ring fault

outer ring fault

rolling element fault

inner ringrolling element fault and inner ringouter ring fault of rolling element bearing were simulated on a twin spool rotor test rig. The acoustic emission signals of each fault were collected. Single fault and coupling faults of rolling element bearings were successfully diagnosed using the information exergy distance method for acoustic emission signals. The diagnosis results show that the method has higher information utilization ratio than the information entropy method. It can diagnose early faults in rolling element bearing clearly and accurately

which is proved more effective than the information entropy distance method.