Hybrid feature selection method for SVM classification and its application for fault diagnosis of wear and peeling in journal bearing with a little muddy water using long-term real data

Rotating machines are widely used as components of various industries around the world, and its normal operation rotating is important. Thus, condition monitoring fault diagnosis have considerable attention in recent years. Industrial statistics illustrate that 40% total large machine breakdowns happened due to broken bearings, while for small machines, analogous number reaches up 90%. This study aimed at researching journal bearings using support vector (SVM) method. The experimental systems vertical horizontal shafts were developed. There was no adding any initial artificial failure bearing, a little muddy water used, long-term vibration data both obtained until bearing damages occurred (3-hour tests conducted repeatedly 128 datasets shaft 24 obtained). A feature selection method focused, hybrid by combining Fisher score (FS) sequential forward (SFS) proposed. Its accuracy efficiency proved experimentally with 97.14% 100% shaft. Furthermore, result SVM model method, most important rotor system clarified mean value RMS, only this can give good result. It useful suggestion selecting features machines.