Fault Diagnosis of Ball bearing by Using Fault probability Distribution Function and Support Vector Machine

Because of many mechanical problems are initially recognized by a change in machinery vibration amplitudes; a maintenance process where the condition of equipment with regard to vibration is monitored for early signs of impending failure. In addition, type and severity of the problems can be recognized by the frequency of vibration, plus the location and direction of the vibratory motion. The objective of this research was condition monitoring of an Ball bearing to investigate the correlation between vibration analysis, fault probability of data and fault prognosis. In this research Grms (Root-Mean-Square Acceleration) and the fault probability amount of an Ball bearing at 1450 RPM in different situations including health, inner, outer and ball were calculated. Probability Distribution Function is a mathematical function can be used to model the frequencies and their probabilities of occurrences over time. Exponential distribution function was used for calculating the Ball bearing fault probability. In this research, the technique of Support Vector Machine(SVM) in the 1450 RPM Ball bearing faults are classified.