Intelligence Fault Diagnosis of Electromotor

Because of many mechanical problems are initiallyrecognized by a change in machinery vibrationamplitudes; a maintenance process where thecondition of equipment with regard to vibration ismonitored for early signs of impending failure. Inaddition, type and severity of the problems can berecognized by the frequency of vibration, plus thelocation and direction of the vibratory motion. Theobjective of this research was condition monitoringof an Electromotor to investigate the correlationbetween vibration analysis, fault probability of dataand fault prognosis. In this research Grms (Root-Mean-Square Acceleration) and the fault probabilityamount of an Electromotor at 1450 RPM in differentsituations including health, unbalance, misalign andlooseness were calculated. Probability DistributionFunction is a mathematical function can be used tomodel the frequencies and their probabilities ofoccurrences over time. Exponential distributionfunction was used for calculating the Electromotorfault probability. In this research, the technique ofWavelet Neural Networks (WNN) in the 1450 RPMElectromotor faults are classified.

Keywords:

Fault Diagnosis , Electromotor , Wavelet Neural Networks