Fault Detection of the Electrical Motors Based on Vibration Analysis

Detection and diagnosis of faults while the system is running can help to reduce all kind of losses. Mechanical faults, usages, slackness's, cause different noises and vibrations with different amplitude and frequency against the normal sound and movement of the equipment. Induction motors are present in any processes and systems and for this it is important to known the types of faults which can occur during its working. This paper presents briefly some electrical and mechanical faults in induction motors and how these faults influence the motor vibration. Each type of fault, electrical faults or mechanical faults, produces a vibration whit a specific frequency. Through monitoring and analyzing the vibration spectrum these specific frequencies can be detect and through these we can detect usages, slackness and forces which act and produce the fault. It is an undestroyed method to find out the functioning problems. If the amplitude of these vibrations reach to a certain level the fault can be detect and identify. The paper presents a possible two degree of freedom model for electrical motors. Motion equations and transfer function are deduced and through simulation can be followed how the vibration spectrum is changing for some important faults, like electrical imbalance, mechanical unbalance, looseness, bearings, external effects, base structure or resonance, which can be reduce to mass or spring variation in the system model. To obtain the behaviour of the system, the equations of the motion and Matlab simulations was used.