Fault Diagnosis of Rotating Machines Based on the MathematicalModel of a Rotor Bearing-Mass System

This work deals with fault diagnosis techniques in rotating machines under unbalance and misalignment based on the mathematical model of a rotor bearing-mass system (RBMS). The fault diagnosis is approached as an observation problem where faults are reconstructed by means of the Luenberger observer. In order to validate the performance of the given methodology a laboratory setup prototype has been implemented employing a mechatronic framework. The fault diagnosis system consists of: Two bearings, a shaft, a disk, two inertial measurement units, a computer, a single-phase induction motor and a start-stop module. The disk is placed along the shaft, in the middle of the two bearings in order to cause a rotor fault due to unbalance, on the other hand, misalignment is caused by means of a misaligned bearing.