Indentification of multiple fault Parameters in rotor-Bearing-Coupling systems Based on Forced Response Measurments
No Thumbnail Available
This work focuses in two main facets of the rotor dynamics field, the analysis of rotor-bearingcoupling system (i.e., the direct problem) and the fault detection and diagnosis (i.e., the inverse problem). The linear analysis of rotor-bearing system is quite mature and now trend is towards the numerical simulation of complicated rotor systems. The fault detection and diagnostic is not straight forward since different faults may produce similar vibration signature. Frequent faults that appear in rotating machinery are the residual unbalances, shaft bow/bent, shaft misalignment, bearing faults, gear faults, rotor cracks, motor faults, etc. Among identification (i.e., the fault detection and diagnosis) of various faults, the residual unbalance estimation in a rotor system is an ancient problem; however, the research on this field is still active and now the trend for the unbalance estimation is to reduce the number of test runs, number of measurements required, and the optimum placement of measurement sensors and balancing planes, especially for the large turbo-machineries where the downtime is very expensive and the accessibility of measurements is limited. After the residual unbalance, the second most common fault is the misalignment of shafts at bearings and couplings, and it causes reaction forces and moments at couplings as well as at bearings, and is often a major cause of machinery vibration and its failures. Especially in a rotor train, the misalignment at couplings and bearings is a well known problem and its identification is still a challenging area of the research..
Supervisor: R. Tiwari