Modeling and estimation of speed dependent bearing and coupling misalignment faults in a turbine generator system

Abstract

Faulty machinery generates exaggerated vibration that leads to fatigue and catastrophic failure of critical components of machinery and even sometime worker’s life. Hence, the fault free functioning of rotating machinery is the prime objective of many industries. The accurate modeling, thereafter estimation of critical components of any rotating machinery is the best way to achieve the fault free condition. The present research work involves with the development of an algorithm for multi–degrees of freedom turbine–generator system to evaluate speed–dependent bearings and coupling misalignment faults, numerically and experimentally. The modeling and simultaneous estimation of speed–dependent bearing and misalignment parameters in addition with speed independent unbalance parameters are the novelty of the present research. In theoretical analysis, finite element method is applied to derive equations of motion of the dynamic system. Least squares technique is used to evaluate the fault characteristic parameters of the turbine–generator system. To test the algorithm additive noise is added in the numerically generated response, and found the algorithm performs well even in the presence of noisy response. In experimental analysis, at randomly selected spin speeds, three sets of forced responses are captured at four bearing locations. These forced responses are imported into the developed algorithm and characteristic parameters are evaluated experimentally. The maximum standard deviation achieved in the quantification of parameters for three sets of data is 2.3 and 2.5 for 29 Hz and 31 Hz, respectively.