Investigation into the influence of bearings nonlinear forces in unbalance identification

Rotating machines are key components in several industrial sectors, mainly, in energy generation. In this way, a rotating component supported on hydrodynamic bearings creates typical problems, being high vibrations amplitudes due to unbalance, one of the most common. To avoid failures and ensure a safe operation, the rotor should be balanced, and influence coefficient methods are usually used. Since balancing by the traditional influence coefficients method requires trial masses, the machine can sometimes experience a long setup time, causing financial losses. Also, this method assumes a linear rotor response that can hamper the balancing procedure for nonlinear situations, which happens when the rotor experiences high vibrational motion. Thus, this work proposes a balancing identification that considers nonlinear bearings and avoids trial masses. For this, a mixed-integer gradient-based optimization is presented. The theoretical model of a rotor supported by hydrodynamic bearings is obtained using the Finite Element method and solving the Reynolds equation. In order to save computational time, the bearing forces are approximated by a fifth order Taylor series expansion. The presented results show that nonlinear bearing consideration can improve the machine diagnosis. (C) 2020 Elsevier Ltd. All rights reserved.

Automated summary

This study proposes a balancing identification method that considers nonlinear bearings and avoids trial masses. The theoretical model of a rotor supported by hydrodynamic bearings is obtained using the Finite Element method and the bearing forces are approximated by a fifth order Taylor series expansion. The results show that considering nonlinear bearings can improve machine diagnosis.