Eventos Anais de eventos
DINAME 2023
XIX International Symposium on Dynamic Problems of Mechanics
Balancing of a Rotating Machine using the Kriging approach
Submission Author:
Marcus Filipe Sousa Reis , MG
Co-Authors:
Marcus Filipe Sousa Reis, Vinícius Carvalho, Leonardo Sicchieri, Aldemir Ap Cavalini Jr, Valder Steffen Jr
Presenter: Marcus Filipe Sousa Reis
doi://10.26678/ABCM.DINAME2023.DIN2023-0070
Abstract
This work presents a new approach to balancing rotating machines using a technique based on vibration signals. The proposed methodology uses a Kriging surrogate model to perform the balancing of a rotating system. The Kriging formalism requires input and output samples. Therefore, the Influence Coefficient (IC) method is used to obtain the output samples, which are the correction mass and its corresponding angular position. The input samples are the unbalanced vibration amplitudes and associated phases. With these samples, the surrogate model is mounted, and hence for new unbalance conditions, is possible to predict the correction mass and its corresponding angular position, without the need for trial weights. Therefore a reduction of the time spent in the balancing procedure is obtained, whilst the vibration amplitude is also reduced. The obtained results using the proposed approach were compared to the IC method for validation purposes. The proposed approach demonstrated itself to be effective for balancing rotating machines, achieving a satisfactory reduction on the vibration amplitudes of the adopted rotor. The methodology was validated both numerically and experimentally.
Keywords
balancing, Surrogate model, rotating machines, influence coefficients method, dynamics
