Abstract
The article is dedicated to the pattern recognition of unbalanced rotor vibration trajectories. The diagnostics of rotary machines with fluid-film bearings is studied. The feed forward neural networks were used to analyze the measurement data of rotor vibrations and other parameters of the rotor-bearing system. The states of the system were studied at various values of the rotor unbalance. It was shown that the number of training samples and the number of neurons in the input layer have the greatest impact on recognition accuracy. As a result of training the neural network to recognize 3 classes of defects, an accuracy of more than 97% was achieved.
Export citation and abstract BibTeX RIS
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.