Blind separation of vibration components in machine monitoring
The paper presents an approach for machine vibration analysis and health monitoring using blind source separation (BSS). By this approach, the investigated problem is transferred from the original space of the measurements to the space of independent sources, where the reduced number of components will simplify the analysis problem and where different methods can be applied for scalar signals. The BSS will also provide a mixing model of the sources, so it will be possible to see how the evolution of the estimated sources, determined by BSS, are reflected in the original measurements. The assessment of the method on a real machine, a pump driven by an electromotor, is presented.