Rotating equipment vibration signal analysis is currently the most common method of determining its technical condition. For this purpose, in the signal spectrum the harmonics levels, associated with the defect magnitude, are determined at the defects frequencies. To improve the diagnosis accuracy and reliability the analysis should be carried out not by the instantaneous spectrum, which is strongly influenced by noises, but by the average spectrum. However, if the equipment speed mode is non-stationary, a direct averaging of the spectrum will lead to recognition errors due to the defects frequencies dependence on the rotation speed.
The paper considers diagnostics of rotating equipment with non-stationary operating mode by the example of a rolling mill diagnostics. To obtain the averaged spectrum, a scale transformation in frequency, which ratio depends on the mill speed mode, has been used. Thus an ordinal average spectrum in rotation frequency terms has been received. To determine accurately the speed mode two independent channels are used – a hardware speed sensor and a software channel of an automatic rotation frequency recognition according to the case vibration.
Using the ordinal average spectrum and two independent channels for the rotation frequency determining has provided an accurate and reliable system for the mill technical condition diagnostics.
Kostyukov V.N., Boichenko S.N. Vibration diagnostics of rotating equipment with non-stationary speed mode // X International Conference Condition Monitoring and Machinery Failure Prevention Technologies - СМ/MFPT 2013 Download the publication
You can order the translation by e-mail: a@dynamics.ru
|