Algorithm for Selecting Signal Processing Technique in Vibration Based Condition Monitoring System

Abstract

In the sound and vibration analysis, dynamic data acquisition plays important role. Dynamic data acquisition not only means acquire data but analyse it according to the complexity of the signals generating from the source of vibration. Accuracy of result depends upon the accuracy with which the signal processing technique is selected. In present work, a comprehensive condition monitoring algorithm is proposed. This algorithm classify the signal according to its behavior with respect to frequency and time characteristics and then select the signal processing technique to acquire a meaningful results. This system not only selects the signal processing technique but also gives the type of bearing defect generating the vibration signals.

Keywords: Vibration, frequency, order, model, FFT, wavelet