Angle order analysis technique for processing non-stationary vibrations

The basic signal processing technique is based on the classical Fast Fourier Transform (FFT). However, for machinery operating in non-stationary conditions, under varying operating condition, the FFT approach is not directly effective for analysis. Presently, two techniques are used: Order Tracking Technique (OTT) and Time-Frequency Analysis (TFA). OTT is used to avoid a spread of the spectral components of a spectrum obtained from variable speed machinery, however, it presents the disadvantage of averaging the amplitude of the spectral components during acquisition time. The TFA are three-dimensional functions that allow to visualize the frequency and amplitude variations of the spectral components. However, when the analyzed vibration has larger changes of the machine speed during measurement, they become very difficult to analyze. The present paper presents a new approach called Angle Order Analysis Technique (AOAT), which uses the advantages of both OTST and TFA techniques. Basically, the technique involves changing the signal sampled at constant time increments to a signal sampled at constant angular increments and then processes it using Wigner-Ville Distribution. The proposed method is evaluated using experimental data of a gearbox under running-up.