Fault Diagnosis of Rotating Machines in Offshore Plant Based on Morlet Wavelet Transform

초록

The condition monitoring consists of a selection of measurable parameters such as vibration signals which correlate with the health or condition of a machine, and an interpretation of the collected data to determine the machinery fault existence and identify specific components (e.g. gear set, bearings) in the machine that are degrading. Many vibration and unbalanced signals of fault rotating machines in offshore plant have complex time-frequency characteristics. As a timefrequency analysis, wavelet transform is useful for locating transient events, discontinuities and patterns of these faults in non-stationary vibration data. The wavelet transformation can be an efficient way to detect the fault from the signal since the signal always includes the fault information about the rotating and reciprocating equipment inside. This paper introduces the faults diagnosis for offshore rotating machines by the application of Morlet wavelet transformation. Special attention is given to the vibratory and unbalanced faults, namely, rotating unbalance and resonance found in the rotating machine installed in offshore plant. Vibration signals recorded from accelerometer are processed by Morlet wavelet so that the both bearing and misalignment failure diagnosis. Thereafter, time-frequency contour map is introduced into fault diagnosis. The timefrequency contour map can easily show the power distribution of signal in time and frequency domain. Moreover, it is a good way to identify the faults involving a breakdown change. Several typical faults of rotating machines are detected by the time-frequency contour map obtained by the Morlet wavelet. The simulative results show that time-frequency contour map have the capabilities to identify the faults. This method also appeared to be an effective tool to diagnose and to discriminate the different types of machinery faults based on the unique pattern of the wavelet contours.