Fault Diagnosis of Compressor Installed in Offshore Plant Using Wavelet Transformation

초록

This paper introduces Morlet wavelet transformation to fault diagnosis of rotating machinery installed in offshore plants. The fault diagnosis is composed of the processing of factual data such as vibration signals which is quantitative for the actual condition or health of the machine, and an interpretation of the acquired data to determine the machinery fault existence and identify specific components (e.g. gear set, bearing, shaft) inside machine that are degrading. Machinery’s abnormal vibration signal includes its fault information, effective testing and analysis of its vibration signal is the key to realize machinery fault diagnosis. Vibration signals acquired from an industry vibration transducer, such as an accelerometer are processed by Morlet wavelet. Time-frequency spectrum result obtained by Morlet wavelet transformation can show the power distribution of the signal in both time and frequency domain. Special attention is given to the common vibration causes like unbalance and bearing defects in the rotating machines installed in offshore plants. Analysis performed using simulated signal proves that time-frequency contour map have the capabilities for fault identification. Thereafter the root-cause analysis of the original vibration problem, so called failure diagnosis, can be performed using the time-frequency contour map. Acceptance limits for normal operation is decided by the application of vibration standards and vibration alarm limits given by vibration severity guidelines. A Graphical User Interface (GUI) is developed to control the diagnosis assessment process. Also, a case study about the implementation of the continuous wavelet transform to the centrifugal compressor fault diagnosis is discussed. Maintenance is the sum of the actions taken to prevent equipment from unexpected breakdowns or to keep it work properly. Unfortunately, many records over the past decades point out that most of the facilities don’t implement the effective practices to