(641d) Incipient Fault Detection in Offshore Oil and Gas Production Platforms: Hybrid Electro-Mechanical Monitoring of Rotating Equipment

Offshore oil and gas production facilities are uniquely hazardous in that platform personnel have to work in a perilous environment surrounded by extremely flammable hydrocarbons. The harsh marine environment as well as oilfield disturbances makes the failure of production equipment common. Given the highly integrated nature and congestion of the processing facility, a failure in equipment could quickly propagate to others resulting in leaks, fires and explosions, causing loss of life, capital invested and production downtime. The failure of rotating equipments, in particular, on offshore platforms contributes significantly to the risk of an accident and hence their monitoring becomes important. The history of fault diagnosis and protection of rotating equipments is as archaic as the equipments themselves. The manufacturers and producers initially relied on simple acoustic and vibration observations to detect the presence of a fault which were sufficient when these equipments were first introduced at the start of the industrial revolution. However, nowadays, the tasks performed by these equipments and the processes they are utilized in have become increasingly complex. Hence, there is a need to develop intelligent monitoring tools capable of monitoring faults at their inception; as unscheduled downtime can upset deadlines and cause financial losses. Faults in rotating equipment in offshore platforms could be caused due to mechanical, electrical or process failures. Monitoring the overall process for the detection of such equipment faults may not be fruitful as faults may be detected too late and platform safety compromised. Hence, we propose to monitor the equipments characteristics to detect incipient faults early in their evolution.

Traditional techniques for monitoring of rotating equipment have relied on observations of periodically measured dynamic sampled response measurements such as vibration. Plenty of literature is available on rotating equipment monitoring using vibration analysis techniques, the most popular among them being frequency (Fourier, Hilbert, Cepstrum and Bispectrum) analysis and time-frequency (STFT, wavelet) analysis. Rotating equipments have also been monitored based on the performance of the electrical motor using current measurements. Similar to vibration analysis, motor current signature analysis (MCSA) also employ signal processing techniques like those mentioned above. But, there is not much literature on a combined electrical and mechanical fault detection technique for rotating equipments. For instance, it is well known that bearing faults induce eccentricities in both current and vibration signals but attempts to monitor the fault have relied on only either one. This paper describes a novel time-frequency signal processing technique along with a hybrid electro-mechanical approach towards monitoring of rotating equipment.

This method is evaluated on two distinct testbeds; one is an experimental setup in which the characteristic vibration and current signals were observed for a lab scale induction motor under normal and fault conditions. The other testbed is an electromechanical model of a typical pump and compressor, similar to those found on offshore oil and gas platforms. Electrical (broken rotor bar, stator winding) and mechanical (bearing) fault are introduced into this model and the characteristics observed. Using measurement data from these testbeds, the signal processing technique is used to monitor the signals and ascertain the condition of the equipment. The results of the combined electro-mechanical monitoring scheme is compared with monitoring either vibration or current signals alone. The proposed scheme is also benchmarked against other signal processing technique currently available in literature.

References

Thorsen, O.V., Dalva, M., (1995), “A survey of faults on induction motors in offshore oil industry, petrochemical industry, gas terminals and oil refineries” IEEE Transactions on Industry Applications, Vol.31, No 5.

Jardine, Andrew K.S., Lin, Daming and Banjevic, Dragan (2006), “A review on machinery diagnostics and prognostics implementing condition based maintenance”, Mechanical Systems and Signal Processing, 20, 1483-1510

Nandi, S., Toliyat, H.A. and Li, Xiaodong (2005), “Condition monitoring and fault diagnosis of electrical motors”, IEEE Transactions on Energy Conversion, Vol.20, No 4.