In offshore Oil & Gas production processes the undesired severe slug flow regime can be present. The negative impact of severe slug is crucial to the production rate and process safety. In this work, the severe slugs which occur in the well-pipeline-riser system are experimentally and theoretically investigated though mathematical modeling, laboratory experiments, control system design and analysis, numerical simulations and laboratory implementations.
In general, this thesis studies the modeling and control of slugging flows which can occur in offshore well-pipeline-riser systems, from both theoretical and experimental perspectives. Some typical control-oriented mathematical models are reviewed and examined. Some extensions have been proposed for improving the model accuracies. The choice of control structure is analyzed based on the Input-Output (IO) controllability concept. All the respective studied measurements give better results than the frequently used riser topside pressure (Pt). A supervisory self-learning control strategy is developed and the results show that the decision making based on the supervisor drives the system close to the closed-loop bifurcation point, but a faster control scheme can reduce the settling time significantly. A number of anti-slug control strategies are proposed, where the robust control solution shows the best potential in both anti-slug control and production rate improvement. Simulation results show that control solutions with the riser bottom pressure Pb performs better than the ones developed for Pt . Furthermore, an alternative transmitter is experimentally investigated for online slug detection and monitoring. The transmitter is an Electrical Resistance Tomography (ERT) sensor measuring the electrical resistance over the cross-area section of the transportation channel. The results show that the transmitter can be a good alternative to conventional measurements if the oil-to-water ratio is low and the fluids are well-mixed. The severe slug’s influence on the downstream separation process is examined. It is confirmed that the riser-induced slugs entering the gravity separator has significant impact on the pressure-dropratio (PDR) controller’s tracking performance on the de-oiling hydrocyclone.
Most of this thesis’s contributions have been experimentally examined and validated. Some problems and techniques still need further investigations in the future, for example, automatic generations of slug models and corresponding control solutions, with respect to the fact that the slug characteristics of the individual installations in offshore productions can differ significantly from each other.
|Navn||Ph.d.-serien for Det Teknisk-Naturvidenskabelige Fakultet, Aalborg Universitet|
Assoc. Prof. Zhenyu Yang, Aalborg University