Summary
Elaborated models, such as those used for simulation purposes [e.g., in the OLGA® simulator (Bendiksen et al. 1991)], cannot be used for model-based control design because these models use too many state variables and the model equations are not usually available for the user. The focus of this paper is on deriving simple, dynamical models with few state variables that capture the essential dynamic behavior for control. We propose a new simplified dynamic model for severe-slugging flow in pipeline/riser systems. The proposed model, together with five other simplified models found in the literature, are compared with results from the OLGA simulator. The new model can be extended to other cases, and we consider also a well/pipeline/riser system. The proposed simple models are able to represent the main dynamics of severe-slugging flow and compare well with experiments and OLGA simulations.
Introduction
Severe-slugging-flow regimes usually occur in pipeline/riser systems that transport a mixture of oil and gas from the seabed to the surface (Taitel 1986). Such flow regimes, also referred to as “riser slugging,” are characterized by severe flow and pressure oscillations. Slugging problems have also been observes in gas lifted oil wells in which two types of instabilities—casing heading and density wave instability—have been reported (e.g., Hu and Golan 2003).
Slugging has been recognized as a serious problem in offshore oil fields because the irregular flow caused by slugging can cause serious operational problems for downstream surface facilities (e.g., overflow of inlet separators). Therefore, effective methods to handle or remove riser slugging are needed, and many efforts have been made to prevent such occurrences (Courbet 1996; Havre et al. 2000). The conventional solution is to partially close the topside choke valve (choking), but this may reduce the production rate, especially for fields in which the reservoir pressure is relatively low. Therefore, a solution that guarantees stable flow and the maximum possible production rate is desirable.