Experiments and Modelling for the Control of Riser Instabilities with Gas Lift

Three slug types were found in the new lab experiments carried out in a 2" flowline-riser system with 477 litre volume operating at ambient pressure: severe slugging of type I (which has a liquid slug length larger than the riser height), severe slugging of type II (having a liquid slug length smaller than the riser height), and small scale slugs initiated in the flowline that prevent the severe slugging cycle. Slug predictions with the OLGA model are in good agreement with the experiments, as well as with other experiments for similar, but smaller, configurations described in the literature. The model also agrees with the severe slugging cycle of the Gannet system in the North Sea. The lab experiments show that gas lift mitigates the type I slugging cycle, but it does not fully remove it. Experiments for gas lift with the type II severe slugging cycle show that the cycle is removed if the superficial lift gas velocity is above 0.75 m/s. No effect of gas lift is found for the experiments where the severe slugging cycle was absent anyhow due to hydrodynamic slugging in the flowline. The OLGA predictions for the effect of gas lift for the three slug types are in reasonable agreement with the lab experiments. Particularly the experiments and simulations show that gas lift is not very effective for slug control.