Development and Optimisation of Multiphase Flow Models Using Design of Experiments: The Brazilian Pre-Salt Case Available to Purchase

We developed and optimised a mechanistic flow model to predict the flow of oil, gas and water in Offshore Production Facilities (OPFs) in Brazilian pre-salt fields. First, two leading modelling frameworks, referred to as the North Sea and Gulf of Mexico flow models, were tested against a database of 249 pressure drop readings collected in production pipelines and well tubings of six OPFs located in the Campos Basin. Then, the Design of Experiments (DoE) framework was employed to provide an efficient and systematic means to quantify the sensitivity of the best-performing model to its various parameters, considering both primary and interaction effects. Finally, the parameters with the greatest impact on the pressure drop predictions were selected for further investigation, and their closure relationships were replaced one by one by other expressions from the literature. The new combination, when employed with fitted values for the tubing and pipeline surface roughness, reduced the Mean Absolute Percentage Error (MAPE) from 5.59% to 4.23% and left almost 90% of the simulations within the ±10% tolerance range, with only one datum outside the range of ± 20%.